Citation
Giving blood

Material Information

Title:
Giving blood an exploration of the determinants of the gift relationship
Creator:
Casebeer, Adrianne Waldman
Publication Date:
Language:
English
Physical Description:
xiii, 293 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Blood donors -- Social conditions ( lcsh )
Blood -- Collection and preservation -- Social spects ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 258-293).
Statement of Responsibility:
by Adrianne Waldman Casebeer.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
50727615 ( OCLC )
ocm50727615
Classification:
LD1190.P86 2002d .C37 ( lcc )

Downloads

This item has the following downloads:


Full Text
GIVING BLOOD:
AN EXPLORATION OF THE DETERMINANTS OF THE GIFT
RELATIONSHIP
by
Adrianne Waldman Casebeer
B.A. University of California, Irvine, 1992
M.P.P. Georgetown University, 1994
M.A. University of Arizona, Tucson, 1997
A dissertation submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Public Affairs
2002


2002 by Adrianne Waldman Casebeer
All Rights Reserved


This dissertation for the Doctor of Philosophy
Degree by
Adrianne Waldman Casebeer
has been approved
by
7L
Date


Casebeer, Adrianne Waldman (Ph.D., Public Affairs)
Blood Donation: An Exploration of the Determinants of the Gift Relationship
Dissertation directed by Professor Peter deLeon
ABSTRACT
While U.S. blood policy relies for its success on the voluntary actions of
individuals, much of the literature on motivations for donation assumes the rational
choice model of human behavior. Moreover, the provision of copious incentives for
donation presumes that the most effective method for motivating potential donors is
through appeals to self-interest. All the while, there is a lack of solid information on
the factors that motivate the blood donation decision. This is troubling because U.S.
blood donation rates are at an all-time low, as is the volume of whole blood
collected.
The main premise of this dissertation is that a more enlightened view of rational
choice together with theories of altruism and social capital can lead to a richer
understanding of the donation decision. In addition, the policy design literature is
considered, to investigate if an understanding of the judgments of the targets of
public policy might improve our standard policy design practices. The choice of
multiple literature areas reflects the conviction that a deeper understanding of the
motivations for blood donation can best be sought through a multidisciplinary
approach.
Derived from an historical review of blood donation policy and the
multidisciplinary literature review, a series of hypotheses are proposed to shed
cumulative light on the research question: What is the structure of motives that
drives the blood donation decision?. The study design employs two surveys
consisting of multiple research techniques, including policy-capturing, administered
before and after the September 11th bombings.
The results suggest that although stated willingness to donate blood was higher after
September 11, a consistent pattern of decision making was revealed. These
analyses provide support for the notion that individuals are responding to a variety
of motivations; that incentives appealing to altruism would be effective for some
segments of the donor population; and that more organizationally involved
individuals are more likely to donate blood than are less involved individuals.
IV


This abstract accurately represents the content of the candidates dissertation. 1
recommend its publication.
Signed
Peter deLeon
v


DEDICATION
This dissertation is dedicated to my husband Bill, who provided unwavering moral
support and good humor as we simultaneously endeavored to complete two
dissertations while raising two rambunctious little children.


ACKNOWLEDGMENT
My committee members, Peter deLeon, Linda deLeon, Gary McClelland, Laurie
Shroyer and Richard J. Stillman, have been wonderfully supportive throughout this
project. In particular, I would like to acknowledge the contribution of my chair,
Peter deLeon, for his sage advice and guidance, thorough readings of drafts and for
the coincidence of having a spare copy of the The Gift Relationship on his shelf '
when I first approached him about this topic. Laurie Shroyer provided excellent
advice and clear thinking on the methodological component of this dissertation.
Linda deLeon, together with Laurie Shroyer, provided invaluable insights on
questions of analytic methods.
In addition, I would like to thank Gary Grunwald, Ph.D. for some remarkably clear
insights into sticky analytic issues. Brad Warner, Ph.D. and Major Jim Wyznowski,
Ph.D., graciously agreed to meet with me, offered excellent suggestions, and
introduced me to some new analytic methods. Nina Rikowski and Nancy Kinney
introduced me to valuable resources on social capital. Thanks also go to the San
Diego Blood Bank and the San Diego Chapter of the American Red Cross for
helpful information, interviews, and for allowing me to interview some of their
donors.
I could not have completed this dissertation, while raising two small children if it
were not for the continual support of the Grandparents who distracted the little ones
so that I could work. Special thanks to Marsha Waldman for her numerous
demonstrations of altruism. Without her, this dissertation would not have been
possible.


TABLE OF CONTENTS
Figures........................................................xii
Tables.........................................................xiii
CHAPTER
1. INTRODUCTION..................................................1
Introduction..................................................1
Overview.....................................................11
Summary......................................................15
2. A DESCRIPTION OF BLOOD AND A HISTORY OF ITS USES.............16
Introduction.................................................16
Blood and Blood Products: Some Technical Background..........17
The Supply of Blood..........................................21
History of Blood Policy......................................23
On Themes and Metaphors......................................57
Blood Donor Motivations......................................62
Summary..................................................... 68
3. LITERATURE REVIEW............................................70
Introduction.................................................70
Rational Choice Theory.......................................71
Judgment and Decision Making..............................81
viii


Social Dilemmas...........................................82
Risk......................................................88
Altruism........................................................... 93
Social Capital......................................................97
The Development of Social Capital...............................100
Trust...........................................................102
Operationalizing Social Capital.................................104
Trends in Social Capital and Implications for Blood -
Donation and Other Types of Voluntary Behavior..................105
Policy Design......................................................109
Summary............................................................112
4. METHODOLOGICAL FOUNDATIONS..........................................116
Introduction.......................................................116
Hypotheses.........................................................117
Altruism Versus Self-Interest.................................. 118
Incentives......................................................124
Social Capital..................................................126
Trust....................................................128
Feelings of Community....................................129
Associational Involvement................................130
Analytic T echniques.............................................. 131
The Surveys........................................................139
The Study Populations...........................................141
Section 1: Policy Capturing.....................................142
Task Familiarity.........................................142
Task Congruence..........................................144
Refinement of Cues and Determination of Cue Quantity.....145
Sample Size..............................................147
Section 2: A Query into Motivations for Blood Donation..........148
Socio-demographics.......................................150
Donation History.........................................151
Hypothesis Related Variables....................................151
Altruism Versus Self-interest Hypotheses.................151
Incentive Hypothesis.....................................153
Social Capital Hypotheses................................154
Analytic Plan: Portion 1, Policy Capturing Exercise................155
Analytic Plan: Portion 2, Core Survey..............................157
Validity and Reliability...........................................157
IX


Human Subjects Approval.............................................160
Conclusion..........................................................161
5. RESULTS.............................................................162
Introduction........................................................162
The Study Populations...............................................163
Comparability of Survey I and Survey II Populations.................164
Survey-by-Survey Analysis...........................................168
Results from Survey I: by Previous Donation......................169
Results from Survey I: Likely Versus Unlikely Donors.............173
Results from Survey II: by Previous Donation.....................175
Results from Survey II: Likely Versus Unlikely Donors............178
Structure of Decision Making: As Revealed by the Policy
Capturing Exercise..................................................180
Task Congruence and Task Familiarity.............................181
The Dimensions...................................................183
The Regressions..................................................184
The Factor Analyses.................................................186
A Factor Analysis of Survey 1....................................186
A Factor Analysis of Survey II...................................188
A Factor Analysis of the Surveys Combined........................190
Clustering of the Decision Makers by Factor Analytic Results.....193
Cluster Analysis...................................................196
Between Group Analysis Survey I Versus Survey II..................199
Conclusion..........................................................201
6. SYNTHESIS AND CONCLUSION ............................................203
Introduction........................................................203
Evaluation of the Hypotheses........................................204
Evaluating the Altruism versus Self-interest Hypotheses..........204
Evaluating the Incentive Hypothesis..............................210
Evaluating the Social Capital Hypotheses.........................212
Summary..........................................................217
Limitations of the Study............................................218
Theoretical Conclusions.............................................221
Implications for Blood Policy.......................................232
Further Research....................................................239
x


Implications for Other Voluntary Medical Donations............243
Conclusion.................................................. 244
APPENDIX
A. Summary of Policy-Capturing Instrument ...................248
B. Section II of the Survey: Administered
after the Policy Capturing Exercise.......................250
C. Comparison by Previous and Likely Donation: By Survey.....256
BIBLIOGRAPHY.....................................................258
xi


FIGURES
Figure 5.1 Distribution of Median Likelihood of Donation..............169
Figure 6.1 The Relationship Between Perceived Benefits and the
Importance of Cost........................................208
xn


TABLES
Table 4.1 Opportunity Cost Versus Benefit to Others in the
Donation Context...............................................123
Table 4.2 Broad Categorization of Judgment Analysis in Research Contexts:
Using the Dimensions of Task Familiarity and Task Congruence...137
Table 5.1 A Comparison of Survey I and Survey II.........................165
Table 5.2 Summary of the Dimensions Used in the
Policy Capturing Exercise......................................184
Table 5.3 A Brief Summary of the Regression Analyses.....................186
Table 5.4 Results from a Factor Analysis: Survey 1......................187
Table 5.5 Rotated Factor Pattern: Survey 1..............................188
Table 5.6 Results from a Factor Analysis: Survey II.....................189
Table 5.7 Rotated Factor Pattern: Survey II.............................189
Table 5.8 Results from a Factor Analysis of Both Surveys Combined.......191
Table 5.9 Rotated Factor Pattern: Surveys Combined......................192
Table 5.10 Clusters, as Determined by the Factor Analysis, by Survey.....193
Table 5.11 Analysis by Cluster ..........................................195
xiii


CHAPTER 1
INTRODUCTION
Introduction
In 1971, Richard Titmuss, in his influential book The Gift Relationship: From
Human Blood to Social Policy, claimed that the major ills of the American blood
banking system could be basically traced to the reliance on payment for blood
donation. According to Titmuss, the result of the United States reliance on payment
for blood donation is a system that is wasteful of blood, costly, and more likely to
distribute contaminated blood (Titmuss, 1970,314). The United States reliance on
payment for blood stands in contrast to the prohibitions against paid donation in the
UK and most other industrialized countries. In a 1997 afterword to a reprint of The
Gift Relationship. Julian Le Grand1 summarizes Titmusss four main arguments:
1. The market in blood was ... allocatively inefficient. It was highly
wasteful; it created shortages and surpluses. More significantly, it led to 1
1 Le Grand is the Richard Titmuss Professor of Health Policy at the London School of Economics.
1


the production of contaminated blood, that is, it damaged the quality of the
product, with socially disastrous consequences.
2. Second, this market also suffered from inefficiency in production. It was
bureaucratic in operation and administratively costly. In consequence it
provided blood at a much greater expense than a voluntary system would.
3. Third, the market was redistributive, but in the wrong direction. It
distributed blood and blood products from poor to rich, from the
disadvantaged and exploited to the privileged and powerful.
4. Finally, and most devastating of all, a market in blood was ultimately
degrading for society as a whole. It drove out altruistic motivations for
blood donation, replacing them with the crude calculus of self-interest. (Le
Grand, 1997, 334).
Titmuss concluded that blood collection systems should be centralized and should
rely exclusively on voluntary donations (Titmuss, 1970). Many of Titmusss
arguments have been critiqued for being biased and for lacking sufficient
justification.2 Nonetheless, his book has been profoundly influential.
Titmusss book received much attention, not only for his indictment of the
American blood banking system, but because of his challenge to the superiority of the
market model. Moreover, he did not confine his critique of the market model to only
the blood market, he also strongly argues against the introduction of the market model
to other areas of social policy. He demonstrated that low donor rates and the
relatively high incidence of tainted blood could be primarily attributed to the United
States practice of paying blood donors.
2 For instance, both Sapolsky and Finkelstein and Le Grand note that there was not theoretical
justification for his arguments regarding a paid system being inherently more costly (Sapolsky and
Finkelstein, 1979 and Le Grand 1997).
2


Now, over 30 years after Titmuss, problems with the blood donor rate and
blood purity persist, despite the cessation of payment for donations.3 According to
the GAO, [t]he blood supply has decreased over the last decade, and there is some
evidence that in recent years the demand for blood has increased (GAO, 1999).4 The
National Blood Data Resource Center, an affiliate of the American Association of
Blood Banks, predicts sharper downturns in supply, though their estimates are
disputed by the GAO. Until very recently (as a result of the September 11th terrorist
attacks in Washington D.C. and New York), the volume of whole blood collected has
declined each year since it peaked at 14.2 million pints in 1989. However, the
available supply has not decreased as appreciably since the amount of blood rejected
during testing has declined, as has the number of autologous units collected. There
has also been a decline in the rate of donation. In 1997, about one in every 100
persons aged 18-65 had donated blood, yielding the lowest recorded level of blood
donation since 1971 (NBDRC, 1998). It is estimated that less than 5 percent of
3 Since 1978 blood has been labeled as coming from either paid or voluntary donors. This has
essentially put an end to payment for blood (see Chapter Two).
4 Reliable data on blood donation is notoriously hard to come by, as reporting on donations is
voluntary, and there is no single definitive source. Hence, official tallies of the volume of blood
donation from various sources differ.
3


healthy Americans eligible to donate blood actually donate each year (AABB,
1998).5
Reactions to the terrorist attacks of the World Trade Towers in New York and
the Pentagon in Washington D.C. on September 11,2001 included a massive
outpouring of heroic and voluntary behavior by rescue workers, medical personnel,
and ordinary citizens. Large amounts of food, clothing and money were donated.
Likewise, blood donors were waiting over five hours to donate blood, and nearly 25
percent of the U.S. population gave blood or tried to give blood directly after the
bombing (Light, 2002). News directly after the bombings indicated that the previous
blood deficits had been (at least temporarily) eliminated.
The change in the blood donation rate immediately after the bombings should
not result in complacency about the problems that have been plaguing the blood
industry. Lest we become contented, we should look to history as our guide. During
World War II, blood donation became institutionalized and synonymous with
patriotism. But after the War, blood donation dropped off precipitously. Other
dramatic events (e.g. the Gulf War, the shooting at Columbine High School in
Colorado, floods in the Midwest and the Oklahoma City Bombing) have all been met
with similarly dramatic swings in donation rates. In times of great, obvious need,
5 Eligible donors must be at least 17 years of age, 110 pounds, in good health, not have donated blood
in the last 65 days. Donors can be deemed ineligible for medical issues such as anemia; for behavioral
issues, such as having ever used intravenous drugs; or, for travel to areas with risk of diseases, such as
Malaria. Currently, individuals that are uncertain of the sterility of needles used for body piercing are
deferred for 12 months. Tattoos are also cause for a 12-month deferral.
4


Americans donate blood in ample supply. But the civic zeal appears to be transitory.
Blood donations wax and wane following dramatic national events. Hence, even as
the United States witnesses high levels of donation since September 11 we should
remember that the donor rate will likely subside, and we still will need answers to
questions about how to best motivate individuals to donate blood. In fact, a recent
Denver Post article reported that 16 of the 52 members of Americas Blood Centers
reported a critical shortage, meaning that they were down to a one-day supply of
blood (Auge, 2002). The unique opportunity afforded by these tragic events is that
they provide a natural experiment, if you will, for the studying of donor motivations.
Timely research in the wake of the September 11th bombings may reveal more about
how to get people to the blood banks than ever before. We can only hope (for a
number of reasons) that these events do not also result in a continued increase in
demand!
While the supply of donated blood, prior to the bombings, has been declining,
the demand for blood has been increasing. Reasons for the increase in demand
include more surgical procedures reliant on the use of blood, more blood used in each
medical procedure, an aging population dependent on blood products, and greater use
of blood for pharmaceutical products. The end result of this growing mismatch
between supply and demand is a blood deficit. There have been increases in delayed
surgeries as a result of the unavailability of blood and increases in local area and
regional shortages (GAO, 1999). For instance, a Denver Post (June 1999) article
5


reported on Colorados low blood supply and the decrease in the (absolute) donation
level, despite the increase in population (Chergo, 1999).
These quantity issues are augmented be equally troubling quality issues.6
While quality is probably at its all-time best with the introduction of new tests for
various diseases, concerns persist. When Titmuss was conducting his research,
Hepatitis B was first being identified in the blood supply. Now, blood centers must
contend with the presence of Hepatitis B, Hepatitis C and HIV, as well as the
emergence of other diseases that could potentially contaminate the blood supply.7 8 In
addition, the ineptitude with which the AIDS epidemic was handled has left many
concerned that emergent diseases will receive similar treatment (see Starr, 1998 and
g
Shifts, 1987). This perhaps explains many governments (in particular the U.S.A.
and Canada) strong reactions to the threat of Creutzfeldt-Jakobs Disease (CJD, or
alternatively mad-cow disease). The American Red Cross severally restricted
donations from visitors to England and parts of Europe, and the Canadian Red Cross
has adopted similar policies.
6 By quality I mean that the rate of infection (of all kinds) from blood transfusions is at an all-time
low.
7 For instance, Creutzfeldt Jakob Disease was originally viewed as a potential contaminant. However,
there is still no proof that CJD can be transmitted via the blood supply (Day, 1999).
8 Examples of the problems with how the AIDS epidemic has been handled include delays in notifying
the public and delays in instituting sufficient precautions even after there was wide agreement that HIV
could be passed through the blood supply. According to Starr, the French and Japanese governments
were particularly egregious about not alerting the public (in particular hemophiliacs) about the dangers
of the blood supply. A series of court cases in France found French officials to be guilty of deception
over the quality of the product (333); they received 4-year prison sentences (Starr, 1998).
6


The result of this confluence of trends is a troubling situation in which blood
banks are having increasing difficulty finding sufficient blood to meet increasing
demand. In response to the situation, blood banks have been pursuing new policies
aimed at attracting new donors. These policies often involve the provision of
incentives designed to induce potential donors. While not forms of direct pecuniary
payment, the incentives are becoming increasingly valuable. For instance, a recent
blood drive in San Diego offered the following incentive for donation:
One free pass to the Birch Aquarium at Scripps
Commemorative Blood Drive t-shirt or Touch Light
San Diego Chargers 2-for-l Pre-season Ticket Voucher
Entry in grand prize drawing
While this incentive package is not directly convertible to cash, it seems clear that the
incentives are valuable and are bordering on having pecuniary worth.
The basic underlying argument for this dissertation is that the main methods
by which blood banks are attempting to recruit new donors are both damaging and
ineffective. Faced with falling donation rates, blood banks are investing increasing
time and money into the provision of incentives. The assumption seems to be that
individuals will give blood when their self-interest is sufficiently addressed. To some
extent, of course, this is correct. But the reliance on incentives implies an overly
simplistic reading of an individuals willingness to donate (i.e., of a personal utility
function) and ultimately a damaging view of human motivation. It is overly
simplistic because people are not motivated only by self-interest. The personal cost
7


of donation does matter, to a degree. However, there are a variety of non-pecuniary
factors that motivate the decision to donate blood. Costs function as a sort of gate,
making donation less likely when the cost of donation is high. Beyond cost,
motivations such as altruism and social capital are important and worthy of
nurturance. After the September 11th bombings, we saw that individuals were willing
to incur substantial costs (of the time and convenience variety) for donation. It seems
clear that the perceived benefit of their actions changed the decision-making calculus
about acceptable costs for donation.
The reliance on these increasingly valuable incentives to encourage blood
donation may be ultimately damaging for several reasons. First, we know from the
recent past that blood donations for payment are more likely to carry contagious
diseases. As incentives become closer to pecuniary, it seems likely (and surely
testable) that blood quality will decline. Individuals that would be more likely to give
blood because of the incentives offered are also likely to be those whose blood is at
greater risk for carrying communicable diseases.9 Second, the focus on incentives
may be a focus on the wrong motive. If it is the case that individuals donate blood
because they recognize the good accomplished by their voluntary act, then that -
rather than t-shirts is what should be emphasized. The focus on incentives has the
danger of sullying what should be a very satisfying, positive experience. When
Titmuss (1970) wrote his classic exegesis on donation, he noted that in the United
9 This is Titmusss argument, renewed.
8


States, where payment for donation was the norm, supply problems were common.
In the U.K. where donations were voluntary, supply problems were rare. He
concluded that the payment for donation was, in essence, squeezing voluntary donors
out of the market. It is plausible that something akin to this is happening again. The
United States over-reliance on the market model is precluding considerations of other
methods for increasing the supply of blood. This is surprising given the long history
between blood donation and voluntary behavior (see Chapter 2 A History of Blood
Policy).
Despite the long-standing (albeit ambiguous) reliance on altruism, there is
very little known about what factors and conditions motivate donors to donate blood
and what keeps non-donors from donating. Previous literature on motivations for
blood donation can be roughly classified into two eras. In the first era, prior to the
cessation of payment for donation, there was research that attempted to differentiate
motives for donors and non-donors. However, since donors at that time were paid for
their blood, it is likely that their motivational responses would be different than
todays donors, who are not paid.10
For the last 20 years, research on the motivations for blood donation has
focused, almost exclusively, on the differences between single-time and repeat
donors. The assumption was that if reliable donors were found, it would be more
cost-effective and safer to nurture relationships with these donors than try to recruit
10 This, of course, was one of Titmusss main arguments (1970). Payment for donation was
encouraging donation for self-interested reasons as opposed to more laudatory motivations.
9


new donors. This policy has been largely successful. However, as repeat donors age
and retire from donation, there is concern that there will not be enough new donors to
replace them. Nationwide, only 20 percent of donations come from first-time donors
(NBDRC, 1998). Perhaps because of blood banks focus on recruiting habitual
donors, there is virtually no current information on non-donors.
As a result of these observations, the basic research question addressed in this
dissertation is:
What is the structure of motivations that drives the blood donation decision?
This research question rests on the supposition that the motivations for
donation are multi-faceted and complex. That is, it is assumed that potential donors
are responding to a variety of motivations, and that there are types of donors,
relying upon similar donation strategies.
It should be noted that this dissertation is focusing on issues of acquisition of
blood, including the solicitation of new donors. The current blood distribution system
is reviewed as it relates to issues of acquisition, but is not be a major focus of this
dissertation. Moreover, the policy recommendations derived from this analysis are
designed to focus on acquisition issues in non-emergency times. Hence, although this
dissertation addresses the impact of the September 11th bombings on the structure of
10


blood donation decision making, the focus will be on improving the donor rate in
more nonnal times.
In order to adequately address this research question, the history of blood
policy is reviewed in order to ascertain the historical roots of our current predicament.
In addition, multiple literature areas that can collectively improve our understanding
of potential motivations for donation are considered. Building on this background, a
survey designed to query individuals about their motivations for blood donation is
developed. The specific plan for addressing this research question is outlined in more
detail below.
Overview
In addition to Chapter One, Introduction, Chapter Two, entitled A History
of Blood Policy, provides background on the history of the medical uses of blood
and information on the nature of blood. The current blood banking system and its
policies are a consequence of both the nature of blood products as well as the history
of past blood donation policies and customs. The chapter includes a discussion of the
relevant characteristics of blood itself that affect how it can be collected and
distributed. This is followed by a discussion of the history of blood collection in the
United States.
These historical precedents upon which the American system is based
provides much-needed context for understanding the difficulties presently faced by
11


the United States blood banking system. This discussion includes a description of the
organizational players involved in blood policy and an analysis of the major policies
and regulations governing blood donation, collection and distribution. Collectively,
this information illuminates some of the peculiarities of the United States blood
donation and collection system.
Chapter Two also reviews the old argument about the nature of blood: is it a gift
or a commodity? These alternating metaphors or understandings of blood have been
shifting throughout our long history of collecting and using blood for medical
purposes. An understanding of these divergent metaphors sheds light on the gulf in
literature about blood and blood banking. This chapter concludes with a review of
the literature examining motivations for blood donation to set the stage for a proposed
methodology for this dissertation.
Chapter Three, A Review of the Literature, provides an introduction to and
review of the relevant theoretical literature on the motivations for blood donation.
The chapter begins with an introduction to the rational choice literature, the
theoretical corpus upon which the original blood policies were established and which
underpins the necessity of payment for donation. Despite recent blood policys
reliance on altruism, much of the literature on motivations for donation assumes the
rational choice model of human behavior.
However, other factors might mediate altruism, such as trust, feelings of
community and perceptions of risk. The sustaining argument is that a more
12


enlightened (i.e., less restrictive) view of rational choice theory together with
theories of altruism and social capital can lead to a richer understanding of the
donation decision. The choice of multiple literature areas reflects the conviction that
a deeper understanding of the motivations for blood donation can best be sought
through a multidisciplinary approach. As such, there will be no single-discipline,
unifying framework; rather, multiple theories will be assessed and integrated to build
a more robust understanding of the motivations behind the donation decision.
The next body of literature reviewed is policy design. This literature area is
considered in order to investigate whether an understanding of the judgments and
decisions of the targets of public policy might improve the standard policy design
practices. The absence of an understanding of how policy targets perceive a policy is
a crucial oversight, and policy design in general could benefit by the inclusion of such
considerations.
In Chapter Four, Methodological Foundations, the analytic plan for this study
is presented. To address the research question, a two-part survey is administered to
two different populations at two different time periods before and after the
September 11th bombings. The purpose is to use multiple analytic techniques with
different populations as a means of triangulating on an understanding of motivations
for blood donation. Moreover, this design allows an inquiry into the impact of
September 11lh on the structure of decision making regarding blood donation.
13


Primarily, this study will be relying on analytic methods derived from the Judgment
and Decision Making (JDM) research tradition.
Chapter Five provides an analysis of the results from this study. This chapter
begins with basic descriptive information about the populations from both surveys.
Next, socio-demographic characteristics are compared across the surveys to assess the
basic comparability of the two populations. This is a necessary step that will permit
further between-group comparisons. Third, bivariate analyses of predictor and
control variables by both previous donation and willingness to donate are
presented. Next, several distinct groups of decision makers are identified and
described. Finally, the results of a between group analysis, before and after
September 11th, are presented.
In Chapter Six, Synthesis and Conclusions, each of the hypotheses
presented in Chapter Four is evaluated. The results are considered in relation to
insights derived from the historical analysis and the literature review. Next, a
discussion of the theoretical insights gained from this study is presented. This is
followed by a consideration of what these findings can offer to our understanding of
other voluntary medical donations including cord-blood, organ, eye, mothers milk
and others. These findings should find applicability to the expanding options in
medical donations. In addition, a discussion of the public policy relevance of this
research is explored, as are areas for further research.
14


Summary
In summary, the blood donation rate has been declining consistently for nearly
three decades (NBDRC, 1998). Traditional accounts of this decline and of blood
donor motivations have relied on the rational actor model of human behavior, and
have not proven sufficient. As a result, blood centers are in a quandary. Periodically,
blood supplies are dangerously low, and centers need useful information on the
factors that motivate an individual to donate blood.
Relying on a multi-disciplinary perspective, a range of factors that might
motivate the individual donation decision has been proposed. A more complex and
complete picture of the blood donation decision will emerge from the reliance on
lessons from the public choice, altruism and social capital literatures.
15


CHAPTER 2
A DESCRIPTION OF BLOOD
AND A HISTORY OF ITS USES
Introduction
The current United States blood banking system and its policies are a
consequence of both the nature of blood products as well as the history of past blood
donation policies and customs. This chapter provides some background materials
on the nature of blood and the history of blood banking policy. Included is a
discussion of the relevant characteristics of blood itself that affect how it can be
collected and distributed. This is followed by a discussion of the troubling history
of blood collection in the United States. The historical precedents upon which the
American system is based provides a much-needed context for understanding the
difficulties presently faced by the United States blood banking system. This
discussion includes a description of the organizational players involved in blood
policy and an analysis of the major policies and regulations governing blood
donation, collection and distribution. This chapter concludes with a review of the
16


literature on the motivations for blood donation. Collectively, this information
illuminates some of the peculiarities of the U.S. blood donation and collection
system.
Blood and Blood Products:
Some Technical Background
This next section briefly explains what occurs to a pint of blood when it is
donated. This includes an introduction to the specific products that whole blood is
broken into and the manner in which it is distributed, followed by a discussion of
the feasibility of blood substitutes.
Less than two percent of blood is transfused as whole blood. The rest is
separated into different components that are then typically transfused to different
patients (AABB, 2000). Components of blood include red blood cells, platelets,
white blood cells, cryoprecipitates and plasma. The components of blood have
differing shelf life and uses.
One of the most commonly derived components of whole blood is red blood
cells. This is the portion of the blood rich in hemoglobin (which gives blood its red
color) that transports oxygen throughout the body. Manufactured in the bone
marrow, red blood cells are continuously being produced and broken down. They
live for approximately 120 days in the circulatory system and are eventually
removed by the spleen (AABB, 2000). Patients who benefit most from
17


transfusions of Red Blood Cells include those with chronic anemia resulting from
disorders such as kidney failure, malignancies, or gastrointestinal bleeding and those
with acute blood loss resulting from trauma or surgery (AABB, 2000). Red blood
cells have a very long shelf life. They can be refrigerated for up to 42 days or
frozen and kept for up to ten years.
A second component of whole blood is platelets; primarily used as
coagulants by leukemia and other cancer patients. Platelets can either be derived
from the whole blood or via a process known as apheresis. This entails drawing
blood from patients, separating out the platelets via centrifuge and returning the rest
of the blood to the donor. With either method of extraction, platelets have a very
short life of only five days.
White blood cells are a portion of blood whose primary use has been
treatment of patients undergoing chemotherapy. These cells are responsible for
protecting the body from invasion by foreign substances such as bacteria, fungi and
viruses (AABB, 2000) and are used for infections that are unresponsive to
antibiotic therapy. They have a very short shelf life that requires transfusion within
24 hours after collection; according to the American Association of Blood Banks,
the effectiveness of white blood cell transfusion is still unclear (AABB, 2000).
An additional component derived from whole blood is cryoprecipitates.
This clotting factor used to control bleeding, often with hemophiliacs, has a shelf
life of one year (American Red Cross, 1995).
18


transfusions of Red Blood Cells include those with chronic anemia resulting from
disorders such as kidney failure, malignancies, or gastrointestinal bleeding and those
with acute blood loss resulting from trauma or surgery (AABB, 2000). Red blood
cells have a very long shelf life. They can be refrigerated for up to 42 days or
frozen and kept for up to ten years.
A second component of whole blood is platelets; primarily used as
coagulants by leukemia and other cancer patients. Platelets can either be derived
from the whole blood or via a process known as apheresis. This entails drawing
blood from patients, separating out the platelets via centrifuge and returning the rest
of the blood to the donor. With either method of extraction, platelets have a very
short life of only five days.
White blood cells are a portion of blood whose primary use has been
treatment of patients undergoing chemotherapy. These cells are responsible for
protecting the body from invasion by foreign substances such as bacteria, fungi and
viruses (AABB, 2000) and are used for infections that are unresponsive to
antibiotic therapy. They have a very short shelf life that requires transfusion within-
24 hours after collection; according to the American Association of Blood Banks,
the effectiveness of white blood cell transfusion is still unclear (AABB, 2000).
An additional component derived from whole blood is cryoprecipitates.
This clotting factor used to control bleeding, often with hemophiliacs, has a shelf
life of one year (American Red Cross, 1995).
18


controversial policy. The United States is the only country in the world allowing
payment for donation of plasma. However, this policy has resulted in making the
United States a major exporter of plasma to the rest of the world. According to
Douglas Starr, the author of Blood: An Epic History of Medicine and Commerce.
the liberal collection laws for plasma in the United States has made the U.S. the
OPEC of the blood industry (Starr, 1998, xi).
Payment for donation of plasma raises important ethical and safety issues,
amplified by the concern that the plasma collected from paid donors is likely to be
less safe than plasma donated by voluntary donors. Although payment for plasma is
a critical issue, the present inquiry will focus on blood policy only (as opposed to
both blood and plasma policy). However, some of the findings from this analysis of
blood policy might have relevance to plasma policy. Hence, this issue is briefly
addressed again in the conclusion of this dissertation.
The main observation to be drawn from this cataloguing of products made
from blood is that a wide variety of components are derived from a single donation
of blood. A single pint can be transfused to multiple patients, and in the case of
plasma, can be used for the production of a variety of pharmaceutical products.
This proliferation of uses for blood hints that even if the supply of blood had not
decreased, demand would nonetheless have outstripped supply.
20


The Supply of Blood
Blood products are regularly collected and used for medical purposes and
have been for nearly 100 years (OTA, 1985: 79). About 13.9 million units
(including approximately 695,000 autologous donations) of Whole Blood are
donated in the United States each year by approximately eight million volunteer
blood donors. These units are transfused to about 4.5 million patients per year
(AABB, National Blood Data Resource Center, 2001). And, according to a World
Health Organization report, over 90 million units of blood are collected worldwide
(Stowell and Tomasulo, 1998, 1). Of the current U.S. supply, unremunerated
voluntary donors donate the vast majority. However, a small portion is imported
from Europe (Stowell and Tomasulo, 1998, 2).
Blood is acquired through three main sources: community blood banks;
hospitals; and the American Red Cross. The American Red Cross, with its 45
regional blood centers, accounts for over 45 percent of all blood collected. The next
largest contributors are community blood centers. Forbes and Laurie estimate that
there were 189 civilian community blood centers operating in the United States
from 1988 through 1992 (1994, 392). They account for 43 percent of the blood
supply (McCullough, 1993). Thus, as McCullough noted, a small number of
regional blood centers collect most of the blood and a large number of hospitals,
each collecting a small amount of blood, account for the remainder (McCullough,
1993, 2239), roughly 12 percent.
21


The supply issues plaguing the blood collection system would be of less
concern if blood alternatives were a more likely near-term option. Blood substitutes
have the potential to be very valuable if, as anticipated, they could reduce the risk of
transmission of infections diseases, reduce cost, alleviate difficulties associated with
collection and storage, and eliminate cross-matching difficulties (Tomasulo, 1995,
10). While much excellent research has been conducted, no safe alternatives are
currently available (Winslow, 2000). Furthermore, Tomasulo identifies
disadvantages associated with the use of blood replacements, including inevitable
differences from normal physiology (1995, 10). Moreover, there is currently no
indication that an alternative to blood could provide hemodynamic stability and
hemostasis (Bowersox and Hex, 1995,49).
The United States has an idiosyncratic blood collection and distribution
system, unique in its reliance on the non-profit sector for the acquisition and
distribution of blood. The majority of blood is donated to non-governmental
agencies. As mentioned previously, the United States does not have any central
be it state or federal entity responsible for the collection and distribution of blood
(or for the collection of data regarding blood). This is highly unusual when
compared to other nations (e.g., U.K., Canada, or Japan) blood collection and
distribution systems. Nonprofit organizations are the recipients of voluntary
donations. These organizations then break down the blood into its various
components for sale and distribution to a variety of buyers. These buyers include
22


hospitals, pharmaceutical companies in the United States, and in some cases,
overseas purchasers. The blood is donated voluntarily (without pecuniary
reimbursement) and then sold. The industry reports that the charge is not for the
blood itself but rather to cover the cost of processing and testing the blood. The
AABB reports that the average national cost for processing blood is between $65
and $75 dollars (AABB, 2000). Starr reports that the average retail price for blood
is between $150 and $200 (Starr, 1998, 350). Neither Starr nor the AABB indicate
how they arrived at their figures. Hence, the discrepancy between the cost and
charge for blood replacement could reflect profit, or may be an artifact of differing
data collection methodologies. This highlights the challenges presented by the
United States lack of publicly available national reporting data on the acquisition
and distribution of blood.
History of Blood Policy
The most thorough exegesis on the history of blood is Douglas Starrs
Blood: An Epic History of Medicine and Commerce (1998). This section derives
much of its material from Starrs book. As the title suggests, the book describes the
complex entanglements of money and medicine that have plagued the blood
banking system since blood was first used for medicinal/curative purposes. Starrs
careful history provides a starting place from which to begin the review of the
history of blood policy. The second section presented is a discussion of a recurrent
23


theme that has confused and impeded the discussion of blood banking namely the
differing metaphors for blood.
For most of modem history, bloodletting has dominated medical sciences
ideas about the curative uses of blood.
Phlebotomy, or bloodletting, originated in the ancient civilizations of Egypt
and Greece, persisted throughout the medieval, Renaissance, and
Enlightenment periods, and lasted through the second Industrial Revolution.
It flourished in Arabic and Indian Medicine. In terms of longevity, no other
practice comes close. Germ theory, the basis for modem Western medicine,
was formulated about 130 years ago. The modem practice of transfusion is
about seventy-five years old. Bloodletting was faithfully and
enthusiastically practiced for more than twenty-five hundred years (Starr,
1998, 17).
There are several remnants from the long history with bloodletting. The
honored British medical journal Lancet was named after the tool used to cause the
bleeding. The striped poles outside barbershops were originally used as an
advertisement for bloodletting: the red signifying blood, the white signifying the
bandage. One of the signers of the Declaration of Independence, physician
Benjamin Rush, was a strong advocate of bleeding as a therapy. And the first
American President, George Washington, died after copious bleedings administered
for what appeared to be strep throat (Starr, 1998).
Starr detailed a couple of developments and advancements that led to the
abandonment of bloodletting. There were a series of typhoid epidemics in Britain in
the early 1830s. Typhoid made patients quite weak, and the bleeding would cause
fainting. Doctors could not help but notice that their ministrations were not having
24


positive effects on their patients. A significant advancement was the development
of an area of inquiry called medical statistics, which called for measurement and
record keeping as opposed to reliance on the impression of physicians. Finally, the
three giants of bacteriology Louis Pasteur in France, Joseph Lister in Scotland,
and Robert Koch in Germany showed that microbes, not humors or other
intangibles, cause disease. Germ theory became the basis of modem medicine
(Starr, 1998, 29).
In the late 17th century, important breakthroughs in the understanding of the
human body were achieved that helped to lead the way to the idea of transfusing
blood from one living organism to another. One such breakthrough was discovery
of the difference between arteries and veins and the recognition that the heart was a
simple pump. This revolutionized scientific thinking about the body, and led to
some early experimentation mostly with transfusions between dogs. In the 17th
and 18th centuries, however, blood was still believed to contain the essence of the
creature from which it came, and hence thinkers at this time believed that
characteristics or tendencies could be passed through the blood. To test this,
experiments were designed in which blood was drained from one dog until it was
nearly dead. Then the blood from a second dog was transfused to the first.
Observers noted that the dog virtually came back from the dead (Starr, 1998, 9).
Blood was transfused from several dogs into one dog, and from one species to
another. Scientists pondered questions such as Would a fierce dog become tame?
25


by being ... stocked with the blood of a cowardly dog .. ? Would a trained dog
forget how to fetch if transfused from an animal that did not know how? Would the
recipients fur color somehow change to that of the donor? (Starr, 1998, 9, citing
Boyle, Robert, 1666k
Advances were also made with the instruments of transfusion. Reeds, an
early favorite, were replaced by a series of silver cylinders with a sack between the
cylinders. Attempts were also made at stitching the donating artery to the receiving
vein (Starr, 1998, 10).
While transfusion was hypothesized to be a potential cure for a wide variety
of illnesses, physicians in the 17th century were much against the use of human
blood.
It would be a very barbarous Operation, to prolong the life of some, by
abridging that of others. Animals on the other hand, did not seem to suffer
unduly from giving blood, and farmers could provide a limitless supply.
Beyond that, animal blood must surely be healthier than mans, which
undoubtedly was debased by debauchery and irregularities in eating and
drinking. After all, sadness, Envy, Anger, Melancholy and Disquiet...
corrupt the whole substance of the blood (Starr, 1998, 11 citing Boyle,
Robert, 1666).
Antione Mauroy was the first modem recipient of the blood transfusion.
Thought to be mad, he was found in 1667 wandering the streets of Paris, naked
(Starr, 1998, 3). He was taken to Jean-Baptiste Denis, a physician to Louis XIV
who had been experimenting with transfusing blood from animal to animal and had
been toying with the idea of transfusing blood from animal to human. Denis
26


transfused calf blood into Mauroy via a silver tube. Mauroy went back to work the
next day. Two days later, he received a second transfusion; larger than the first. It
was by luck that Mauroy survived the shock his body experienced as a result of this
second transfusion (Starr, 1998, 6).
Denis conducted several other animal-to-animal transfusions during 1666
and reported his results in Philosophical Transactions (July 22, 1667). He received
a scathing response from a colleague across the channel who had conducted some of
the early transfusion experiments with dogs. This rivalry between the French and
English physicians and an unfortunate lawsuit (described below) would put an end
to transfusion for nearly 150 years.
Once again Mauroy, the recipient of the first transfusion, and his wife,
Perrine, showed up on Deniss doorstep requesting a third transfusion. Denis
refused since Mauroy did not appear strong enough to withstand the transfusion.
Sometime later, Denis received a conciliatory letter from Perrine asking him to
exercise the charity to come to her home (Starr, 1998, 14). When Denis arrived
he found everything prepared for him to conduct a transfusion (including his
assistant and tools) on a very sick Mauroy. Denis relented, but as he was about to
begin, Mauroy began shaking, and the transfusion never occurred. The next
27


morning Mauroy died.3 The news of Mauroys death spread quickly throughout
France, where detractors began distributing books and pamphlets accusing Denis of
murder. Perrine attempted to blackmail Denis, and in an attempt to clear his good
name Denis filed a libel suit against Perrine. Denis was found not guilty, and it
was discovered that Perrine had been poisoning her husband with arsenic (Starr,
1998)!
As an apparent afterthought to the ruling, the judge also ruled that all future
transfusions would require permission from the Faculty of Medicine, a conservative
and powerful group of physicians (Starr, 1998, 15). Since this group was not in
favor of transfusions, this ruling effectively put an end to transfusion for over 150
years. Starr reports, transfusions were effectively terminated in Europe, when
[t]wo years later, the French parliament officially banned all transfusions involving
human beings, with the English following suit. When two men died from
transfusion in Rome, the pope banned the practice throughout most of Europe
(Starr, 1998, 15).
In the intervening one and a half centuries without transfusion, significant
advances would be made in the understanding of blood and in the art of medicine
that would improve the success of transfusions when transfusions would reemerge
3 It appears that the blood transfusions might actually have been beneficial for Mauroy. An observer
to the experiments concluded that Mauroy had syphilis. Syphilis is caused by... a bacterium that
cannot tolerate high temperatures... Dennis might have triggered in Mauroy a strange but feasible
chain of events: He gave the transfusion; the patient reacted; the fever that nearly killed him impeded
the bacteria. And for a couple of months, the madman was sane (Starr, 1998, 16).
28


as a medical procedure in the twentieth century. This hiatus in the practice of
transfusions did not prevent theorizing about transfusions. For instance, Erasmus
Darwin (the father of Charles) published some speculations on the uses of
transfusion in 1794 (Starr, 1998, 36).
Then, in 1818, the first modem transfusion of human blood occurred and
the patient died. The physician was James Blundel and he had been concerned
about the high mortality rates of hemorrhaging mothers. His experimentation on
animals would lead to two important decisions, (1) only human blood should be
employed, and (2) transfusions should not be used to cure madness or change
character, only to replace blood (Starr, 1998, 37). Brundels subsequent
experimentation with transfusions, from human to human, would renew interest in
blood transfusion. New transfusion tools were invented and improved upon. [B]y
the second half of the nineteenth century, transfusion was becoming popular again,
with hundreds reported throughout Europe (Starr, 1998, 38). Transfusions were
attempted during the American Civil War for leg amputations. During Canadas
great cholera plague, some doctors gave milk transfusions in the belief that the
white corpuscles of milk were capable of being transformed into red blood
corpuscles (Starr, 1998, 38 citing, Jennings, 1888). These early transfusions were
all done without an understanding of blood types, methods for the prevention of
clotting, or of the importance of sanitation.
29


In 1900, Viennese pathologist Karl Landsteiner differentiated the four blood
types (A, B, AB, and O) based on clever and methodical experiments in which he
combined blood from various donors and noted the presence or absence of clotting.
Landsteiner won the Nobel Prize for Medicine in 1930 for his discovery of human
blood groups (Starr, 1998, 39). However, his work did not affect the world of
transfusion for some time.
Direct transfusion, whereby the artery of the donor was sutured to the vein of
the recipient, was the standard method for preventing clotting during transfusion
(this worked since it avoided having the blood come in contact with air). Advances
were made in this procedure. For instance, Dr. George Washington Crile, in
Cleveland developed a metal ring through which the recipients vein could be
drawn and cuffed making it easier to attach the donors artery (Starr, 1998, 41).
This painful, messy method was abandoned when, in 1913, Dr. Edward Lindeman
of New Yorks Belleveue Hospital developed the multiple-syringe that
circumvented the need to cut open the patients arm. He slipped a sharp, hollow
needle into the arms of patient and donor, puncturing the skin and entering the
veins. The needles remained in place while he shuttled back and forth from donor
to recipient, withdrawing and reinjecting blood with a syringe (Starr, 1998, 43).
These advances were quickly followed by others including rubber tubing between
30


the syringes, valves to control the flow of blood and safe forms of anticoagulants.4
From here, transfusion became a standard part of medical practice.
The concept of banking blood was the idea of an employee of the British
Red Cross in 1921, Percy Lane Oliver (Starr, 1998, 53). Before this time, blood was
being transfused on the spot through donations from friends and relatives. Oliver
had the idea that a city bureau could be established that could have volunteers who
had been both screened and typed, and ready to donate should Oliver give them a
call. In the first full year of operation he had 13 calls for blood just four years
later there were nearly 800 calls (Starr, 1998, 54).
The increased demand for blood can be largely attributed (according to
Starr) to Dr. Geoffrey Keynes, brother of economist John Maynard Keynes. Dr.
Keynes was a prominent British surgeon who had become interested in transfusions
during World War I, having learned the techniques from American physicians
stationed in Europe (Starr, 1998, 55). Dr. Keynes was responsible for introducing
transfusion into standard British practice and for publishing the first modem
textbook on the subject in 1941 (Keynes, 1941).
The collaboration of Percy Lane Oliver and Dr. Geoffrey Keynes marked the
beginning of modem blood banking. Oliver became an advocate for the donors who
at times were receiving unfavorable treatment from physicians. Physicians assumed
that donors were there for pay (they were not), and hence did not use the most
4 See, Starr (1998) pages 47-49 for an in depth discussion of these advances.
31


humane of methods (for example, they surgically opened the vein before insertion
of the needle). Oliver insisted on the humane use of a needle for the extraction of
blood. He engaged in efforts to educate physicians about the voluntary status of the
donors, since many physicians at this time assumed that donors were being paid.
Oliver also emphasized the personal nature of the donation, stressing the link
between donor and recipient... (Starr, 1998, 55). He reprinted thank you letters
from donors and their families in his quarterly newsletter5 of the in order to
illuminate the relationship between donor and recipient. Olivers major contribution
to blood donation practices was recognition of the humanity of the voluntary act.
Largely because of the success of Olivers and Keynes efforts in London,
similar arrangements were established around the world in the late 1920s and 1930s.
Interestingly, while much of the industrialized world (e.g., England, France, Russia,
Germany, Japan, Austria, Belgium, Australia, etc.) was developing blood donor
systems that relied on voluntary donations the United States was unique in its
reliance on and acceptance of payment for donation. Payment for donation in the
U.S. was consistent with the market model that dominated thinking at the time. In
1923, the New York Times reported that the standard payment was $35-$50, a
healthy sum in those days (Starr, 1998, 58)! A New York Times article described
professional blood donation as a potential supplement to a career field. During
5 The letters were printed in the Blood Transfusion Services Quarterly Circular, of the British Red
Cross.
32


the Great Depression, blood donation gained in popularity, and, according to Starr,
blood donation bureaus had established themselves across New York City,
collecting blood from indigents (Starr, 58, 1998).
A group of New York City doctors was outraged that blood was being
collected from populations likely to carry syphilis and other diseases. To address
these concerns, they established the Blood Transfusion Betterment Association,
whose goal was to raise the standards of professional donation (Starr, 1998, 58).
The Association established a highly disciplined system of donation, requiring
donors to have current physical examinations, syphilis tests and a telephone so
that donors could be reached when needed. The Association attempted to exclude
donors that had a history of communicable diseases, or drug or alcohol abuse. They
established a maximum frequency of donation (no more than one time in five
weeks), staunchly advocated healthy living, encouraged exercise and cleanliness of
their donors, and would not allow donations from donors that did not appear to be
meeting their standards (Starr, 1998, 60).
Starr concluded that both the voluntary system in London and the
remunerated system in New York City were producing favorable results in this era
(late 1930s). Both systems were acquiring and distributing safe blood for patients in
need. However, since the United States system charged money for the blood that it
provided ($35 plus $6 commission for the Association), funds were available for a
more professional approach (e.g., with a medical director).
33


During World War II, there was a rapid growth of the blood industry and the
development of blood banks (where the donated blood was actually stored). Starr
reported that despite Britains extensive preparations for World War II, it was
surprisingly complacent about blood. London still relied on donors-on-the-hoof,
a system that, barely adequate in peacetime, would be swamped in the first days of
any conflict (Starr, 1998, 84).
The United States first efforts at large scale bleeding came in 1940 with
the shipment of plasma to allied troops in Britain (Starr, 1998, 93). Plasma was sent
instead of whole blood because of its longer shelf life,6 and because plasma, unlike
whole blood, could be used regardless of the blood type of either donor or recipient
(Dulles, 1950, 414). Dried plasma had the additional advantage of not requiring
refrigeration. Methods had been developed to turn the plasma into a dry powder,
that, when mixed with distilled water, could be administered anywhere along the
battle front (Dulles, 1950, 414). By 1941, both dried plasma and albumin were
being sent to the front lines. These two products required more blood donations per
unit of transfusable blood and so the demand for blood rose yet again; the demand,
of course, was assisted by the increasing casualty rate. The Red Cross began setting
up more donation centers and began an extensive advertising campaign (e.g., He
gave his blood. Will you give yours?) (Starr, 1998).
6 Fresh plasma has a 28 day shelf life.
34


A serious social problem was plaguing U.S. blood donation at this point -
the controversy over colored versus white blood. The Red Cross, in accordance
with military desires, was turning away African-American donors. After Pearl
Harbor, the Red Cross was able to convince the military to accept African-American
blood so long as it was labeled as such. The justification was that [t]his way those
receiving transfusions may be given blood from those of their own race (Starr,
1998, 108). Understandably, this policy was considered controversial, and by many,
offensive. In 1942, the Red Cross held a meeting to reconsider its position and
concluded that there was no difference in the blood of the races, yet, found it
impossible to overcome the assumption that most men of the white race objected to
blood of Negroes injected in their veins (Starr, 1998, 109 citing Fletcher, History
of the American National Red Cross).7
The United States population was not alone in its racism. The Germans
were hampered by their unwillingness to use anything other than Aryan blood,
and the Japanese were still using arm-to-arm transfusion, as Starr implies, for blood
purity reasons (1998, 115).
Despite these social problems, the war effort led to great advances in blood
banking. By the middle of the war, the United States had elaborate systems for
collecting blood, shipping it to companies able to convert the blood into dried
7 The year of publication for Fletcher was not listed by Starr. Despite numerous attempts the original
document has not been identified.
35


plasma, packaging and shipping it off to the war effort. Likewise, albumin was
being shipped in a lighter package and an easier to use liquid form. There was
extensive cooperation between universities and industry. Up until this point, the
United States still did not have any government intervention (at any level of
government) in the market for blood.
Around this same time, serious scientific research was searching for
replacements for blood. In 1942/1943, plasma and albumin were being routinely
used. However, some military surgeons were becoming convinced that whole blood
was needed for the American troops. When their requests were repeatedly denied,
they endeavored to establish their own system for providing whole blood to troops
in North Africa. Starr speculates that the denials were likely because of the
logistical difficulties associated with transporting the whole blood and because the
current Army Surgeon General, Norman T. Kirk, lacked the patience for laboratory
science. As far as he was concerned, the question of transfusion had been answered
with plasma, and was loath to reopen it (Starr, 1998, 127). Dr. Edward Churchill, a
consultant to the Army regarding the troops in North Africa, believed that part of
the resistance from official channels was also a result from the U.S. Armys
investment in blood substitutes. Churchill wrote:
A huge vested interest had been built up starting from assumptions and
erroneous thinking ... Civilians were busy helping with the war effort and
many had their prestige at stake in their collection and use of plasma.
Publicity had been launched to provide plasma for the wounded soldiers, and 8
8 For a discussion of earlier, less scientific endeavors, see Starr, 1998.
36


the Red Cross as well as the N.R.C. [National Research Council] was behind
it. Edwin Cohn9 was working to improve plasma and was trying to get
albumin solution into production ... It gained size and momentum like a
rolling snowball (Starr, 1998, 127 citing Churchill, 1972).
Despite these official impediments, Churchill and his colleagues set up their
own system in North Africa. Donors received $ 10 per pint, and the blood was not
tested for diseases as was traditionally done. Supplies were harvested from
whatever sources were available. This work was all done outside the normal chain
of command. Ad hoc blood banks were established near battles in Sicily in the
summer of 1943. Churchill repeatedly attempted to receive assistance and approval
from Washington, which was not forthcoming. He became desperate and
encouraged the New York Times to print an article that argued that plasma was not
medically sufficient and that blood banks that provided whole blood (as opposed to
plasma) were saving lives. The article was entitled, Live-blood Banks Save
Soldiers Lives in Sicily When Plasma Proves Inadequate (Starr, 1998, 128,
referring to a 1943 New York Times article). Even more provocatively, Churchill
ordered supplies through official channels that had not been approved. He wrote
memos on official letterhead requesting supplies; these letters indicated that the U.S.
Surgeon General was in support of this position when, in actuality, he was staunchly
opposed (Starr, 1998, 129).
9 Dr. Cohn was a prominent physician best known for his groundbreaking work in the fractionation
of plasma.
37


By 1943, over 2.5 million packages of dried plasma had been shipped to the
combat zone, not to mention the nearly 125,000 ampoules of albumin (Starr,
1998, 121). As the war progressed, methods for collecting and shipping blood had
become more sophisticated. However, some Army and medical personnel were
concerned that plasma was not the right answer and pushed for finding ways to get
whole blood to the front was investigated and to some degree was mastered. A
highly developed system for blood was functioning by the end of the war. The U.S.
had been using blood in more forms than any other country, establishing the
foundations for a large-scale blood complex.
Immediately after the war, the blood industry began to shut down.
Donations declined as the sense of commitment and community declined after the
war. The American Red Cross (ARC) was $16 million dollars in debt (Starr, 1998,
165) from financing its national blood effort and so closed its blood collection sites.
These closures did not last long as the board of the ARC realized in 1947 that its
organization had gained celebrity and a reputation synonymous with patriotism for
efforts with blood during the war. As a means of re-establishing financial viability,
the ARC worked to create a national program. Its strategy was to establish blood
banks across the country, but initially only in cities that did not already have
established blood banks.
The American Association of Blood Banks (AABB) was founded soon after
as a reaction to what was considered by some to be predatory behavior by the
38


American Red Cross. Existing blood banks had become concerned that the ARC
would push them out of business and would gain a monopoly on blood donation and
distribution. Starr (1998, 174) observes [tjheir ostensible purpose was to trade
expertise and technical information, but the political agenda quickly became clear.
Their real raison detre was to oppose the Red Cross movement... Some
denounced the Red Cross movement as socialistic (Starr, 1998, 174). At the first
meeting of the AABB, the members passed a resolution inviting the Red Cross to
join. However, the wording of the resolution was intended to indicate what the
AABB board saw as the proper function of the Red Cross an aid in the
procurement of donors. The implication was that all else should be left to the
professionals (Starr, 1998,174).
This conflict between the Red Cross and the AABB was, basically, about
power. But it was also about differing conceptions of blood and how blood should
be managed (Starr, 1998). The AABB was premised on the belief that blood
collection should be managed and controlled by physicians. In contrast, the Red
Cross was intending to mobilize its considerable volunteer brigade to fill this role.
Second, the AABB saw blood as a commodity that should be left to the free market,
whereas, the Red Cross saw blood as being a community responsibility. The Red
Cross stressed the civic and voluntary nature of the donation. Third, the AABB
believed in local autonomy and vehemently resisted the idea of central control over
blood. The Red Cross preferred a national organization that set standards and
39


established control at the national level, aided by local chapters. The AABB
stressed technical education; the Red Cross did not. Finally, the AABB advocated a
policy of individual responsibility in which recipients of blood were responsible
for either paying for or replenishing the blood they used. The hope was that
recipients would feel an obligation to round up sufficient donors on their own, so as
to avoid the monetary fees. The Red Cross held that blood was a community
resource, available to those in need. Demands were not placed on the recipient for
replacing the blood they used.
As Starr observed, (1998, 175), there were some points of agreement. Both
organizations shared some fundamental principles: that blood should not be
considered a commodity, for example; that the processing fees be kept as low as
possible; and that the enterprise should run on a nonprofit basis (Starr, 1998, 175).
In the transition to a peace-time blood complex, a system was revealed that
was highly unorganized, unregulated, and contained large gaps and inconsistencies.
As Starr explains:
Unlike most Western European countries, which had enacted blood
legislation following the war, the Americans never established a national
policy. They left blood collection to the marketplace, with no central
coordination, inventory, or control. The American Red Cross and American
Association of Blood Banks each attempted to monopolize the market, yet
neither was able to prevail. Their inability to cooperate divided the nation
into a patchwork of territories. Each system had its own set of rules free
blood in the Red Cross areas and loaned blood in territories of the AABB.
Neither respected the rules of the other (Starr, 1998, 188).
40


These gaps and inconsistencies had left room for entrepreneurs who saw
blood as a commodity and recognized blood banking as a growth industry.
Individuals with no medical background had set up and run blood banks across the
country. They paid for donations and were under the jurisdiction of no medical or
regulatory authority. On a national level, these for-profit banks had never
constituted more than a fifth of the nations collection capacity, but in some areas
they became particularly strong (Starr, 1998, 189). These donors-for-cash were
often down on their luck, and had much greater risks of carrying diseases like
malaria and hepatitis. Moreover, some of these for-profit banks had become
notorious for their unsanitary conditions and for the poor quality of their blood.
One particularly bad case was in Kansas City. When local hospitals had
boycotted the for-profit bank and established their own blood bank, the Federal
Trade Commission stepped in. The inspector for the FTC had seen the situation
only in economic terms; public health and safety were not part of his considerations.
The FTC subsequently charged [the doctors] with illegally conspiring to interfere
with free trade (Starr, 1998, 192). In the trial that followed, the government's case
was that the physicians and hospitals were attempting to squeeze out a competitor.
The physicians case was that they had acted out of a sense of public responsibility.
Again, we see the differing metaphors about blood. The FTC saw blood as a
commodity and hence their argument was that blood should be bought, sold, and
processed like other drugs (Starr, 1998, 193). This was a very troubling metaphor
41


for the physicians and had some devastating implications. If blood were to be
considered like any other drug, then blood, and those who are involved with blood,
would be subject to other product related statutes. The most threatening of them,
Starr noted, was the Uniform Product Code, a federal regulation adopted by all the
states mandating that anything sold as an article of commerce carried an implied
warranty... If the consumer is harmed by the product... he can sue the
manufacturer for violating the implied warranty (Starr, 1998, 193). This
understandably alarmed physicians, since there was no way to be sure if, for
instance, blood contained the hepatitis virus.
The physicians contended that blood was not a commodity, but a human
organ; likewise, they defined transfusion not as an economic transaction but as a
medical procedure (or alternatively as a service). This was to no avail. The verdict
from the FTC was that the for-profit blood bank was a legal, fully licensed
business entitled to the complete protection of the law (Starr, 1998, 195).
In response to the outcry resulting from this ruling, legislation was
introduced in Congress in 1964 to exempt blood banks from federal antitrust
lawsuits. The hearings were contentious and, in the end, the bill never made it out
of committee. In addition, an appeal to a five-panel FTC commission upheld the
original ruling. This was all occurring amidst scandal in the blood blanking
industry. Blood banks were being accused of tampering with expiration dates;
hepatitis was being found in the blood; the common payment for donation was
42


causing concern over safety of the blood supply; the squabbling between the AABB
and the Red Cross had not gone unnoticed; and, the reputation of the blood banks
was turning from heroic to scurrilous (Starr, 1998). Finally, in 1969, a Federal
Appeals Court ruled that the FTC had overstepped its authority and that the blood
banks were outside of its jurisdiction.10 Starr pointed out that although the 1969
ruling allowed the physicians to form their own blood banks, it did not resolve the
issue of the status of blood. Was it a commodity, like any drug, or did it have some
special status?
At this same time, European nations had already turned to voluntary
donation and had organized national blood donation systems. After a scandal in
which a U.S. Ambassador to Japan was stabbed and then given tainted blood (he
developed hepatitis as a result), Japan also made the change to a nationalized
system, with voluntary donation. The United States, however, remained unique in
its reliance on paid donation and its lack of a central authority for blood banking. In
the mid-1960s only seven states were licensing blood banks and only five states
were inspecting them (Starr, 1998, 204). There was still no national policy.
The blood business boomed in the 1960s and 70s (Starr, 1998, 207). The
crazy-quilt fashion in which the U.S. blood industry was established meant that no
one knew precisely how much blood was collected. Starr reported that most
10 By 1969, however, the Kansas blood bank at the center of the controversy was no longer in
business.
43


estimates were well above six million pints a year in the U.S. The business was
expanding, and also bifurcating into two separate systems: one for whole blood and
another plasma. The development of plasmapheresis was the catalyst for this split.
This procedure allowed the removal and then centrifuging of the patients blood.
The red blood cells were then re-infused back into the donor. The process was
painful and time consuming but had the advantages of not causing anemia, and a
shorter turn around time before the donor could donate again.11 Donors could now
donate 104 times per year, instead of six (Starr, 1998, 208).
The plasma industry exploded, and with the explosion came corruption and
unethical practices. Donor centers were opened in the high-risk parts of downtown
areas, attracting those that were largely unfit for donation; e.g., donors were paid
with vouchers redeemable at liquor stores (Starr, 1998). A particularly unethical
practice was the gathering of gamma globulin from prisoners. Gamma globulin
could be derived from plasma, but it was far more efficient to find someone who
had been exposed to a disease and had produced a high concentration of the
antibodies in question (Starr, 1998, 210). This could be accomplished by searching
the population for individuals exposed to the right diseases. Or, you could expose
people, wait for their bodies to build up sufficient antigens, and then pay them to
extract their plasma. According to Starr (1998), the latter was regularly done with 11
11 Plasma can be replenished in a few days as opposed to the weeks it takes the body to replenish red
blood cells.
44


the prison population. The firms running the plasma drives in prisons were exempt
from federal safety inspections, since they only collected the plasma and did not
process it. Not surprisingly, disease rates skyrocketed at the prisons where these
sorts of practices were going on (Starr, 1998,211). Despite the well-known
unsanitary conditions, according to Starr, the major pharmaceutical companies
remained loyal clients.
Researchers had suspected that blood from professional donors was more
likely to be tainted by disease than blood by voluntary donors. However, the
difference in disease rates had never been determined. In 1966, Dr. Garrott Allen
published a study showing that professional blood was ten times more likely to be
infected than was voluntarily donated blood (Starr, 1998,220). Allen recommended
the use of single donors (rather than pooling blood from multiple donors), rapidly
moving away from professional donors, and a reduction in the amount of blood
given for each transfusion. He also recommended the labeling of blood that came
from high-risk populations (namely skid row and prisons).
Allens recommendations did not sit well with certain members of the blood-
banking establishment. In fact, a few months after he published his study,
the California and Los Angeles Medical Associations denounced his
suggestions as impractical, unworkable, and cause for concern. (Starr,
1998, 220)
The media soon became involved, asserting that the blood-and-plasma
industry was engaging in a game of transfusion roulette with products that might
45


transmit hepatitis (Starr, 1998, 221, citing Altman, L., 1970). Allen began a
massive letter-writing campaign, with one of the recipients being Elliot Richardson,
then- Secretary of the U.S. Department of Health, Education and Welfare.
Richardson presided over the Division of Biologies Standards, the lax regulator of
the blood banking establishment (Starr, 1998, 225). In their correspondence, Allen
presented statistics and exhorted Richardson to require the labeling of blood as
being from either paid or volunteer donors.
In 1971, Allen sent Richardson a copy of a new book on the topic of blood
donation, Richard Titmusss The Gift Relationship: From Human Blood to Social
Policy, which compared the blood donation systems between the United States and
Britain. The final chapter of Titmusss book contains an often-cited tirade against
payment for donation. In it, he first attacked the commercialization of blood first
for ethical reasons, arguing that the commercialization of blood donation was odious
because it represses the expression of altruism, erodes the sense of community...
[and] places immense social costs on those least able to bear them Titmuss, 1970,
314). Titmuss went on to critique the commercialization of blood on four other
criteria: economic efficiency, administrative efficiency, price and quality. He
concluded that on these four testable non-ethical criteria, the commercialized blood
market is bad (Titmuss, 1970, 314). He contended that commercialized blood
systems are wasteful, administratively inefficient, costly and more likely to produce
contaminated blood.
46


Critics have contended that Titmuss work was biased, comparing the worst
of the U.S. to the best of the U.K., nonetheless, The Gift Relationship hit a
responsive chord with the American public and with Secretary of DHEW
Richardson. Secretary Richardson established a task force to look at new ways of
managing the American blood supply. A few months later, President Nixon,
declaring blood a unique national resource, ordered DHEW to conduct an
intensive study of better ways to manage it (Starr, 1998, 228). Several blood reform
bills were also introduced into Congress. Several months later, Richardson gave the
FDA regulatory oversight of the blood banks. The FDA had much more authority
than the Division of Biologies Standards, this meant that the blood banks would
now be regulated and inspected.
As Starr and Titmuss pointed out, commercialization of the U.S. blood
banking industry had failed. The for-profit system was wasteful (29 percent of
blood was going bad on the shelves) and was producing unsafe blood. A more
centralized, bureaucratized system had some appeal, but was anathema to the free
market ethos of the time. So, despite the serious failures of the blood banking
system, it was not reorganized. Instead, in 1974, the HEW proposed a National
Blood Policy, an attempt to reform and unify the blood system without resorting to
federal control (Starr, 1998, 251). The industry was called upon to develop a plan.
The blood bankers founded the American Blood Commission, a nonprofit
47


voluntary body, to serve as a forum for all blood-related debates, and to issue
decisions that members would voluntarily follow (Starr, 1998, 251).
The continuing acrimony between the AABB and the Red Cross
balkanized the proceedings. Their perennial, territorial disputes were augmented
by fights over the use of the non-replacement fee an issue that the organizations
were using to characterize what they saw as differing philosophies on blood. The
AABB and Red Cross both charged monetary fees for the blood. The AABB would
waive the fee if the blood used by the patient was replaced by blood donated in their
name. The American Blood Commission sponsored a study of the replacement
fees. The report, with little factual support and much interjected politics (Starr,
1998, 253), concluded that replacement fees were tantamount to selling blood
(Starr, 1998, 253 citing American Blood Commission, 1977, Recommendation for
Unified Donor Recruitment. July 26, 1977). Of course, the AABB members were
outraged; the President, Bernice Hemphill issued a minority report, arguing either
doctrine could succeed, depending on the experience of blood banks in their region
(Starr, 1998, 253, citing National Blood Policy papers, MS C 393).
The Red Cross withdrew from the AABB, after 16 years of membership,
partially as a result of continuing conflict regarding competition for donors and the
controversy over replacement fees (Starr, 1998, 254). It left with a significant
financial debt to the AABB, since the Red Cross had withdrawn more blood from
the clearinghouse than it had replaced or paid for. Moreover, its departure disrupted
48


the replacement credits many Americans had built up. Many donors had built up
credits donating at Red Cross banks, assuming that this would waive their
replacement fees in the eventuality that they needed blood. When the Red Cross
withdrew, all AABB credits accrued at the Red Cross were eliminated. The public
was understandably angry. A series of heated letters, accusations and lawsuits
followed. The Red Cross, in turn, accused the AABB of profiteering. The American
Blood Commission voted to phase out the non-replacement fee but the
Commission had no enforcement authority. The Commission disintegrated in the
late 1970s, acrimony remained the rule and little changed.
On the safety front, the world was already witnessing the explosion of
communicable diseases amongst the hemophiliac population as a result of the large
amounts of blood required to treat their disease. A large portion of the worlds
plasma was now being imported from the poorest comers of Latin America and was
then purchased by large U.S. based companies who cornered the international
market in plasma (Starr, 1998, 241). European nations and Japan were almost
entirely dependent on the United States for their plasma. Payment for whole blood
donations in the United States was also still common, and a cause for concern.
Then, in 1978, the FDA issued a regulation that effectively put an end to
paid donation for whole blood. The regulation was a compromise position between
the status quo and an outright prohibition against paid donation. Blood banks were
now required to label each unit of blood as coming from either paid or voluntary
49


donors. Regulators had been loath to outlaw paid donation, since plasma banks
relied almost exclusively on remunerated donors. Paid blood disappeared since no
hospital was willing to buy the implicitly more risky blood. Hepatitis rates declined
rapidly, at least for type B.
Whole blood donated in the U.S. was not being shipped overseas because of
the shorter shelf life. However, it was being shipped around the country and much
profiteering was the result. A Pulitzer Prize-winning expose on the blood banking
industry by Gilbert Gaul (1989) found that some non-profit blood banks were
staging blood crisis campaigns and then selling the excess blood to areas in need
at substantial mark-up. Indicative of American capitalism, blood was becoming big
business.
By the early 1980s, the HIV/AIDS epidemic had begun. Early researchers
had made the connection between gay men and infected blood. However,
representatives for the gay community did not want gay men to be labeled
unacceptable donors. They argued for the screening of blood, not people. The
drug companies did not agree.12 There was no test for AIDS, but because of the
high degree of correlation between AIDS and Hepatitis B, screening for hepatitis
would catch most of the AIDS. The major blood banking organizations (AABB,
ARC and others) issued a document insisting that the case for blood borne
transmission was inconclusive, and offering several reasonable measures for blood
50


banks and physicians to follow. These included educating donors about AIDS,
allowing autologous donations in which patients could set aside blood for their
own future use, and discouraging donations among groups that may have a high
incidence of AIDS (Starr, 1998, 275). They did not advocate additional testing of
the blood, but instead recommended asking the donors questions about risky
behavior.
At this point, the medical establishment was confident that AIDS was
transmitted via blood, that it was highly contagious and that there was a way to
screen for it (the hepatitis test, which would cost about $5 per test). Nonetheless,
blood was not tested this was partially due to public resistance to confront the
disease. Shilts (1987) argues that this reluctance stemmed from the public
discomfort with discussing the risk factors for AIDS and because, at least in the
beginning, it was marginalized populations that were mostly likely infected
(homosexuals and intravenous drug users). The donation rates dropped, as people
feared that they could become infected by donating. In 1983, the U.S. Public Health
Service issued its first AIDS-related recommendations, including a call for
sexually active homosexual or bisexual men with multiple partners to refrain from
donating plasma or blood (Starr, 1998,277). The FDA then issued guidelines for
the blood industry recommending self-exclusion. That is, blood banks would now
p For a very thorough analysis of these issues, see, And the Band Played on: Politics. People and the
AIDS Epedmic. by Randy Shilts, 1987.
51


be required to ask more questions and have people sign a form indicting they
understood the risk of AIDS;13 however, no additional testing would be conducted.
There remained an enormous backlog of contaminated blood products. In
1984, researchers discovered that heating plasma proved effective in reducing
infection (this did not work with whole blood). The FDA permitted the phasing in
of new heated blood products (like factor VIII for hemophiliacs) and did not
require the issuance of a recall. Also in 1984, Dr. Robert Gallo announced that he
had identified the virus that caused AIDS, identifying AIDS in the blood would be
much easier. By 1985, an inexpensive test was developed (ELISA). There were
two concerns with the test: (1) the false positive rate, (2) and it could only detect
AIDS once the donor had become symptomatic. There was also growing public fear
about AIDS, as the death and disease tolls began to rise. In 1987, a poll conducted
by the American Association of Blood Banks found that 27 percent of the people
they surveyed thought they could get AIDS merely by giving blood (Starr, 1998,
317, emphasis in original).
Government regulation and oversight of the blood banks were never
extensive to begin with and under the Reagan Administration became more lax. A
blood bank could expect to be visited by the FDA about once every two years (Starr,
1998, 317). As a result of the AIDS crisis, the FDA stepped up inspections and
13 The idea behind self-selection is that at-risk individuals would identify themselves as being high
risk, and hence, would decide not to donate blood.
52


found gross violations especially in the Red Cross. In almost every region,
inspectors found that Red Cross blood banks had released blood that had failed the
ELISA test or neglected to notify the recipients of bad blood (Starr, 1998, 317).
Blood was being purposely mislabeled (reactive blood being labeled as non-
reactive) and then shipped off to pharmaceutical companies. In 1988, the FDA
secured an agreement with the Red Cross, in which it promised to correct its
deficiencies. This did not occur but was eventually remedied by a court order and a
tight schedule to which the Red Cross was required to adhere (Starr, 1998).
In 1993, the Subcommittee on Oversight and Investigations of the
Committee on Energy and Commerce of the House of Representatives held hearings
on the safety of the United States blood supply. Preceding these hearings, there had
been growing perceptions and documentation regarding the safety of the blood
supply. Not surprisingly, these hearings occurred in the early 1990s, when
information about AIDS and the spread of other communicable diseases (namely
hepatitis) via blood transfusion were becoming more widely recognized. Also,
concerns over the persistent blood supply deficit were sparking additional interest in
the issue. This subcommittee had met previously to discuss issues of blood safety
and supply. At the committee hearing, policies recommended and implemented
based on previous committee hearing recommendations were re-evaluated. Also,
the entire issue of the safety of the blood supply was revisited.
53


There was a great deal of acrimony in this subcommittee hearing, stemming
from fundamental differences over the conceptualization of blood as well as
differences in opinion on acceptable levels of risk. Furthermore, there were
disagreements over the proper role of the FDA. FDA Commissioner David Kessler
summarized the regulatory communitys position in the following manner:
FDAs belief that a voluntary agreement with the American Red Cross
would be sufficient was, I think, emblematic of our collegial approach to
regulated industry at the time. Those days are behind us. We have shifted,
from relying on voluntary agreements to the use of agreements involving
court supervision and sanctions where necessary, from concentrating on jaw
boning with industry to writing enforceable regulations (Kessler, 1993, 22).
The FDA was rather severely criticized by Congress for the perceived
shortcomings in enforcement of blood safety regulations. The lack of success of the
FDA can be viewed as a problem of poor regulatory performance or, as
Representative John Dingell (D- MI) presented it, it could be a case of insufficient
authority. If the problem was the level of authority, then the solution was to give
the FDA broader power over blood donation. The argument was that the
enforcement problems were not caused by anything the FDA had or had not done;
critics claimed that it was the partnership type approach that was the culprit.
Thus, a stricter, more regulatory approach was necessary. They argued that the
problem was that the FDA has treated the blood community differently than they
had other industries. If you accept this metaphor, then there was no reason that a
partnership was necessary-the clear solution to improving the safety of the blood
54


supply was to strictly regulate the industry, as other similar industries were
regulated. The FDAs conceptualization of blood as a commodity rather a human
product underpinned this important distinction about whether and how the industry
should be regulated.
Increased official concern over the blood quality and quantity was
evidenced from the numerous hearings and meetings on the topic in the 1990s. For
instance in 1990, 1991, and 1993, there was a series of Congressional hearings on
blood safety, focused on the FDAs role in regulating the blood collection and
distribution process. Then, in 1996, the National Academy of Sciences Institute of
Medicine released a series of monographs from the Forum on Blood Safety and
Availability.
As a result of these hearings and especially the AIDS crisis, the current
blood supply is safer.14 The FDA policies have been effective, and the tests are
becoming more accurate. Invasive questioning about risky behavior is required as a
screen before donation. The basic structure of the blood banking industry has not
changed, but there is more regulatory control now. The FDA and its Center for
Biologies Evaluation and Research, is now responsible for the establishment of
standards (this is done in collaboration with the Public Health Service) and for the
regulation of blood products and blood banks. The FDA has set more stringent
14 In fact, there has only been one identified case of AIDS resulting from blood donation in the last
three years (Associated Press, 2002).
55


guidelines with the help of its Blood Products Advisory Committee stipulating that
the committee must only deliberate over safety and efficacy not over cost
effectiveness, as in the past (Starr, 1998, 349). The committee must also include
consumers. The National Heart, Lung and Blood Institute of the National Institutes
of Healths Division of Blood Diseases and Resources supports research related to
blood transfusion issues, though they have no regulatory authority.
Still, much remains the same, and new risks are apparent. The unfolding
saga with CJD (aka mad cow disease) has given the blood industry another
opportunity to be bold in its protection of public safety. The blood industry has
been intrepid, severely restricting donations from individuals who have spent time
in Europe. These new policies are particularly challenging for some metropolitan
areas, such as New York, and for the armed forces. The armed services have
adopted even more stringent restrictions than those recommended by the FDA.
Given the mobile nature of their populations, these changes will have dramatic
impacts.
Another important trend has been an increase in the commercialization of
non-profit blood centers. Blood centers sell voluntarily collected blood to hospitals,
pharmaceutical companies and other customers. In the past, at least some blood
centers saw themselves as performing a community service; and since the
environment did not force it, the organizations focus was not on sales. However,
drastic changes in health care have made even the market for blood more
56


competitive and have put blood centers under increasing financial pressure.
Regulators and purchasers of blood are putting great scrutiny on blood centers.
Once-stable relations between blood centers and hospitals are being disrupted as
hospitals seek to buy blood from the lowest bidder (an understandable position, to
be sure, but something relatively new to blood banking/hospital relationships).
Trends are now toward more competition between blood centers in a given locale,
more adversarial relationships between the main organizational players, and
consolidation of markets.15 Continuing acrimony between the Red Cross and the
AABB manifests itself in competition for both donors and purchasers of blood, and,
battles over lucrative markets (Frantz, 1996; Clark, 1998).
On Themes and Metaphors
One of the recurrent themes through this history of blood policy is the
disagreement over the nature of blood. There are two classic views previously
discussed (one more overtly than the other) in this historical review. One view is
that blood is a human product, given altruistically. The competing view is that
blood is a manufactured product. The difference is that although the raw material
(blood) is donated by humans, it is in every other way a manufactured good. The
argument is that the process of preparing the blood products for transfusion means
that it is no different from other pharmaceutical products, and hence should be
1;> San Diego Blood Bank (1999) Personal communication.
57


regulated like any other pharmaceutical. These competing views of blood greatly
influence how the stakeholders in the blood industry believe that blood should be
regulated, and the proper role for government in relation to securing a safe supply of
blood. Below, we explore the importance of these divergent blood metaphors.
FDA Commissioner Kessler was a vocal supporter for the blood as a
manufactured commodity metaphor. In fact, he specifically encourages its
adoption. His contention is that the adoption of this metaphor would force the
recognition of the importance of regulating blood in a fashion similar to any other
pharmaceutical or manufactured products.
We are dealing with a product, the product that is going to get transfused
(Kessler, 1993, 35).
Like any regulated industry blood banks are responsible for ensuring the
safety of their products (Kessler, 1993,20).
The product needs the same kind of rigorous quality control and adherence
to good manufacturing practices that are in place for any of our regulated
industries that manufacture products (Kessler, 1993, 35).
Peter Tomasulo, M.D. (1995), in his discussion of transfusion alternatives,
mentions some of the difficulties associated with considering blood to be a
manufactured product that can be regulated by the FDA just as any other drug. First
of all, drug manufacturers try to make absolutely pure compounds with a small
number of components that have well understood effects. Compared to most drugs,
blood is complicated and nonspecific. It has too many components and too many
58


effects (desirable as well as potentially dangerous) (Tomasulo, 1995, 3). He also
points out that while there are substances in blood that are non-beneficial to the
patient, these cannot be removed (given the current state of technology) as they
often can be with commercially synthesized drugs.
In addition, the processes for ensuring the quality of source material are very
different. Tomasulo points out that it is difficult to imagine a drug manufacturer
would ever depend on interviews of thousands of volunteers over whom they have
no control for the quality of their raw materials (1995, 93). He claims that blood
products, because of their nature, cannot undergo the same end-product testing that
other manufactured drugs can (1995, 93). Finally, Tomasulo points out that drug
manufacturers are required to list all of the components of their drugs. With blood,
this is particularly difficult to do, because of its many components and because
knowing what is in the blood relies on individuals reporting information that they
may not know (e.g. that they have been exposed to a particular communicable
disease) or may be reluctant to report (e.g. that they have engaged in a risky
behavior). Moreover, it might simply be the case that the relevant information was
not requested for instance, have you been to England in the last 2 years. The
points that Tomasulo addresses is that blood is made by a human, is very complex,
and the blood industry does not have a large degree of control over the product. It
is difficult for those responsible for collecting and distributing blood to meet the
exact requirements of other drug manufacturers.
59


Lori Andrews and Dorothy Nelkin, in Body Bazaar: The Market for Human
Tissue in the Biotechnology Age, provide a third view of blood (or in fact all body
parts). They document the ways in which demands for skin, blood, placenta,
gametes, biopsies tissue and sources of genetic material are expanding (2001, 2)
and discuss the many ways in which these samples are being taken, sometimes
without consent, and used to produce lucrative biological products.16 Often the
donors received no benefit from their biological samples, and may not even be
aware of the uses to which their body samples have been put. Hence, people's
tissues, cells and genes are increasingly being perceived as natural resources to be
harvested and transformed into value-added commodities (Weiss, 2001). The
blood as natural resource metaphor is a new one, and one likely to grow as the
profits to be reaped from the genetic information contained in blood continues to
grow.
These competing metaphors for blood are, of course, nothing new. Titmuss
remarks on them (except for the blood as natural resource metaphor) in his book
The Gift Relationship. What is important is that the debate about which metaphor is
most appropriate for framing the discussion is still a contentious one. The
regulatory community seems to be currently holding sway and the blood as product
metaphor seems to be quite strong. In The Gift Relationship. Titmuss concludes
u> This is happening both in the United States and abroad. Vivid international examples include the
Chinese system where organs can be purchased from donors that may not have consented, and the
Singaporian system where donor consent is presumed.
60


that the commercial, capitalistic nature of the U.S. blood donation system
jeopardized the safety of the blood supply. Specifically, he notes that rates of
infection were higher in blood donated by paid donors than by voluntary donors.
He also noted safety differences by for- and not-for-profit blood banks. His
conclusion was that the United States needed fundamentally to rethink how it was to
conceptualize blood donations (Titmuss, 1970).
While critics have contended that Titmuss overstated his case,17 many of
Titmusss recommendations have been embraced. Since the publication of his
book, there have been fewer paid donations, and more not-for-profit blood banks.
However, now, over 30 years later, the American blood community finds itself
going back to the very metaphor and mindset that had been considered the bane of
the blood banking industry in the past. Granted, there has not been much serious
talk of paying donors (there is always some talk of this), but it is clear that some
significant reconceptualization of blood as an industry is occurring. The irony is
that both the shift away from and now back towards this commercial metaphor has
been justified as an attempt to improve the safety of the United States blood supply.
In summary, this background on blood donation has thus far reviewed the
nature of blood, the history of blood policy and some competing metaphors for
17 Sapolsky and Finkelstein (1979), for instance, critique Titmuss for overstating the deleterious
impacts of capitalism.
61


blood. Missing from this analysis is a description of blood donors, and a
consideration of previous research on why people donate blood. These two issues
will be addressed, in turn, below.
Blood Donor Motivations
There has been a modicum of research into blood donation issues, including
a large number of published surveys on the characteristics of blood donors. These
reviews have considered a wide range of factors that can be roughly clustered under
the categories of socio-demographics, personality and motivations. A brief review
of the literature is presented, describing blood donors and why individuals donate
blood, since this helps set the stage for the proposed methodology and study plan,
which will then follow.
In 1977, Robert Oswalt published a review of the literature on blood
donation (much of the literature was his), identifying motivations for and against
donation. He concluded that the results of previous studies were remarkably similar
for the last 20 years and that nothing new could be added (meaning since the 1950s).
He summarized the major motivations for donation as altruism, personal or family
62


credit,18 social pressure, and reward.19 He noted that reasons for not donating
include fear, medical concerns, apathy, and inconvenience (Oswalt, 1977, 123).
While Oswalts work focused mostly on differentiating between the motives for
donors and non-donors, he recommended that blood centers focus on the retention
and management of existing donors rather than on the recruitment of new donors.
However, much of the work reviewed by Oswalt may now be obsolete since it came
from an era of payment for blood donation.
Ten years later, Jane Piliavin wrote a review that covered relevant literature
published after Oswalts review (note, much of the work was hers). Piliavin
focused her literature review on how first-time donors became regular or habitual
donors (Piliavin, 1987).20 Her explanation was that the emergence of AIDS had
encouraged blood centers to transition their efforts from recruiting new donors to
maintaining previous donors who had safe blood. While her work provided many
useful insights, reconsideration of the blood donor literature is warranted, once
again, due to environmental circumstances.
18 Family credit refers to the system in which donors could repay use of blood with either money or
blood donations made in their name.
19 It is interesting to note that in an era of payment for donation, Oswalt (1977) did not think that the
payment was a major motivator. While there are no national data, past research indicated a very high
turnover rate after the cessation of payments. A study in New Mexico found a nearly 100 percent
turnover in the donor pool indicating that payment might have been more important than Oswalt
hypothesized. See Piliavin and Callero (1991) for a discussion of the turnover rate.
20 Piliavin (1989) claimed that there had been two important changes, as a result of AIDS, that
justified a reconsideration of the blood donor literature. First, there was a fear of getting AIDS by
donation. Second, there had been a precipitous rise in autologous blood donation.
63


Blood centers have been finding that a strict reliance on repeat donors is no
longer sufficient to maintain the current supply of blood (at least prior to the
September 11th bombings). Apparently, the older generation of donors is not
being replaced by younger donors. Once again, blood centers need to understand
what differentiates and motivates donors and non-donors. Below, we compare
briefly the findings from Oswalts and Piliavins studies as a foundation for what
information to incorporate into the present inquiry.
Both Oswalt and Piliavin found that the past literature considered a host of
demographic factors. Apparently, the early researchers thought that a demographic
profile of donors might have been used in a predictive sense, while later researchers
use demographic information either as simple descriptive information, or as control
variables in the very few cases where modeling was conducted.
Both Piliavin and Oswalt reviewed demographic factors including age,
gender, race and marital status. Piliavin found the average age of donors to be
between 33 to 38.21 She found that donorship rates declined dramatically for those
over age 50, even though 65 was the legal cutoff.22 Oswalt found that 70-75 percent
of donors were male; Piliavin found that percent to be going down, in fact, women
now predominate among current beginning donors (Piliavin, 1989, 445). Women,
21 This is based on studies that asked for the exact age. Most studies, however, asked for ages in a
range.
22 Piliavin also points out that there is currently little justification for the 65 cutoff.
64


however, were much less likely to be multi-gallon donors. Both researchers found
that donors were more likely to be married than were non-donors.
As to race, Piliavin found a higher proportion of Caucasians than other races
donate blood. The difference in donation rate by race increased between Oswalts
and Piliavins reviews. Piliavin questioned if this race disparity might be due to
blood centers recruitment practices. If blood recruitment personnel presumed that
Caucasians were more likely to donate blood, they might have solicited donations
from other races either less often or less earnestly, thereby creating a self-fulfilling
prophecy. The differences in donation rate, by race, may also be a residual of the
World War II controversy regarding race and blood.
Oswalt considered neither occupation nor education in his review of the
literature. Piliavin tried, but found that the data were not often comparable.
However, a study of 15 blood center areas found that frequent donors had incomes
roughly 30% higher than nondonors (Piliavin, 1990, 445). Also, she found that
frequent donors were better educated than occasional or non-donors. Interestingly,
very frequent donors reversed the trend. They were more likely to be high school or
trade school educated than less frequent donors (Piliavin, 1990, 445).
Both Piliavan and Oswalt reviewed the role of altruism in relation to blood
donation. Oswalt noted that altruism was the most often cited reason for the
donation decision. However, he wondered if this were an explanation or a
rationalization. Sympathetic to the egoistical account of altruism, Oswalt cited
65


research from Osborne and Bradley (1975) that stated, 77 per cent of donors
reported feeling pride, virtue, worthwhile, or having a halo for what they did
(Osborne and Bradley, 1975, 125). Piliavin seemed to agree with Oswalts
assessment and called for more theory-based research in the area of altruism.
While Piliavin seemed unimpressed with the literature on altruism, she
seemed more impressed with research on motivations for donation.23 She
considered extrinsic and intrinsic rewards and incentives for donation. Altruism, in
its pure sense, for instance, was an intrinsic reward, while social pressure was an
extrinsic incentive. Piliavin commented that the literature was contradictory on
which rewards and incentives were most effective.24
A few studies looked at perceived community needs and support as
motivation for donation (Piliavin, 1990). Building on such work, Piliavin found that
communities that perceived strong support for donation were rewarded with higher
donation levels (1990). In addition, social pressure had been considered as a
motivator for donation (Drake, Finkelstein and Sapolsky, 1982). Finally, Piliavin
suggested that there might be something like an addiction to donation (1987, 1990).
In such instances, individuals donated blood because of the literal adrenal-high
that they experienced after donation. This high became addictive, and hence,
perpetuated the donation behavior.
2? Piliavin does not consider motivations for altruism.
66


Piliavin concluded her review of the literature on blood donation with the
following critical summary:
The quality of the research in this area is not too high. In 1977, Oswalt
noted that few studies had begun to manipulate and experimentally control
the variables involved in donating. In reviewing the studies since that time,
we have noticed little change.... [Djetails of research design, sampling,
questionnaire return rates, etc. are often missing. When the details of design
and analysis are available, the work is often of poor quality.... In addition,
data are usually presented as simple numbers or percentages of donors who
reported certain motivations or behaviors. Except in journals not easily
accessible to donor recruitment professionals, it is rare to find studies that
relate motivations to actual behavior. Also controlled analyses, such as
multiple regressions, are almost never performed on data to determine which
of many aspects of motivation or experiences are most important in affecting
donation (Piliavian, 1990, 26).
In review, there is a large body of literature on blood donor motivations.
However, the literature has several important limiting characteristics. The early
literature (pre-1975) focuses often on differences between paid and unpaid donors.
While there is some literature that compares donors to non-donors, it is difficult to
interpret its usefulness since some studies indicate a nearly 100 percent turnover rate
when payment for donation is discontinued. Such a high turnover rate implies that
the current donor pool of non-paid individuals is substantially different from the
paid donor pool of the early 1970s. By the early 1980s, blood centers (as per
Oswalt, 1977) began to focus their efforts on retaining current donors rather than on
recruiting new donors. This trend continues to the present. As a result, the bulk of 24
24 As in most literature on donor recruitment, incentives refer to those items designed to encourage
donation whereas rewards are items designed to show appreciation for past donation. In practice, the
difference is less than clear.
67


current literature focuses on differences between single-time and multiple-time
donors as did the vast majority of Piliavins work. Unfortunately, there are no
published studies after 1980 that consider the reluctance of non-donors. Moreover,
the research often lacks a sound theoretical base and there are very few cases that
use multivariate techniques.
Summary
In summary, the importance of human blood has long been recognized. For
nearly 2500 years, practically across the globe, bloodletting was a common practice
as an intervention against disease and in some cases as a form of prophylactic
medicine. In the middle of the seventeenth century, early experimentation with
transfusion occurred, with donors (and often recipients) being animals. It has only
been in the last 125 years that human-to-human transfusions have been attempted.
The advances in the short period of time are truly remarkable, and donating and
receiving blood have become a relatively commonplace activity. Since World War
II, there has been a dramatic transformation in the infrastructure surrounding the
acquisition and distribution of blood for medical purposes. Blood has become big
business. While still being reliant upon voluntary action, blood donation has
become inseparable from the profit motive. Until a viable substitute for blood in its
many uses of blood is developed, we can expect the demand for blood to continually
25 See Piliavin and Callero, (1991) for a discussion of this.
68


increase and hence we will have continued concerns about supply. Issues over the
supply of blood can not be adequately addressed without a clearer understanding of
the motivations for blood donation. The review of the literature on motivations for
blood donation highlights several important gaps. For instance, there is little
research into the motivational difference between donors and non-donors.
Moreover, limiting assumptions about the nature of human motivation plagues
much of the literature on blood donation. In the next chapter, several literature areas
are introduced that can provide insights into motivations for blood donation.
69


CHAPTER 3
LITERATURE REVIEW
Introduction
Traditional accounts of the blood donation decision presupposed the
rational actor model of human behavior. Hence, payment for donations has
traditionally been seen as the logical method for motivating the donation of whole
blood, with market forces determining the price. In 1978, the FDA required that all
blood and blood components must be labeled as coming from paid or volunteer
donors. This, in effect, eliminated the market for paid blood because paid blood
was presumed to be of poorer quality, and hospitals were concerned about the risks,
both medical and legal (Starr, 1998). A few years later, the American Association
of Blood Banks issued a policy statement in support of a reliance on a totally
voluntary blood supply.1 1
1 AABB position manual, 30.52, 1980.
70


Despite the reliance on voluntary behavior, there has been little
understanding of what factors motivate an individual to donate and keep non-donors
from donating. The predominance of the rational actor model has eclipsed the
consideration of alternative explanations of the motives driving an individuals
behavior. Other possible explanations can be found in neighboring literature areas
including: new literature in the public choice vein, altruism, and social capital. In
addition, it is useful to review the policy design literature for what it says and does
not say regarding the importance of understanding the judgment and decision-
making processes of the targets of public policy.
Rational Choice Theory
Rational choice theory is a model of human behavior that has become hugely
influential across the social sciences. In this section, the assumptions of rational
choice theory are explicated, followed by a discussion of the application of this
economic theory to the study of political behavior. Next, the major criticisms of
rational choice theory are explored. Finally, extensions of rational choice theory
are presented, including: research on social dilemmas and risk, both derived from
the Judgment and Decision-Making body of literature.
Rational choice theory first gained prominence in the 19th century in
attempts to systematize understandings of economic activity on the basis of utility
theory.
71


The theory held that every economic agent could derive a certain amount of
utility or satisfaction from any amount of any commodity. Furthermore, subject
to the limitations on available resources and information, each agent acquired
the bundle of commodities that maximized the agents utility. Rational was
thus defined as the maximization of available utility, and all agents were
assumed to be rational (Rosenberg, 1994, 74).
Utility maximization, according to James Buchanan (1972), is a meaningless
postulate until further restrictions are imposed on the definition of utility or,
technically, on the utility function. Once this step is taken, once the goods that
the individual (in some average or representative sense) values are identified, the
way is open for the derivation of hypotheses that can be tested against observation
(1987, 16). Buchanan does not describe what restrictions need to be imposed on
the utility function, but presumably, they would be rather similar to the assumptions
of rational choice theory identified by Green and Shapiro (1994). Though Green
and Shapiro are strong critics of rational choice theory, they nicely summarize the
theories assumptions, and hence their work will form the basis of the discussion on
rational choice theory, though the discussion will be augmented by the work of
traditional proponents of the theory.
Green and Shaprio (1994) present five generally accepted assumptions of
rational choice theory. First, it is assumed that individuals maximize their own
utility. Second, there is agreement that requirements of consistency are central to
rational choice theory, and should include a rank ordering of preferences and these
preference orderings should be transitive. Third, rational choice theorists generally
72


assume that individuals maximize the expected (as opposed to actual) value of their
payoffs. Fourth, there is agreement, according to Green and Shapiro (1994), that
the relevant maximizing agents are individuals. Finally, rational choice theorists
generally assume that their models apply equally to all persons under study that
decisions, rules and tastes are stable over time and similar among people (Green
and Shaprio, 1994, 17, citing Stigler and Becker, 1977, 76).
The concept of a social optimum (or Pareto optimality) is also a core tenet of
rational choice theory (Coleman, 1994). The presumption is that when individuals
pursue personal ends, they inadvertently promote the ends of society. [P]ursuit of
ones interests in exchange leads to an improvement for all those involved in the
exchange with no loss to others. When no more voluntary exchanges are possible,
a social optimum has been achieved (Coleman, 1994,41). As Green and Shapiro
(1994) summarize:
In sum, rational choice theorists generally agree on an instrumental
conception of individual rationality, by reference to which people are
thought to maximize their expected utilities in formally predictable ways. In
empirical applications, the further assumption is generally shared that
rationality is homogeneous across the individuals under study (Green and
Shaprio, 1994, 17).
One recurrent theme by proponents of rational choice theory involves
clarifying the presumption of rationality as used by rational choice theory. Herbert
Simon (1987) distinguishes between substantive and procedural rationality.
Substantive rationality is viewed in terms of the choices it produces. Procedural
73


rationality is concerned with process employed in decision making. He claims that
economics is based upon substantive rationality, and the other social sciences
presume procedural rationality. This seems to be similar to the comment made by
Buchanan, that the economic model of behavior is almost entirely predictive in
content rather than prescriptive (Buchanan, 1987, 17) and argues that most of the
criticisms of rational choice theory stem from a misunderstanding about this.
Buchanan continues:
Failure to understand the descriptive and predictive content of economic
theory along with a proclivity to interpret all social science in prescriptive
terms has caused many critics to deplore the dismal science and to rail
against the crass materialism that economic behavior allegedly represents.
The appropriate response of the economist to such criticism should be (but
perhaps too rarely has been) that he is wholly unconcerned, as a professional
scientist, about the ethically relevant characteristics of the behavior that he
examines. To the extent that men behave as his model predicts, the
economist can explain uniformities in social order. To the extent that men
behave differently, his predictions are falsified. It is as simple as that.
(Buchanan, 1987, 17).
In a further discussion of the criticism economists have received for the use
of their models outside traditional market activities, Buchanan contends that the
criticisms generally stem from a misunderstanding of the descriptive versus
predictive nature of economic reasoning. Moreover, he contends that criticisms are
a result of inappropriately conflating the distinction between is and ought
(Buchanan, 1987, 18).
74


Arrow (1987) argues for a contextual understanding of rational choice
theory, noting that rationality is not a property of the individual alone. Rather, it
gathers not only its force but also its meaning from the social context in which it is
embedded (Arrow, 1987, 201). It should be noted that Arrow also contends that
economic theory need not always be based on rationality as a matter of principle
(Arrow, 1987, 201). Not only is it possible to devise complete models of the
economy on hypotheses other than rationality, but in fact, virtually every practical
theory of macroeconomics is partly so based (Arrow, 1987, 202). He argues that
the use of rational theory in some economic models (e.g. models by Keyenes,
Friedman and Tobin) is ritualistic, not essential (203).
Green and Shapiro, citing Ferejohn (1991), distinguish between thick
versus thin accounts of rationality. In the thin account, agents are assumed to be
rational only in the sense that they efficiently employ the means available to purse
their ends (Green and Shapiro, 1994, 17, citing Ferejohn, 1991, 282). Thick
accounts of rationality are exemplified by utilitarianism and classic economics,
where the assumption has been that agents in a wide variety of situations value the
same sorts of things: for example, wealth, income, power, or the perquisites of
office (Green and Shapiro, 1994, 17, citing Ferejohn, 1991,282). Green and '
Shapiro conclude that that much of the work in rational choice theory rests on
unambiguously thick-rational assumptions (1994, 19) though there is
75


considerable variation in both what is explicitly stated and what is implicitly
assumed about the nature of rationality.
An area of dispute within the rational choice corpus concerns the amount of
relevant information that agents can normally be presumed to possess and act
upon (Green and Shapiro, 1994, 19). Zey (1998) discusses the central role of
information in making rational choices, noting that traditionally, rational choice
theory presumed perfect information. Green and Shapiro (1994) also note the
traditional assumption of the consumers ability to understand and use that
information (19). However, important research, beginning with Simons concept
of bounded rationality (1979), has demonstrated that individuals make decisions
with imperfect information, and may misuse information to which they have access.
Rational choice theory was original a theory of the market, but it had gained
significant prominence as a predictor of a wide variety of behaviors ranging from
voting to substance abuse (see Green and Kagel, 1996). The argument for the use
of rational choice theory beyond market analyses is that individuals make decisions
in the same way in all these domains.
Classic work applying economic reasoning to political issues includes Anthony
Downs An Economic Theory of Democracy (1957), and James Buchanan and
Gordon Tullocks The Calculus of Consent (1969). Their general argument is that
traditional political scientists have been studying the appropriate phenomena with
inappropriate methods. The major flaw, according to rational choice advocates, has
76


been ignoring the micro-foundations of political behavior (Green and Shapiro,
1994, 3). Green and Shapiro note that:
[RJational choice theorists have worked out the microfoundations of these
behaviors with a parsimony and rigor that has not hitherto been attempted by
political scientists. By working through the logic of the incentives
confronting political actors in a variety of structural settings, rational choice
theorists have sought to enrich our understanding of the nature of politics
and of the possibilities and limits of political reform (1994, 4).
Hence, Anthony Downs, in An Economic Theory of Democracy (1957),
conceptualizes a homo politicus that is analogous to the homo economicus
prominent in economic theory. In Downs conception, homo politicus is faced with
uncertainty about the future (and therefore does not have complete information),
and intends to strike an accurate balance between costs and returns (Downs,
1957, 9). Downs book represents one of the earliest applications of rational
choice theory outside of the market. It is important to note that Downs was not
only considering the political actor to be rational, but also political parties,
interests groups, and governments (Downs, 1957, 6). With this theoretical
framework, Downs proposes two main hypotheses: parties act to maximize votes
and ... citizens behave rationally in politics (Downs, 1957, 300).
Similarly, James Buchanan and Gordon Tullock in The Calculus of Consent
(1969) apply rational choice theory to the study of the state and develop what they
call a theory of collective choice. Relying on rather restrictive assumptions about
human nature (e.g. perfect information and rationality), they argue that although
77


decision makers often do not have complete information and do not act rationally
in the real world, these differences will tend to cancel out so that the overall
outcome will be rational. In the end, Buchanan and Tullock argue that the
application of rational choice theory does advance the understanding of political
phenomena in a way that is not possible relying on traditional social science
theories.
The major contribution of Downs (1957), and Buchanan and Tullock (1969)
is this application of economic theory to what were traditionally considered non-
market activities (i.e., the political arena). Their work opened to the door to the
application of the rational actor model to a wide array of problems in a wide variety
of disciplines. Another important contribution of this sort of work includes
attempts to consider apparently unselfish behavior in a systematic manner within
the framework of the model. Hence, issues such as volunteering, charity and other
forms of philanthropy received serious attention (see Buchanan, 1972, 19, for a
listing of some attempts). The approach required turning prescriptive norms into
testable hypotheses, which then could be subjected to the rigors of positive
analysis. This approach to the study of non-selfish behavior will receive more
attention later in the discussion of altruism.
It is perhaps the critique and extensions of rational choice theory, however,
that provide the most promise for improving our understanding of the motives for
blood donation. Critiques of rational choice theory are of two main types. First,
78


there are criticisms levied against rational choice theorys explanatory power.
Green and Shapiro summarize:
To date, a large portion of the theoretical conjectures of rational choice
theorists have not been tested empirically. Those tests that have been
undertaken have either failed on their own terms or garnered theoretical
support for positions that, on reflection, can only be characterized as banal;
they do little more than restate existing knowledge in rational choice
terminology (1994, 6).
Second, there are concerns with the behavioral assumptions of rational
choice theory (see Zey, 1992). One concern is that if the theory does not
adequately describe behavior, then it will be unable to predict action. For instance,
rational choice theorys assumption of a single preference ordering (utility) has
drawn a great deal of attention.
A person is given one preference ordering, and as and when the need arises
this is supposed to reflect his interests, represent his welfare, summarize his
ideas of what should be done, and describe his actual choices and behavior.
Can one preference ordering do all these things? (Sen, 1990, 37, emphasis in
original).
These concerns are perhaps best summarized by Sens comment that [t]hepurely
economic man is indeed close to being a social moron (Sen, 1990, 37, emphasis in
original). Equally important are the concerns that rational choice theorys
presumption of self-interested behavior does not adequately explain human
behavior. Rational choice theorists would argue that helping, giving gifts and the
like are actually motivated by selfish motives. As Zey laments, [t]he anomalies of
79


unwise, value-laden, altruistic, emotion based decisions do not limit the theory of
rational choice for its most committed adherents (1998, 88).
Another important limitation has been the assumption of value neutrality.
The idea of value neutrality has appeal to the social sciences as they have long
desired to become more scientific (Rosenberg, 1995). For instance, in A Primer
for Policy Analysis, Edith Stokey and Richard Zeckhauser, (1978) simultaneously
advocate the use of rational choice theory as a means of policy analysis, and for the
value neutrality of policy analysis without discussing the incompatibility of these
positions. Likewise, Buchanan (1972) presents a rational choice theory where the
economist (or other users of economic tools) is seen as a neutral scientist applying a
positive value free tool for policy analysis.
In summary, rational choice theory has been found to have insufficient
explanatory power, plagued by incorrect assumptions about human motivation
(Green and Shapiro, 1994). Moreover, the researchers using rational choice theory
have often and inappropriately assumed it to be value neutral. Despite these
limitations, the theory has been profoundly influential both in the academic
literature and in policy formation (Tullock, 1972).
In relation to blood policy, adherents to rational choice theory would argue
that the logical way to increase the blood supply is to pay (or in some other fashion,
properly incentivize) donors. The idea is that an appeal to the individuals self-
interest represents the best way to motivate behavior. This is in contrast to an
80


appeal to other motives that might encourage the desired behavior (e.g. altruism,
duty, community responsibility). Consistent with the self-interested view is a large
body of literature calling for the reinstatement of paid donations as a means of
expanding the blood supply (e.g. Eckert and Wallace, 1985; von Schubert, 1994;
Solow, 1971; Stewart, 1984; Cooperand Culyer, 1968).
While substantial amounts of policy analysis and theorizing are still done
within the rational choice framework, theoretical extensions abound. One of the
most promising of these areas is judgment and decision-making.
Judgment and Decision-Making
The judgment and decision-making (JDM) literature is often seen as an
extension of rational choice theory.2 JDM looks at the ways in which the
psychology of decisions differs from the deliberative model presumed by rational
choice theory. These ideas gained prominence with the publication of the oft-cited
work of Kahneman and Tversky (e.g. 1973, 1974; also Kahneman, Slovic and
Tversky, 1982) who attempt to show the biases and heuristics common in human
decision-making.3 Also, Slovics work on judgments involving risk, and JDM in
2 However, Coleman sees the heuristics and biases work by Kahneman and Tversky (1981) as a
caveat to the narrowly construed rational choice theory (Coleman, 1990, 14).
3 For a wonderful review of the criticisms of their theory see Hammond, Human Judgment and Social
Policy (1996).
81


economic/fmancial situations, has popularized these ideas to a growing and
increasingly receptive audience (see, for example Slovic, 1972, 1980).
There are two veins of research within JDM that warrant further
consideration for this dissertation. The first is the insights into social dilemmas; the
second is work on risk. The work on social dilemmas is pertinent, since it provides
an alternative framework for the blood donation problem, as well as alternative
resolutions. The work on risk is relevant because blood donation is viewed by many
potential donors as being inherently risky. This literature area provides a theoretical
framework for understanding and addressing the perceptions of risk associated with
blood donation.
Social Dilemmas. Voting, attending a political rally, or, for that matter,
donating blood are actions that are not adequately explained by rational choice
theory. Green and Shapiro (1994, 68), for instance, demonstrate the lack of insights
gained from the application of rational choice theory to voting, despite the huge
amount of empirical work on the subject. The individual vote is not likely to
determine the collective outcome, yet people regularly engage in these activities. A
given individual is not likely to receive any benefit from the act of donation
(assuming that there is no market for blood in the foreseeable future) or individually
reverse the shortage, yet the system depends on this very act.
82


Hardin nicely introduced the original concept of a social dilemma into the
literature in his description of problems with grazing rights known as the tragedy
of the commons. In this classic example, Hardin describes a situation in which
individuals share grazing rights on a common green.
Picture a pasture open to all. It is to be expected that each herdsman will try
to keep as many cattle as possible on the commons. ..Asa rational being
each herdsman seeks to maximize his gain. Explicitly or implicitly, more or
less consciously, he asks, what is the utility to me of adding one more
animal to my herd? (Hardin, 1968, 162, emphasis in original).
It is in each individuals self-interest to add one more animal to herd, since
the herdsman will receive all of the benefit of the sale of the extra cattle. The
negative consequences (externalities), however, are shared by all herdsmen. If each
individual acts without consideration of others, he or she would over-graze the
commons. This, then, is the prototypical social dilemma. Hardins solution is
mutual coercion: mutually agreed upon. But alternate solutions can be found.
Theorists from a variety of disciplines have identified alternate methods for
eliciting the cooperation necessary for resolving social dilemmas. Dawes, van de
Kragt, and Orbell (1997) summarize what they consider to be four main strategies
for resolving these sorts of social dilemmas.
1. Leviathan (Hobbes, 1651/1947). However it is established, a central state
mandates cooperation by punishing defection...
2. Reciprocal altruism (Axelrod, 1984). Through some mechanism--perhaps
biological (Trivers, 1971) cooperation on the part of one individual in a
dilemma situation enhances the probability that others will cooperate later
83


in the same situation or a similar one. Thus, an individuals enlightened
self-interest is to cooperate in hopes of eliciting reciprocity.
3. Mutual coercion mutually agreed upon (Hardin, 1968). Rather than being
punished for defection by a (potentially arbitrary) central authority, freely
choosing people agree to provide punishments to each other for choosing a
dominating defecting choice.
4. Socially instilled conscience and self-esteem (Campbell, 1975). While
externally provided payoffs may define a social dilemma, social training
can lead to such a bad conscience for choosing a dominant defecting
strategy or to such heightened self-esteem for eschewing such strategies
in favor of cooperation that the individual is better off cooperating,
irrespective of external consequences (Dawes, Van De Kragt and Orbell,
1997, 379, emphasis in original).
Dawes, Van De Kragt and Orbell (1997) summarized their presentation of
the four solutions to social dilemmas with the following remarks:
These four solutions have one characteristic in common; they turn an
apparent dilemma into a non-dilemma by manipulation (conscious or
automatic) of the consequences accruing to the individual for cooperation or
defection. Manipulation of behavior through the egoistic payoffs resulting
from such consequences is compatible with: (i) psychoanalytic beliefs in the
preeminence of primitive drives, (ii) behaviorist beliefs in the automatic and
omnipotent effects of rewards ad punishment, (iii) conservative economic
theory, (iv) social exchange theory, (v) the insistence of sociobiologists that
altruism be compatible with inclusive fitness, and (vi) the obvious success
in U.S. current society of appeals to personal payoffs (Dawes, Van De Kragt
and Orbell, 1997, 379, emphasis in original).
This listing provides a useful starting point, but it is interesting to note what
is missing from this list. The four solutions for resolving social conflicts assume
that resolutions must in some way appeal to an individuals self-interest. This is a
point, also made by Jane Mansbridge in The Relation of Altruism and Self-
Interest:
84


Other solutions derive from love or duty. That is, they require one or more
of the interacting parties either to make the others good their own or to be
committed to a principle or course of action that requires cooperation. These
two distinct motivations which I together call the unselfish or altruistic
motivations, have been variously labeled sympathy and commitment
(Sen), love and duty (Elster), empathy and morality (Jencks), we-
feeling and conscience (Dawes, van de Graft, and Ordbell), or affection
and principle (Hume) (Mansbridge, 1990, 135).
Mansbridge points out that even this list is only illustrative and not
exhaustive. She indicates that Rawls, for instance, suggests that prisoners
dilemmas can be overcome via commitment to a common purpose (1971, 260) and
Taylor (1987) suggests altruism can be a solution. From a philosophical orientation,
there are a number of possible resolutions: for instance, a rule-utilitarian approach
would lead us to conclude that cooperation could be encouraged simply by asking
those in the dilemma to formulate the rule or law that, if followed, would produce
the greatest amount of happiness for the greatest number and follow it (see Mill,
1863/1993, or Brandt 1975). The writings of Immanuel Kant also provide an
alternative strategy for resolving social dilemmas (1785/1993). Here, the decision
rule we use to choose actions would be act as though the maxim of our actions
were to become a universal law. So, one ends up with non-egoistic motivations,
not because they are only better for the individual, but because it is what morality
demands.
Alternate resolutions to social dilemmas can also be identified from a broad
reading of Robert Axelrods reciprocal altruism solution summarized by Dawes and
85


colleagues (1997). This explanation from Axelrods classic work on eliciting
cooperation in games has decidedly evolutionary roots. However, evolutionary
theorists, starting with Darwin, offer a much more nuanced understanding of
cooperation in social dilemmas.4 Axelrods approach to evolutionary theory
considers the individual to be the adaptive unit. Alternatively, we can consider both
the individual and the group to be adaptive units. The latter alternative is much less
tidy, but is also much more realistic. Sober and Wilson argue that there is a long
history of seeing groups as adaptive units, subject to the laws of evolution (1998).
If the group is the adaptive unit, then resolutions of social dilemmas no longer look
self-interested.
The purpose of this discussion is not to identify the most suitable solution,
but to recognize that alternatives to egoistic self-serving resolutions exist.
Researchers within the JDM tradition have also identified factors, not easily
identifiable as egoistic, that increase the chances of cooperation.
Dawes, Van De Kragt and Orbell (1997) find, based on ten years of research,
that group identity alone can radically alter individuals willingness to cooperate in
social dilemmas. In their game-theory style research, the authors systematically
alter the group identity of the subjects in relatively simple ways, including letting
individuals chat or dividing subjects into groups before an experiment (of a
4 The arguments will be only briefly summarized here, as they will be covered more extensively in
the discussion of altruism later in this chapter.
86


modified prisoners dilemma sort). Comparison to situations in which group
identity is not manipulated reveals dramatic increases in cooperative behavior. The
authors conclude that group identity, even in these simple games, significantly
increases cooperation: this finding, they argue, fits with an evolutionary framework
in which sociality is primary for humans (Dawes et al., 1997, 379, emphasis in
original). They conclude, [i[t is not just the successful group that prevails, but the
individuals who have a propensity to form such groups (Dawes et ah, 1997, 390).
The authors note that previous research has considered what factors have lead to
sociality, but then point out that there is no indication that people were ever not
social. Finally, they point out, as well they should, that sociality is not morality, as
numerous examples of amoral and immoral groups can be readily identified.
Interestingly, Aristotle, in his Nicomachean Ethics, is one of the first theorists to
propose that humans are social animals (zooa politica) (Aristotle, -350
B.C./1999).
Other work from JDM provides us with insights into the factors that might
enhance cooperation in social dilemmas. For instance, Axelrod notes that
communication and expectations of future interactions increase the likelihood that
an individual will adopt cooperative strategies in game situations (Axelrod, 1984).
Contrarily, Frank, Gilovich and Regan find that studying economics decreases
cooperation in games (2000).
87


Full Text

PAGE 1

GIVING BLOOD: AN EXPLORATION OF THE DETERMINANTS OF THE GIFT RELATIONSHIP by Adrianne Waldman Casebeer B.A. University of California, Irvine, 1992 M.P.P. Georgetown University, 1994 M.A. University of Arizona, Tucson, 1997 A dissertation submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs 2002

PAGE 2

2002 by Adrianne Waldman Casebeer All Rights Reserved

PAGE 3

This dissertation for the Doctor of Philosophy Degree by Adrianne Waldman Casebeer has been approved by Peter deLeon 7 l ?Nc...fl! ,, Date

PAGE 4

Casebeer, Adrianne Waldman (Ph.D., Public Affairs) Blood Donation: An Exploration of the Determinants of the Gift Relationship Dissertation directed by Professor Peter deLeon ABSTRACT While U.S. blood policy relies for its success on the voluntary actions of individuals, much of the literature on motivations for donation assumes the rational choice model of human behavior. Moreover, the provision of copious incentives for donation presumes that the most effective method for motivating potential donors is through appeals to self-interest. All the while, there is a lack of solid information on the factors that motivate the blood donation decision. This is troubling because U.S. blood donation rates are at an all-time low, as is the volume of whole blood collected. The main premise of this dissertation is that a more enlightened view of rational choice together with theories of altruism and social capital can lead to a richer understanding of the donation decision. In addition, the policy design literature is considered, to investigate if an understanding of the judgments of the targets of public policy might improve our standard policy design practices. The choice of multiple literature areas reflects the conviction that a deeper understanding of the motivations for blood donation can best be sought through a multidisciplinary approach. Derived from an historical review of blood donation policy and the multidisciplinary literature review, a series of hypotheses are proposed to shed cumulative light on the research question: "What is the structure of motives that drives the blood donation decision?". The study design employs two surveys consisting of multiple research techniques, including policy-capturing, administered before and after the September 11th bombings. The results that although stated willingness to donate blood was higher after September 111 a consistent pattern of decision making was revealed. These analyses provide support for the notion that individuals are responding to a variety of motivations; that incentives appealing to altruism would be effective for some segments of the donor population; and that more organizationally involved individuals are more likely to donate blood than are less involved individuals. IV

PAGE 5

This abstract accurately represents the content of the candidate's dissertation. I recommend its publication. Signed Peter deLeon v

PAGE 6

DEDICATION This dissertation is dedicated to my husband Bill, who provided unwavering moral support and good humor as we simultaneously endeavored to complete two dissertations while raising two rambunctious little children.

PAGE 7

ACKNOWLEDGMENT My committee members, Peter deLeon, Linda deLeon, Gary McClelland, Laurie Shroyer and Richard J. Stillman, have been wonderfully supportive throughout this project. In particular, I would like to acknowledge the contribution of my chair, Peter deLeon, for his sage advice and guidance, thorough readings of drafts and for the coincidence of having a spare copy of the The Gift Relationship on his shelf when I first approached him about this topic. Laurie Shroyer provided excellent advice and clear thinking on the methodological component of this dissertation. Linda deLeon, together with Laurie Shroyer, provided invaluable insights on questions of analytic methods. In addition, I would like to thank Gary Grunwald, Ph.D. for some remarkably clear insights into sticky analytic issues. Brad Warner, Ph.D. and Major Jim Wyznowski, Ph.D., graciously agreed to meet with me, offered excellent suggestions, and introduced me to some new analytic methods. Nina Rikowski and Nancy Kinney introduced me to valuable resources on social capital. Thanks also go to the San Diego Blood Bank and the San Diego Chapter of the American Red Cross for helpful information, interviews, and for allowing me to interview some of their donors. I could not have completed this dissertation, while raising two small children if it were not for the continual support of the Grandparents who distracted the little ones so that I could work Special thanks to Marsha Waldman for her numerous demonstrations of altruism. Without her, this dissertation would not have been possible.

PAGE 8

TABLE OF CONTENTS Figures .......................................................................................................... xii Tables ................................................................................................................... xiii CHAPTER 1. INTRODUCTION ....................................................................... 1 Introduction ................................................................................ 1 Overview ................................................................................ 11 Summary ................................................................................ 15 2. A DESCRIPTION OF BLOOD AND A HISTORY OF ITS USES ............. I 6 Introduction ............................................................................. 16 Blood and Blood Products: Some Technical Background ....................... 17 The Supply of Blood ................................................................... 21 History of Blood Policy .............................................................. .23 On Themes and Metaphors ........................................................... 57 Blood Donor Motivations ............................................................. 62 Summary ........................................................................ ........ 68 3. LITERATURE REVIEW .............................................................. 70 Introduction ............................................................................. 70 Rational Choice Theory ............................................................... 71 Judgment and Decision Making ................................................. 81 viii

PAGE 9

Social Dilemmas ............................................................................. 82 Risk .............................................................................. 88 Altruism .................. .. ............................................................ 93 Social Capital. .......................................................................... 97 The Development of Social Capital ............................................ 1 00 Trust ................................................................................ 102 Operationalizing Social Capital. ............................................... 1 04 Trends in Social Capital and Implications for Blood Donation and Other Types of Voluntary Behavior .......................... 105 Policy Design ............................................................................... } 09 Summary ............................................................................... 112 4. METHODOLOGICAL FOUNDATIONS ......................................... 116 Introduction ................................................................................................... 116 Hypotheses ...................................................................... ......... .......... ........ 117 Altruism Versus SelfInterest .. .. .. .. .... .. .. .. .. .. .. . . . . . . 118 Incentives ....... ........................... ......................................... ................. 124 Social Capital ..................................................................... 126 Trust. .................................... .. ..................... .. ......... 128 Feelings of Community ................................................ 129 Associational Involvement ............................................ 130 Analytic Techniques ........................... ....................................... 131 The Surveys ............................................................................ 139 The Study Populations .......................................................... 141 Section I: Policy Capturing ................................................... 142 Task Familiarity ........................ .................................................. 142 Task Congruence ......................................................................... 144 Refinement of Cues and Determination of Cue Quantity ........... 145 Sample Size .............................................................. 14 7 Section 2: A Query into Motivations for Blood Donation ................. 148 Socio-demographics .................................................... 150 Donation History ........................................................ 151 Hypothesis Related Variables .................................................................. 151 Altruism Versus Self-interest Hypotheses ............................ 151 Incentive Hypothesis ...................................................... 153 Social Capital Hypotheses ................................................ 154 Analytic Plan: Portion I, Policy Capturing Exercise ................ ............. 155 Analytic Plan: Portion 2, Core Survey .............................................. .......... 157 Validity and Reliability ........................................ ........................................ 157 lX

PAGE 10

Human Subjects Approval. ............................................................................. 160 Conclusion .......... .......................................................................................... 161 5. RESULTS ....................................................................................................... 162 Introduction ........................................... ..................................................... 162 The Study Populations ................................................................................... l63 Comparability of Survey I and Survey II Populations .................................. 164 Survey-by-Survey Analysis ........................................................................... 168 Results from Survey I: by Previous Donation ......................................... 169 Results from Survey I: Likely Versus Unlikely Donors ......................... 173 Results from Survey II: by Previous Donation ........................................ l75 Results from Survey II: Likely Versus Unlikely Donors ........................ 178 Structure of Decision Making: As Revealed by the Policy Capturing Exercise ...... ................................................................................. 180 Task Congruence and Task Familiarity ........................................... ........ 181 The Dimensions ....................................................................................... 183 The Regressions .............. .............. .............. ... ............... ......... .................. 184 The Factor Analyses ...................................................................................... 186 A Factor Analysis of Survey I ...... ........................................................... 186 A Factor Analysis of Survey II.. .............................................................. 188 A Factor Analysis ofthe Surveys Combined .......................................... 190 Clustering ofthe Decision Makers by Factor Analytic Results .............. 193 Cluster Analysis ............................................................................ ................ 196 Between Group Analysis-Survey I Versus Survey II ................................. 199 Conclusion ..................................................................................................... 20 I 6. SYNTHESIS AND CONCLUSION ............................................................. 203 Introduction ................................................................................................... 203 Evaluation of the Hypotheses ........................................................................ 204 Evaluating the Altruism versus Self-interest Hypotheses ....................... 204 Evaluating the Incentive Hypothesis ....................................................... 210 Evaluating the Social Capital Hypotheses ............................................... 212 Summary .................................................................................................. 217 Limitations ofthe Study ................................................................................ 218 Theoretical Conclusions ................................................................................ 221 Implications for Blood Policy ....................................................................... 232 Further Research ...................................... ..................................................... 239 X

PAGE 11

Implications for Other Voluntary Medical Donations .................... ....... ...... 243 Conclusion .......................... ...................... ................................. .... ............ 244 APPENDIX A. Summary ofPolicy-Capturing Instrument ..... ................... .... ... ......... 248 B. Section II of the Survey: Administered after the Policy Capturing Exercise ............................................ ............ 250 C. Comparison by Previous and Likely Donation: By Survey ............ ...... 256 BIBLIOGRAPHY ... ............. ........................... .............. ............ ................. 258 xi

PAGE 12

FIGURES Figlire 5.1 Distribution ofMedian Likelihood ofDonation .... ......................... l69 Figure 6.1 The Relationship Between Perceived Benefits and the Importance of Cost ......................................................................... 208 Xll

PAGE 13

TABLES Table 4.1 Opportunity Cost Versus Benefit to Others in the Donation Context ............................................................................... 123 Table 4.2 Broad Categorization of Judgment Analysis in Research Contexts: Using the Dimensions ofTask Familiarity and Task Congruence ..... 137 Table 5.1 A Comparison of Survey I and Survey II .... ................... ..... ............. 165 Table 5.2 Summary of the Dimensions Used in the Policy Capturing Exercise .......... ................................. ............ ... ...... 184 Table 5.3 A Brief Summary of the Regression Analyses ................................... 186 Table 5.4 Results from a Factor Analysis: Survey!.. .... ................................... 187 Table 5.5 Rotated Factor Pattern: Survey I ......................... ............... ........... 188 Table 5.6 Results from a Factor Analysis: Survey II ........ ...................... ........ 189 Table 5.7 Rotated Factor Pattern: Survey II ....................................................... 189 Table 5.8 Results from a Factor Analysis ofBoth Surveys Combined .............. 191 Table 5.9 Rotated Factor Pattern: Surveys Combined ....................................... 192 Table 5.10 Clusters, as Determined by the Factor Analysis, by Survey ............ 193 Table 5.11 Analysis by Cluster ............................................................... ..... .... 195 Xlll

PAGE 14

CHAPTER 1 INTRODUCTION Introduction In 1971, Richard Titmuss, in his influential book The Gift Relationship: From Human Blood to Social Policy, claimed that the major ills of the American blood banking system could be basically traced to the reliance on payment for blood donation. According to Titmuss, the result of the United States' reliance on payment for blood donation is a system that is wasteful of blood, costly, and more likely to distribute contaminated blood (Titmuss, 1970, 314). The United States' reliance on payment for blood stands in contrast to the prohibitions against paid donation in the UK and most other industrialized countries. In a 1997 afterword to a reprint of The Gift Relationship, Julian Le Grand 1 summarizes Titmuss' s four main arguments: 1. The market in blood was ... "allocatively inefficient." It was highly wasteful; it created shortages and surpluses. More significantly, it led to 1 LeGrand is the Richard Titmuss Professor of Health Policy at the London School of Economics.

PAGE 15

the production of contaminated blood, that is, it damaged the quality of the product, with socially disastrous consequences. 2. Second, this market also suffered from inefficiency in production. It was bureaucratic in operation and administratively costly. In consequence it provided blood at a much greater expense than a voluntary system would. 3. Third, the market was redistributive, but in the wrong direction. It distributed blood and blood products from poor to rich, from the disadvantaged and exploited to the privileged and powerful. 4. Finally, and most devastating of all, a market in blood was ultimately degrading for society as a whole. It drove out altruistic motivations for blood donation, replacing them with the crude calculus of self-interest. (Le Grand, 1997, 334). Titmuss concluded that blood collection systems should be centralized and should rely exclusively on voluntary donations (Titmuss, 1970). Many ofTitmuss's argliments have been critiqued for being biased and for lacking sufficient justification.2 Nonetheless, his book has been profoundly influential. Titmuss's book received much attention, not only for his indictment of the American blood banking system, but because ofhis challenge to the superiority of the market model. Moreover, he did not confine his critique of the market model to only the blood market, he also strongly argues against the introduction of the market model to other areas of social policy. He demonstrated that low donor rates and the relatively high incidence of tainted blood could be primarily attributed to the United States practice of paying blood donors. 2 For instance, both Sapolsky and Finkelstein and Le Grand note that there was not theoretical justification for his arguments regarding a paid system being inherently more costly (Sapolsky and Finkelstein, 1979 and LeGrand 1997). 2

PAGE 16

Now, over 30 years after Titmuss, problems with the blood donor rate and blood purity persist, despite the cessation of payment for donations. 3 According to the GAO, "[t]he blood supply has decreased over the last decade, and there is some evidence that in recent years the demand for blood has increased" (GAO, 1999).4 The National Blood Data Resource Center, an affiliate of the American Association of Blood Banks, predicts sharper downturns in supply, though their estimates are disputed by the GAO. Until very recently (as a result of the September 11th terrorist attacks in Washington D.C. and New York), the volume of whole blood collected has declined each year since it peaked at 14.2 million pints in 1989. However, the available supply has not decreased as appreciably since the amount ofb1ood rejected during testing has declined, as has the number of autologous units collected. There has also been a decline in the rate of donation. In 1997, about one in every 1 00 persons aged 18-65 had donated blood, yielding the lowest recorded level ofblood donation since 1971 (NBDRC, 1998). It is estimated that "less than 5 percent of 3 Since 1978 blood has been labeled as coming from either paid or voluntary donors. This has essentially put an end to payment for blood (see Chapter Two). 4 Reliable data on blood donation is notoriously hard to come by, as reporting on donations is voluntary, and there is no single defmitive source. Hence, official tallies of the volume ofblood donation from various sources differ. 3

PAGE 17

healthy Americans eligible to donate blood actually donate each year" (AABB, 1998).5 Reactions to the terrorist attacks of the World Trade Towers in New York and the Pentagon in Washington D.C. on September 11, 2001 included a massive "outpouring" of heroic and voluntary behavior by rescue workers, medical personnel, and ordinary citizens. Large amounts of food, clothing and money were donated. Likewise, blood donors were waiting over five hours to donate blood, and nearly 25 percent of the U.S. population gave blood or tried to give blood directly after the bombing (Light, 2002). News directly after the bombings indicated that the previous blood deficits had been (at least temporarily) eliminated. The change in the blood donation rate immediately after the bombings should not result in complacency about the problems that have been plaguing the blood industry. Lest we become contented, we should look to history as our guide. During World War II, blood donation became institutionalized and synonymous with patriotism. But after the War, blood donation dropped off precipitously. Other dramatic events (e.g. the GulfWar, the shooting at Columbine High School in Colorado, floods in the Midwest and the Oklahoma City Bombing) have all been met with similarly dramatic swings in donation rates. In times of great, obvious need, 5 Eligible donors must be at least 17 years of age, II 0 pounds, in good health, not have donated blood in the last 65 days. Donors can be deemed ineligible for medical issues such as anemia; for behavioral issues, such as having ever used intravenous drugs; or, for travel to areas with risk of diseases, such as Malaria. Currently, individuals that are uncertain of the sterility of needles used for body piercing are deferred for 12 months. Tattoos are also cause for a 12-month deferral. 4

PAGE 18

Americans donate blood in ample supply. But the civic zeal appears to be transitory. Blood donations wax and wane following dramatic national events. Hence, even as the United States witnesses high levels of donation since September 111 \ we should remember that the donor rate will likely subside, and we still will need answers to questions about how to best motivate individuals to donate blood. In fact, a recent Denver Post article reported that "16 of the 52 members of America's Blood Centers reported a 'critical shortage,' meaning that they were down to a one-day supply of blood" (Auge, 2002). The unique opportunity afforded by these tragic events is that they provide a natural experiment, if you will, for the studying of donor motivations. Timely research in the wake of the September 11th bombings may reveal more about how to get people to the blood banks than ever before. We can only hope (for a number of reasons) that these events do not also result in a continued increase in demand! While the supply of donated blood, prior to the bombings, has been declining, the demand for blood has been increasing. Reasons for the increase in demand include more surgical procedures reliant on the use ofblood, more blood used in each medical procedure, an aging population dependent on blood products, and greater use of blood for pharmaceutical products. The end result of this growing mismatch between supply and demand is a blood deficit. There have been increases in delayed surgeries as a result of the unavailability ofblood and increases in local area and regional shortages (GAO, 1999). For instance, a Denver Post (June 1999) article 5

PAGE 19

reported on Colorado's low blood supply and the decrease in the (absolute) donation level, despite the increase in population (Chergo, 1999). These "quantity" issues are augmented be equally troubling "quality" issues.6 While quality is probably at its all-time best with the introduction of new tests for various diseases, concerns persist. When Titmuss was conducting his research, Hepatitis B was first being identified in the blood supply. Now, blood centers must contend with the presence of Hepatitis B, Hepatitis C and HIV, as well as the emergence of other diseases that could potentially contaminate the blood supply.7 In addition, the ineptitude with which the AIDS epidemic was handled has left many concerned that emergent diseases will receive similar treatment (see Starr, 1998 and Shilts, 1987).8 This perhaps explains many governments' (in particular the U.S.A. and Canada) strong reactions to the threat ofCreutzfeldt-Jakobs Disease (CJD, or alternatively "mad-cow disease"). The American Red Cross severally restricted donations from visitors to England and parts of Europe, and the Canadian Red Cross has adopted similar policies. 6 By "quality" I mean that the rate of infection (of all kinds) from blood transfusions is at an all-time low. 7 For instance, Creutzfeldt Jakob Disease was originally viewed as a potential contaminant. However, there is still no proof that CJD can be transmitted via the blood supply (Day, 1999). 8 Examples of the problems with how the AIDS epidemic has been handled include delays in notifying the public and delays in instituting sufficient precautions even after there was wide agreement that HIV could be passed through the blood supply. According to Starr, the French and Japanese governments were particularly egregious about not alerting the public (in particular hemophiliacs) about the dangers of the blood supply. A series of court cases in France found French officials to be "guilty of deception over the quality of the product" (333); they received 4-year prison sentences (Starr 1998). 6

PAGE 20

The result of this confluence of trends is a troubling situation in which blood banks are having increasing difficulty finding sufficient blood to meet increasing demand. In response to the situation, blood banks have been pursuing new policies aimed at attracting new donors. These policies often involve the provision of incentives designed to induce potential donors. While not forms of direct pecuniary payment, the incentives are becoming increasingly valuable. For instance, a recent blood drive in San Diego offered the following incentive for donation: One free pass to the Birch Aquarium at Scripps Commemorative Blood Drive t-shirt or Touch Light San Diego Chargers 2-for-1 Pre-season Ticket Voucher Entry in grand prize drawing While this incentive package is not directly convertible to cash, it seems clear that the incentives are valuable and are bordering on having pecuniary worth. The basic underlying argument for this dissertation is that the main methods by which blood banks are attempting to recruit new donors are both damaging and ineffective. Faced with falling donation rates, blood banks are investing increasing time and money into the provision of incentives. The assumption seems to be that individuals will give blood when their self-interest is sufficiently addressed. To some extent, of course, this is correct. But the reliance on incentives implies an overly simplistic reading of an individual's willingness to donate (i.e., of a personal utility function) and ultimately a damaging view of human motivation. It is overly simplistic because people are not motivated only by self-interest. The personal cost 7

PAGE 21

of donation does matter, to a degree. However, there are a variety of non-pecuniary factors that motivate the decision to donate blood. Costs function as a sort of gate, making donation less likely when the cost of donation is high. Beyond cost, motivations such as altruism and social capital are important and worthy of nurturance. After the September 11th bombings, we saw that individuals were willing to incur substantial costs (of the time and convenience variety) for donation. It seems clear that the perceived benefit of their actions changed the decision-making calculus about acceptable costs for donation. The reliance on these increasingly valuable incentives to encourage blood donation may be ultimately damaging for several reasons. First, we know from the recent past that blood donations for payment are more likely to carry contagious diseases. As incentives become closer to pecuniary, it seems likely (and surely testable) that blood quality will decline. Individuals that would be more likely to give blood because of the incentives offered are also likely to be those whose blood is at greater risk for carrying communicable diseases.9 Second, the focus on incentives may be a focus on the wrong motive. If it is the case that individuals donate blood because they recognize the good accomplished by their voluntary act, then thatrather than t-shirts --is what should be emphasized. The focus on incentives has the danger of sullying what should be a very satisfying, positive experience. When Titmuss (1970) wrote his classic exegesis on donation, he noted that in the United 9 This is Titmuss's argument, renewed. 8

PAGE 22

States, where payment for donation was the norm, supply problems were common. In the U.K. where donations were voluntary, supply problems were rare. He concluded that the payment for donation was, in essence, squeezing voluntary donors out of the market. It is plausible that something akin to this is happening again. The United States over-reliance on the market model is precluding considerations of other methods for increasing the supply ofblood. This is surprising given the long history between blood donation and voluntary behavior (see Chapter 2-A History of Blood Policy). Despite the long-standing (albeit ambiguous) reliance on altruism, there is very little known about what factors and conditions motivate donors to donate blood and what keeps non-donors from donating. Previous literature on motivations for blood donation can be roughly classified into two eras. In the first era, prior to the cessation of payment for donation, there was research that attempted to differentiate motives for donors and non-donors. However, since donors at that time were paid for their blood, it is likely that their motivational responses would be different than today's donors, who are not paid.10 For the last 20 years, research on the motivations for blood donation has focused, almost exclusively, on the differences between single-time and repeat donors. The assumption was that if reliable donors were found it would be more cost-effective and safer to nurture relationships with these donors than try to recruit 10 This, of course, was one ofTitmuss's main arguments (1970). Payment for donation was encouraging donation for self-interested reasons as opposed to more laudatory motivations. 9

PAGE 23

new donors. This policy has been largely successful. However, as repeat donors age and retire from donation, there is concern that there will not be enough new donors to replace them. Nationwide, only 20 percent of donations come from first-time donors (NBDRC, 1998). Perhaps because ofblood banks' focus on recruiting habitual donors, there is virtually no current information on non-donors As a result of these observations, the basic research question addressed in this dissertation is: What is the structure of motivations that drives the blood donation decision? This research question rests on the supposition that the motivations for donation are multi-faceted and complex. That is, it is assumed that potential donors are responding to a variety of motivations, and that there are "types" of donors, relying upon similar donation strategies. It should be noted that this dissertation is focusing on issues of acquisition of blood, including the solicitation of new donors. The current blood distribution system is reviewed as it relates to issues of acquisition, but is not be a major focus of this dissertation. Moreover, the policy recommendations derived from this analysis are designed to focus on acquisition issues in non-emergency times. Hence, although this dissertation addresses the impact ofthe September 11th bombings on the structure of 10

PAGE 24

blood donation decision making, the focus will be on improving the donor rate in more nonnal times. In order to adequately address this research question, the history of blood policy is reviewed in order to ascertain the historical roots of our current predicament. In addition, multiple literature areas that can collectively improve our understanding of potential motivations for donation are considered. Building on this background, a survey designed to query individuals about their motivations for blood donation is developed. The specific plan for addressing this research question is outlined in more detail below. Overview In addition to Chapter One, "Introduction," Chapter Two, entitled "A History of Blood Policy," provides background on the history of the medical uses ofblood and information on the nature of blood. The current blood banking system and its policies are a consequence of both the nature ofblood products as well as the history of past blood donation policies and customs. The chapter includes a discussion of the relevant characteristics ofblood itself that affect how it can be collected and distributed. This is followed by a discussion of the history ofblood collection in the United States. These historical precedents upon which the American system is based provides much-needed context for understanding the difficulties presently faced by 11

PAGE 25

the United States blood banking system. This discussion includes a description of the organizational players involved in blood policy and an analysis of the major policies and regulations governing blood donation, collection and distribution. Collectively, this information illuminates some of the peculiarities of the United States blood donation and collection system. Chapter Two also reviews the old argument about the nature of blood: is it a gift or a commodity? These alternating metaphors or understandings ofblood have been shifting throughout our long history of collecting and using blood for medical purposes. An understanding of these divergent metaphors sheds light on the gulf in literature about blood and blood banking. This chapter concludes with a review of the literature examining motivations for blood donation to set the stage for a proposed methodology for this dissertation. Chapter Three, "A Review of the Literature," provides an introduction to and review of the relevant theoretical literature on the motivations for blood donation. The chapter begins with an introduction to the rational choice literature, the theoretical corpus upon which the original blood policies were established and which underpins the necessity of payment for donation. Despite recent blood policy's reliance on altruism, much of the literature on motivations for donation assumes the rational choice model of human behavior. However, other factors might mediate altruism, such as trust, feelings of community and perceptions of risk. The sustaining argument is that a more 12

PAGE 26

enlightened (i.e., less restrictive) view of rational choice theory -together with theories of altruism and social capital -can lead to a richer understanding of the donation decision. The choice of multiple literature areas reflects the conviction that a deeper understanding of the motivations for blood donation can best be sought through a multidisciplinary approach. As such, there will be no single-discipline, unifying framework; rather, multiple theories will be assessed and integrated to build a more robust understanding of the behind the donation decision. The next body ofliterature reviewed is policy design. This literature area is considered in order to investigate whether an understanding of the judgments and decisions of the targets of public policy might improve the standard policy design practices. The absence of an understanding of how policy targets perceive a policy is a crucial oversight, and policy design in general could benefit by the inclusion of such considerations. In Chapter Four, "Methodological Foundations," the analytic plan for this study is presented. To address the research question, a two-part survey is administered to two different populations at two different time periods before and after the September 11th bombings. The purpose is to use multiple analytic techniques with different populations as a means oftriangulating on an understanding of motivations for blood donation. Moreover, this design allows an inquiry into the impact of September II th on the structure of decision making regarding blood donation. 13

PAGE 27

Primarily, this study will be relying on analytic methods derived from the Judgment and Decision Making (JDM) research tradition. Chapter Five provides an analysis of the results from this study. This chapter begins with basic descriptive information about the populations from both surveys. Next, socio-demographic characteristics are compared across the surveys to assess the basic comparability of the two populations. This is a necessary step that will permit further between-group comparisons. Third, bivariate analyses of predictor and control variables by both "previous donation" and "willingness to donate" are presented. Next, several distinct groups of decision makers are identified and described. Finally, the results of a between group analysis, before and after September 11th, are presented. In Chapter Six, "Synthesis and Conclusions," each of the hypotheses presented in Chapter Four is evaluated. The results are considered in relation to insights derived from the historical analysis and the literature review. Next, a discussion of the theoretical insights gained from this study is presented. This is followed by a consideration of what these findings can offer to our understanding of other voluntary medical donations including cord-blood, organ, eye, mother's milk and others. These findings should find applicability to the expanding options in medical donations. In addition, a discussion of the public policy relevance of this research is explored, as are areas for further research. 14

PAGE 28

Summary In summary, the blood donation rate has been declining consistently for nearly three decades (NBDRC, 1998). Traditional accounts ofthis decline and ofblood donor motivations have relied on the rational actor model of human behavior and have not proven sufficient. As a result, blood centers are in a quandary. Periodically, blood supplies are dangerously low, and centers need useful information on the factors that motivate an 1ndividual to donate blood. Relying on a multi-disciplinary perspective, a range of factors that might motivate the individual donation decision has been proposed. A more complex and complete picture of the blood donation decision will emerge from the reliance on lessons from the public choice, altruism and social capital literatures. 15

PAGE 29

CHAPTER2 A DESCRIPTION OF BLOOD AND A HISTORY OF ITS USES Introduction The current United States blood banking system and its policies are a consequence ofboth the nature ofblood products as well as the history of past blood donation policies and customs. This chapter provides some background materials on the nature ofblood and the history of blood banking policy. Included is a discussion of the relevant characteristics ofblood itself that affect how it can be collected and distributed. This is followed by a discussion of the troubling history ofblood collection in the United States. The historical precedents upon which the American system is based provides a much-needed context for understanding the difficulties presently faced by the United States blood banking system. This discussion includes a description of the organizational players involved in blood policy and an analysis of the major policies and regulations governing blood donation, collection and distribution. This chapter concludes with a review of the 16

PAGE 30

literature on the motivations for blood donation. Collectively, this information illuminates some of the peculiarities of the U.S. blood donation and collection system. Blood and Blood Products: Some Technical Background This next section briefly explains what occurs to a pint ofblood when it is donated. This includes an introduction to the specific products that whole blood is broken into and the manner in which it is distributed, followed by a discussion of the feasibility ofblood substitutes. Less than two percent of blood is transfused as whole blood. The rest is separated into different components that are then typically transfused to different patients (AABB, 2000). Components ofblood include red blood cells, platelets, white blood cells, cryoprecipitates and plasma. The components of blood have differing shelf life and uses. One of the most commonly derived components of whole blood is red blood cells. This is the portion of the blood rich in hemoglobin (which gives blood its red color) that transports oxygen throughout the body. "Manufactured in the bone marrow, red blood cells are continuously being produced and broken down. They live for approximately 120 days in the circulatory system and are eventually removed by the spleen" (AABB, 2000). "Patients who benefit most from 17

PAGE 31

transfusions of Red Blood Cells include those with chronic anemia resulting from disorders such as kidney failure, malignancies, or gastrointestinal bleeding and those with acute blood loss resulting from trauma or surgery" (AABB, 2000). Red blood cells have a very long shelflife. They can be refrigerated for up to 42 days or frozen and kept for up to ten years. A second component of whole blood is platelets; primarily used as coagulants by leukemia and other cancer patients. Platelets can either be derived from the whole blood or via a process known as apheresis. This entails drawing blood from patients, separating out the platelets via centrifuge and returning the rest of the blood to the donor. With either method of extraction, platelets have a very short life of only five days. White blood cells are a portion of blood whose primary use has been treatment of patients undergoing chemotherapy. These cells are responsible "for protecting the body from invasion by foreign substances such as bacteria, fungi and viruses" (AASB, 2000) and are used for infections that are unresponsive to antibiotic therapy. They have a very short shelflife that requires transfusion within 24 hours after collection; according to the American Association of Blood Banks, the effectiveness of white blood cell transfusion is still unclear (AABB, 2000). An additional component derived from whole blood is cryoprecipitates. This clotting factor used to control bleeding, often with hemophiliacs, has a shelf life of one year {American Red Cross, 1995). 18

PAGE 32

Ill/ _. transfusions of Red Blood Cells include those with chronic anemia resulting from disorders such as kidney failure, malignancies, or gastrointestinal bleeding and those -!!! with acute blood loss resulting from trauma or surgery" (AABB, 2000). Red blood cells have a very long shelflife. They can be refrigerated for up to 42 days or frozen and kept for up to ten years. A second component of whole blood is platelets; primarily used as coagulants by leukemia and other cancer patients. Platelets can either be derived from the whole blood or via a process known as apheresis. This entails drawing blood from patients, separating out the platelets via centrifuge and returning the rest of the blood to the donor. With either method of extraction, platelets have a very short life of only five days. White blood cells are a portion of blood whose primary use has been treatment of patients undergoing chemotherapy. These cells are responsible "for protecting the body from invasion by foreign substances such as bacteria, fungi and viruses" (AABB, 2000) and are used for infections that are unresponsive to antibiotic therapy. They have a very short shelflife that requires transfusion within 24 hours after collection; according to the American Association of Blood Banks, the effectiveness of white blood cell transfusion is still unclear (AABB, 2000). An additional component derived from whole blood is cryoprecipitates. This clotting factor used to control bleeding, often with hemophiliacs, has a shelf life of one year (American Red Cross, 1995). 18

PAGE 33

controversial policy. The United States is the only country in the world allowing payment for donation of plasma. However, this policy has resulted in making the United States a major exporter of plasma to the rest of the world. According to Douglas Starr, the author ofBlood: An Epic History of Medicine and Commerce, the liberal collection laws for plasma in the United States has made the U.S. the OPEC of the blood industry (Starr, 1998, xi). Payment for donation of plasma important ethical and safety issues, amplified by the concern that the plasma collected from paid donors is likely to be less safe than plasma donated by voluntary donors. Although payment for plasma is a critical issue, the present inquiry will focus on blood policy only (as opposed to both blood and plasma policy). However, some of the findings from this analysis of blood policy might have relevance to plasma policy. Hence, this issue is briefly addressed again in the conclusion of this dissertation. The main observation to be drawn from this cataloguing of products made from blood is that a wide variety of components are derived from a single donation ofblood. A single pint can be transfused to multiple patients, and in the case of plasma, can be used for the production of a variety of pharmaceutical products. This proliferation of uses for blood hints that even if the supply ofblood had not decreased, demand would nonetheless have outstripped supply. 20

PAGE 34

The Supply of Blood Blood products are regularly collected and used for medical purposes and have been for nearly 100 years {OTA, 1985: 79). "About 13.9 million units (including approximately 695,000 autologous donations) of Whole Blood are donated in the United States each year by approximately eight million volunteer blood donors. These units are transfused to about 4.5 million patients per year" (AABB, National Blood Data Resource Center, 2001). And, according to a World Health Organization report, over 90 million units ofblood are collected worldwide (Stowell and Tomasulo, 1998, 1). Ofthe current U.S. supply, unremunerated voluntary donors donate the vast majority. However, a small portion is imported from Europe (Stowell and Tomasulo, 1998, 2). Blood is acquired through three main sources: community blood banks; hospitals; and the American Red Cross. The American Red Cross, with its 45 regional blood centers, accounts for over 45 percent of all blood collected. The next largest contributors are community blood centers. Forbes and Laurie estimate that there were 189 civilian community blood centers operating in the United States from 1988 through 1992 ( 1994, 392). They account for 43 percent of the blood supply (McCullough, 1993). Thus, as McCullough noted, "a small number of regional blood centers collect most of the blood and a large number ofhospitals, each collecting a small amount of blood, account for the remainder" (McCullough, 1993, 2239), roughly 12 percent. 21

PAGE 35

The supply issues plaguing the blood collection system would be ofless concern if blood alternatives were a more likely near-term option. Blood substitutes have the potential to be very valuable if, as anticipated, they could reduce the risk of transmission of infections diseases, reduce cost, alleviate difficulties associated with collection and storage, and eliminate cross-matching difficulties (Tomasulo, 1995, 1 0). While much excellent research has been conducted, no safe alternatives are currently available (Winslow, 2000). Furthermore, Tomasulo identifies disadvantages associated with the use of blood replacements, including "inevitable differences from normal physiology" ( 1995, 1 0). Moreover, there is currently no indication that an alternative to blood could "provide hemodynamic stability and hemostasis" (Bowersox and Hex, 1995, 49). The United States has an idiosyncratic blood collection and distribution system, unique in its reliance on the non-profit sector for the acquisition and distribution of blood. The majority of blood is donated to non-governmental agencies. As mentioned previously, the United States does not have any centralbe it state or federal -entity responsible for the collection and distribution of blood (or for the collection of data regarding blood). This is highly unusual when compared to other nations' (e.g., U.K., Canada, or Japan) blood collection and distribution systems. Nonprofit organizations are the recipients of voluntary donations. These organizations then break down the blood into its various components for sale and distribution to a variety of buyers. These buyers include 22

PAGE 36

hospitals, pharmaceutical companies in the United States, and in some cases, overseas purchasers. The blood is donated voluntarily (without pecuniary reimbursement) and then sold. The industry reports that the charge is not for the blood itselfbut rather to cover the cost of processing and testing the blood. The AABB reports that the average national cost for processing blood is between $65 and $75 dollars (AABB, 2000). Starr reports that the average retail price for blood is between $150 and $200 (Starr, 1998, 350). Neither Starr nor the AABB indicate how they arrived at their figures. Hence, the discrepancy between the cost and charge for blood replacement could reflect profit, or may be an artifact of differing data collection methodologies. This highlights the challenges presented by the United States' lack of publicly available national reporting data on the acquisition and distribution ofblood. History of Blood Policy The most thorough exegesis on the history ofblood is Douglas Starr's Blood: An Epic History ofMedicine and Commerce (1998). This section derives much of its material from Starr's book. As the title suggests, the book describes the complex entanglements of money and medicine that have plagued the blood banking system since blood was first used for medicinal/curative purposes. Starr's careful history provides a starting place from which to begin the review of the history of blood policy. The second section presented is a discussion of a recurrent 23

PAGE 37

theme that has confused and impeded the discussion of blood banking--namely the differing metaphors for blood. For most of modern history, bloodletting has dominated medical sciences' ideas about the curative uses ofblood. Phlebotomy, or bloodletting, originated in the ancient civilizations of Egypt and Greece, persisted throughout the medieval, Renaissance, and Enlightenment periods, and lasted through the second Industrial Revolution. It flourished in Arabic and Indian Medicine. In terms of longevity, no other practice comes close. Germ theory, the basis for modern Western medicine, was formulated about 130 years ago. The modern practice of transfusion is about seventy-five years old. Bioodletting was faithfully and enthusiastically practiced for more than twenty-five hundred years (Starr, 1998, 17). There are several remnants from the long history with bloodletting. The honored British medical journal Lancet was named after the tool used to cause the bleeding. The striped poles outside barbershops were originally used as an advertisement for bloodletting: the red signifying blood, the white signifying the bandage. One of the signers of the Declaration of Independence, physician Benjamin Rush, was a strong advocate ofbleeding as a therapy. And the first American President, George Washington, died after copious bleedings administered for what appeared to be strep throat (Starr, 1998). Starr detailed a couple of developments and advancements that led to the abandonment ofbloodletting. There were a series of typhoid epidemics in Britain in the early 1830s. Typhoid made patients quite weak, and the bleeding would cause fainting. Doctors could not help but notice that their ministrations were not having 24

PAGE 38

positive effects on their patients. A significant advancement was the development of an area of inquiry called medical statistics, which called for measurement and record keeping as opposed to reliance on the impression of physicians. "Finally, the three giants of bacteriology-Louis Pasteur in France, Joseph Lister in Scotland, and Robert Koch in Germany-showed that microbes, not humors or other intangibles, cause disease. Germ theory became the basis of modem medicine" (Starr, 1998, 29). In the late 1 ih century, important breakthroughs in the understanding of the human body were achieved that helped to lead the way to the idea of transfusing blood from one living organism to another. One such breakthrough was discovery of the difference between arteries and veins and the recognition that the heart was a simple pump. This revolutionized scientific thinking about the body, and led to some early experimentation-mostly with transfusions between dogs. In the 1 ih and l81h centuries, however, blood was still believed to contain the essence of the creature from which it came, and hence thinkers at this time believed that characteristics or tendencies could be passed through the blood. To test this, experiments were designed in which blood was drained from one dog until it was nearly dead. Then the blood from a second dog was transfused to the first. Observers noted that the dog "virtually came back from the dead" (Starr, 1998, 9). Blood was transfused from several dogs into one dog, and from one species to another. Scientists pondered questions such as "Would a fierce dog become tame? 25

PAGE 39

'by being ... stocked with the blood of a cowardly dog .. ?" Would a trained dog forget how to fetch if transfused from an animal that did not know how? Would the recipient's fur color somehow change to that of the donor?" (Starr, 1998, 9, citing Boyle, Robert, 1666). Advances were also made with the instruments of transfusion. Reeds, an early favorite, were replaced by a series of silver cylinders with a sack between the cylinders. Attempts were also made at stitching the donating artery to the receiving vein (Starr, 1998, 1 0). While transfusion was hypothesized to be a potential cure for a wide variety of illnesses, physicians in the 17'h century were much against the use of human blood. 'It would be a very barbarous Operation, to prolong the life of some, by abridging that of others.' Animals on the other hand, did not seem to suffer unduly from giving blood, and farmers could provide a limitless supply. Beyond that, animal blood must surely be healthier than man's, which undoubtedly was debased by 'debauchery and irregularities in eating and drinking.' After all, 'sadness, Envy, Anger, Melancholy and Disquiet ... corrupt the whole substance of the blood' (Starr, 1998 II citing Boyle Robert, 1666). Antione Mauroy was the first modern recipient of the blood transfusion. Thought to be mad, he was found in I667 wandering the streets of Paris, naked (Starr, 1998 3). He was taken to Jean-Baptiste Denis, a physician to Louis XIV who had been experimenting with transfusing blood from animal to animal and had been toying with the idea of transfusing blood from animal to human. Denis 26

PAGE 40

transfused calf blood into Mauroy via a silver tube. Mauroy went back to work the next day. Two days later, he received a second transfusion; larger than the first. It was by luck that Mauroy survived the shock his body experienced as a result of this second transfusion (Starr, 1998, 6). Denis conducted several other animal-to-animal transfusions during 1666 and reported his results in Philosophical Transactions (July 22, 1667). He received a scathing response from a colleague the channel who had conducted some of the early transfusion experiments with dogs. This rivalry between the French and English physicians and an unfortunate lawsuit (described below) would put an end to transfusion for nearly 150 years. Once again Mauroy, the recipient of the first transfusion, and his wife, Perrine, showed up on Denis's doorstep requesting a third transfusion. Denis refused since Mauroy did not appear strong enough to withstand the transfusion. "Sometime later, Denis received a conciliatory letter from Perrine" asking him to ''exercise the charity" to come to her home (Starr, 1998, 14). When Denis arrived he found everything prepared for him to conduct a transfusion (including his assistant and tools) on a very sick Mauroy. Denis relented, but as he was about to begin, Mauroy began shaking, and the transfusion never occurred. The next 27

PAGE 41

morning Mauroy died.3 The news ofMauroy's death spread quickly throughout France, where detractors began distributing books and pamphlets accusing Denis of murder. Perrine attempted to blackmail Denis, and--in an attempt to clear his good name -Denis filed a libel suit against Perrine. Denis was found not guilty, and it was discovered that Perrine had been poisoning her husband with arsenic (Starr, 1998)! As an apparent afterthought to the ruling, the judge also ruled that all future transfusions would require permission from the Faculty of Medicine, a conservative and powerful group of physicians (Starr, 1998, 15). Since this group was not in favor of transfusions, this ruling effectively put an end to transfusion for over 150 years. Starr reports, transfusions were effectively terminated in Europe, when "[t]wo years later, the French parliament officially banned all transfusions involving human beings, with the English following suit. When two men died from transfusion in Rome, the pope banned the practice throughout most of Europe" (Starr, 1998, 15). In the intervening one and a half centuries without transfusion, significant advances would be made in the understanding of blood and in the art of medicine that would improve the success of transfusions when transfusions would reemerge 3 It appears that the blood transfusions might actually have been beneficial for Mauroy. An observer to the experiments concluded that Mauroy had syphilis. "Syphilis is caused by ... a bacterium that cannot tolerate high temperatures.. Dennis might have triggered in Mauroy a strange but feasible chain of events: He gave the transfusion; the patient reacted; the fever that nearly killed him impeded the bacteria. And for a couple of months, the madman was sane" (Starr, 1998, 16). 28

PAGE 42

as a medical procedure in the twentieth century. This hiatus in the practice of transfusions did not prevent theorizing about transfusions For instance, Erasmus Darwin (the father of Charles) published some speculations on the uses of transfusion in 1794 (Starr, 1998, 36). Then, in 1818, the first modem transfusion ofhuman blood occurred-and the patient died. The physician was James Blundel and he had been concerned about the high mortality rates of hemorrhaging mothers. His experimentation on animals would lead to two important decisions, "( 1) only human blood should be employed, and (2) transfusions should not be used to cure madness or change character, only to replace blood" (Starr, 1998, 37). Brundel 's subsequent experimentation with transfusions, from human to human, would renew interest in blood transfusion. New transfusion tools were invented and improved upon. "[B]y the second half of the nineteenth century, transfusion was becoming popular again, with hundreds reported throughout Europe" (Starr, 1998, 38). Transfusions were attempted during the American Civil War for leg amputations. "During Canada's great cholera plague, some doctors gave milk transfusions in the belief that the 'white corpuscles of milk were capable ofbeing transformed into red blood corpuscles"' (Starr, 1998, 38 citing, Jennings, 1888). These early transfusions were all done without an understanding of blood types, methods for the prevention of clotting, or of the importance of sanitation. 29

PAGE 43

In 1900, Viennese pathologist Karl Landsteiner differentiated the four blood types (A, B, AB, and 0) based on clever and methodical experiments in which he combined blood from various donors and noted the presence or absence of clotting. Landsteiner won the Nobel Prize for Medicine in 1930 for his discovery ofhuman blood groups (Starr, 1998, 39). However, his work did not affect the world of transfusion for some time. Direct transfusion, whereby the artery of the donor was sutured to the vein of the recipient, was the standard method for preventing clotting during transfusion (this worked since it avoided having the blood come in contact with air). Advances were made in this procedure. For instance, Dr. George Washington Crile, in Cleveland developed a metal ring "through which the recipient's vein could be drawn and cuffed making it easier to attach the donor's artery" (Starr, 1998, 41 ). This painful, messy method was abandoned when, in 1913, Dr. Edward Lindeman ofNew York's Belleveue Hospital developed the "multiple-syringe" that circumvented the need to cut open the patient's arm. "He slipped a sharp, hollow needle into the arms of patient and donor, puncturing the skin and entering the veins. The needles remained in place while he shuttled back and forth fiom donor to recipient, withdrawing and reinjecting blood with a syringe" (Starr, 1998, 43). These advances were quickly followed by others including rubber tubing between 30

PAGE 44

the syringes, valves to control the flow of blood and safe forms of anticoagulants.4 From here, transfusion became a standard part of medical practice. The concept of banking blood was the idea of an employee of the British Red Cross in 1921, Percy Lane Oliver (Starr, 1998, 53). Before this time, blood was being transfused on the spot through donations from friends and relatives. Oliver had the idea that a city bureau could be established that could have volunteers who had been both screened and typed, and ready to donate should Oliver give them a call. In the first full year of operation he had 13 calls for blood -just four years later there were nearly 800 calls (Starr, 1998, 54). The increased demand for blood can be largely attributed (according to Starr) to Dr. Geoffrey Keynes, brother of economist John Maynard Keynes. Dr. Keynes was a prominent British surgeon who had become interested in transfusions during World War I, "having learned the techniques from American physicians stationed in Europe" (Starr, 1998, 55). Dr. Keynes was responsible for introducing transfusion into standard British practice and for publishing the first modem textbook on the subject in 1941 (Keynes, 1941 ). The collaboration of Percy Lane Oliver and Dr. Geoffrey Keynes marked the beginning of modem blood banking. Oliver became an advocate for the donors who at times were receiving unfavorable treatment from physicians. Physicians assumed that donors were there for pay (they were not), and hence did not use the most 4 See Starr (1998) pages 47-49 for an in depth discussion of these advances 31

PAGE 45

humane of methods (for example, they surgically opened the vein before insertion of the needle). Oliver insisted on the humane use of a needle for the extraction of blood. He engaged in efforts to educate physicians about the voluntary status of the donors, since many physicians at this time assumed that donors were being paid. Oliver also "emphasized the personal nature of the donation, stressing the link between donor and recipient ... (Starr, 1998, 55). He reprinted thank you letters from donors and their families in his quarterly newsletter5 of the in order to illuminate the relationship between donor and recipient. Oliver's major contribution to blood donation practices was recognition of the humanity of the voluntary act. Largely because ofthe success of Oliver's and Keynes' efforts in London, similar arrangements were established around the world in the late 1920s and 1930s Interestingly, while much of the industrialized world (e.g., England, France, Russia, Germany, Japan, Austria, Belgium, Australia, etc.) was developing blood donor systems that relied on voluntary donations the United States was unique in its reliance on and acceptance of payment for donation. Payment for donation in the U.S. was consistent with the market model that dominated thinking at the time. In 1923, the New York Times reported that the standard payment was $35-$50, a healthy sum in those days (Starr, 1998, 58)! A New York Times article described "professional" blood donation as a potential supplement to a career field. During 5 The letters were printed in the Blood Transfusion Services Quarterly Circular of the British Red Cross 32

PAGE 46

the Great Depression, blood donation gained in popularity, and, according to Starr, blood donation bureaus had established themselves across New York City, collecting blood from indigents (Starr, 58, 1998). A group ofNew York City doctors was outraged that blood was being collected from populations likely to carry syphilis and other diseases. To address these concerns, they established the Blood Transfusion Betterment Association, whose goal was to raise the standards of professional donation (Starr, 1998, 58). The Association established a highly disciplined system of donation, requiring donors to have current physical examinations, syphilis tests and a telephone -so that donors could be reached when needed. The Association attempted to exclude donors that had a history of communicable diseases, or drug or alcohol abuse. They established a maximum frequency of donation (no more than one time in five weeks), staunchly advocated healthy living, encouraged exercise and cleanliness of their donors, and would not allow donations from donors that did not appear to be meeting their standards (Starr, 1998, 60). Starr concluded that both the voluntary system in London and the remunerated system in New York City were producing favorable results in this era (late 1930s ). Both systems were acquiring and distributing safe blood for patients in need. However, since the United State's system charged money for the blood that it provided ($35 plus $6 commission for the Association), funds were available for a more professional approach (e.g., with a medical director). 33

PAGE 47

During World War II, there was a rapid growth of the blood industry and the development ofblood banks (where the donated blood was actually stored). Starr reported that despite Britain's extensive preparations for World War II, it was "surprisingly complacent about blood. London still relied on 'donors-on-the-hoof, a system that, barely adequate in peacetime, would be swamped in the first days of any conflict" (Starr, 1998, 84). The United States' first efforts at large scale "bleeding" came in 1940 with the shipment of plasma to allied troops in Britain (Starr, 1998, 93). Plasma was sent instead of whole blood because of its longer shelflife,6 and because plasma, unlike whole blood, could be used "regardless of the blood type of either donor or recipient (Dulles, 1950, 414). Dried plasma had the additional advantage of not requiring refrigeration. Methods had been developed to tum the plasma into a dry powder, that, when mixed with distilled water, could be administered anywhere along the battle front (Dulles, 1950, 414) By 1941, both dried plasma and albumin were being sent to the front lines. These two products required more blood donations per unit of transfusable blood and so the demand for blood rose yet again; the demand, of course, was assisted by the increasing casualty rate. The Red Cross began setting up more donation centers and began an extensive advertising campaign (e.g., "He gave his blood. Will you give yours?") (Starr, 1998). 6 Fresh plasma has a 28 day shelf life 34

PAGE 48

A serious social problem was plaguing U.S. blood donation at this point-the controversy over "colored" versus "white" blood. The Red Cross, in accordance with military desires, was turning away African-American donors. After Pearl Harbor, the Red Cross was able to convince the military to accept African-American blood so long as it was labeled as such. The justification was that "[t]his way those receiving transfusions may be given blood from those of their own race" (Starr, 1998, 108). Understandably, this policy was considered controversial, and by many, offensive. In 1942, the Red Cross held a meeting to reconsider its position and concluded that there was no difference in the blood of the races, yet, "found it impossible to overcome the assumption that 'most men ofthe white race objected to blood of Negroes injected in their veins"' (Starr, 1998, 109 citing Fletcher, History of the American National Red Cross).7 The United States population was not alone in its racism. The Germans were hampered by their unwillingness to use anything other than "Aryan blood," and the Japanese were still using arm-to-arm transfusion, as Starr implies, for blood purity reasons ( 1998, 115). Despite these social problems, the war effort led to great advances in blood banking. By the middle of the war, the United States had elaborate systems for collecting blood, shipping it to companies able to convert the blood into dried 7 The year of publication for Fletcher was not listed by Starr. Despite numerous attempts the original document has not been identified. 35

PAGE 49

plasma, packaging and shipping it off to the war effort. Likewise, albumin was being shipped in a lighter package and an easier to use liquid form. There was extensive cooperation between universities and industry. Up until this point, the United States still did not have any government intervention (at any level of government) in the market for blood. Around this same time, serious scientific research was searching for replacements for blood. 8 In 1942/1943, plasma and albumin were being routinely used. However, some military surgeons were becoming convinced that whole bloodwas needed for the American troops. When their requests were repeatedly denied, they endeavored to establish their own system for providing whole blood to troops in North Africa. Starr speculates that the denials were likely because of the logistical difficulties associated with transporting the whole blood and because the current Army Surgeon General, Norman T. Kirk, "lacked the patience for laboratory science. As far as he was concerned, the question of transfusion had been answered with plasma, and was loath to reopen it" (Starr, 1998, 127). Dr. Edward Churchill, a consultant to the Army regarding the troops in North Africa, believed that part of the resistance from official channels was also a result from the U.S. Army's investment in blood substitutes. Churchill wrote: A huge vested interest had been built up starting from assumptions and erroneous thinking ... Civilians were busy helping with the war effort and many had their prestige at stake in their collection and use of plasma. Publicity had been launched to provide plasma for the wounded soldiers, and 8 For a discussion of earlier, less scientific endeavors, see Starr, 1998 36

PAGE 50

the Red Cross as well as the N.R.C. [National Research Council] was behind it. Edwin Cohn9 was working to improve plasma and was trying to get albumin solution into production ... It gained size and momentum like a rolling snowball (Starr, 1998, 127 citing Churchill 1972). Despite these official impediments, Churchill and his colleagues set up their own system in North Africa. Donors received $10 per pint, and the blood was not tested for diseases as was traditionally done. Supplies were harvested from whatever sources were available. This work was all done outside the normal chain of command. Ad hoc blood banks were established near battles in Sicily in the summer of 1943. Churchill repeatedly attempted to receive assistance and approval from Washington, which was not forthcoming. He became desperate and encouraged the New York Times to print an article that argued that plasma was not medically sufficient and that blood banks that provided whole blood (as opposed to plasma) were saving lives The article was entitled, "'Live-blood' Banks Save Soldiers' Lives in Sicily When Plasma Proves Inadequate" (Starr, 1998, 128, referring to a 1943 New York Times article). Even more provocatively, ordered supplies through official channels that had not been approved. He wrote memos on official letterhead requesting supplies; these letters indicated that the U.S. Surgeon General was in support of this position when, in actuality, he was staunchly opposed (Starr, 1998 129). 9 Dr. Cohn was a prominent physician best known for his groundbreaking work in the fractionation of plasma 37

PAGE 51

By 1943, over 2.5 million packages of dried plasma had been shipped to the combat zone, not to mention the "nearly 125,000 ampoules of albumin" (Starr, 1998, 121 ). As the war progressed, methods for collecting and shipping blood had become more sophisticated. However, some Army and medical personnel were concerned that plasma was not the right answer and pushed for finding ways to get whole blood to the front was investigated and to some degree was mastered. A highly developed system for blood was functioning by the end of the war. The U.S. had been using blood in more forms than any other country, establishing the foundations for a large-scale blood complex. Immediately after the war, the blood industry began to shut down. Donations declined as the sense of commitment and community declined after the war. The American Red Cross (ARC) was $16 million dollars in debt (Starr, 1998, 165) from financing its national blood effort and so closed its blood collection sites. These closures did not last long as the board of the ARC realized in 1947 that its organization had gained celebrity and a reputation synonymous with patriotism for efforts with blood during the war. As a means of re-establishing financial viability, the ARC worked to create a national program. Its strategy was to establish blood banks across the country, but initially only in cities that did not already have established blood banks. The American Association of Blood Banks (AABB) was founded soon after as a reaction to what was considered by some to be predatory behavior by the 38

PAGE 52

American Red Cross. Existing blood banks had become concerned that the ARC would push them out of business and would gain a monopoly on blood donation and distribution. Starr (1998, 174) observes "[t]heir ostensible purpose was to trade expertise and technical information, but the political agenda quickly became clear. Their real raison d'etre was to oppose the Red Cross movement ... Some denounced the Red Cross movement as socialistic" (Starr, 1998, 174). At the first meeting of the AABB, the members passed a resolution inviting the Red Cross to join. However, the wording of the resolution was intended to indicate what the AABB board saw as the proper function of the Red Cross -an "aid in the procurement of donors." The implication was that all else should be left to the professionals (Starr, 1998, 174). This conflict between the Red Cross and the AABB was, basically, about power. But it was also about differing conceptions ofblood and how blood should be managed (Starr, 1998). The AABB was premised on the belief that blood collection should be managed and controlled by physicians. In contrast, the Red Cross was intending to mobilize its considerable volunteer brigade to fill this role. Second, the AABB saw blood as a commodity that should be left to the free market, whereas, the Red Cross saw blood as being a "community responsibility." The Red Cross stressed the civic and voluntary nature of the donation. Third, the AABB believed in local autonomy and vehemently resisted the idea of central control over blood The Red Cross preferred a national organization that set standards and 39

PAGE 53

established control at the national level, aided by local chapters. The AABB stressed technical education; the Red Cross did not. Finally, the AABB advocated a policy of"individual responsibility" in which recipients ofblood were responsible for either paying for or replenishing the blood they used. The hope was that recipients would feel an obligation to round up sufficient donors on their own, so as to avoid the monetary fees. The Red Cross held that blood was a community resource, available to those in need. Demands were not placed on the recipient for replacing the blood they used. As Starr observed, (1998, 175), there were some points of agreement. "Both organizations shared some fundamental principles: that blood should not be considered a commodity, for example; that the processing fees be kept as low as possible; and that the enterprise should run on a nonprofit basis" (Starr, 1998, 175). In the transition to a peace-time blood complex, a system was revealed that was highly unorganized, unregulated, and contained large gaps and inconsistencies. As Starr explains: Unlike most Western European countries, which had enacted blood legislation following the war, the Americans never established a national policy. They left blood collection to the marketplace, with no central coordination, inventory, or control. The American Red Cross and American Association of Blood Banks each attempted to monopolize the market, yet neither was able to prevail. Their inability to cooperate divided the nation into a patchwork of territories. Each system had its own set of rules free blood in the Red Cross areas and "loaned" blood in territories of the AABB. Neither respected the rules of the other (Starr, 1998, 188). 40

PAGE 54

These gaps and inconsistencies had left room for entrepreneurs who saw blood as a commodity and recognized blood banking as a growth industry. Individuals with no medical background had set up and run blood banks across the country. They paid for donations and were under the jurisdiction of no medical or regulatory authority. On a national level, these for-profit banks had "never constituted more than a fifth ofthenation's collection capacity, but in some areas they became particularly strong" (Starr, 998, 189). These donors-for-cash were often down on their luck, and had much greater risks of carrying diseases like malaria and hepatitis. Moreover, some of these for-profit banks had become notorious for their unsanitary conditions and for the poor quality of their blood. One particularly bad case was in Kansas City. When local hospitals had boycotted the for-profit bank and established their own blood bank, the Federal Trade Commission stepped in. The inspector for the FTC had seen the situation only in economic terms; public health and safety were not part of his considerations. The FTC subsequently "charged [the doctors] with illegally conspiring to interfere with free trade" (Starr, 1998, 192). In the trial that followed, the government's case was that the physicians and hospitals were attempting to squeeze out a competitor. The physicians' case was that they had acted out of a sense of public responsibility. Again, we see the differing metaphors about blood. The FTC saw blood as a commodity and hence their argument was that blood should be "bought, sold, and processed like other drugs" (Starr, 1998, 193). This was a very troubling metaphor 41

PAGE 55

for the physicians and had some devastating implications. If blood were to be considered like any other drug, then blood, and those who are involved with blood, would be subject to other product related statutes. "The most threatening of them," Starr noted, "was the Uniform Product Code, a federal regulation adopted by all the states mandating that anything sold as an article of commerce carried an implied warranty ... If the consumer is banned by the product ... he can sue the manufacturer for violating the implied warranty" (Starr, 1998, 193). This understandably alarmed physicians, since there was no way to be sure if, for instance, blood contained the hepatitis virus. The physicians contended that blood was not a commodity, but a human organ; likewise, they defined transfusion not as an economic transaction but as a medical procedure (or alternatively as a service). This was to no avail. The verdict from the FTC was that the for-profit blood bank was "a legal fully licensed business entitled to the complete protection of the law" (Starr, 1998, 195). In response to the outcry resulting from this ruling, legislation was introduced in Congress in 1964 to exempt blood banks from federal antitrust lawsuits. The hearings were contentious and, in the end, the bill never made it out of committee. In addition, an appeal to a five-panel FTC commission upheld the original ruling. This was all occurring amidst scandal in the blood blanking industry Blood banks were being accused of tampering with expiration dates; hepatitis was being found in the blood; the common payment for donation was 42

PAGE 56

causing concern over safety of the blood supply; the squabbling between the AABB and the Red Cross had not gone unnoticed; and, the reputation of the blood banks was turning from heroic to scurrilous (Starr, 1998). Finally, in 1969, a Federal Appeals Court ruled that the FTC had overstepped its authority and that the blood banks were outside of its jurisdiction.10 Starr pointed out that although the 1969 ruling allowed the physicians to form their own blood banks, it did not resolve the issue of the status ofblood. Was it a commodity, like any drug, or did it have some special status? At this same time, European nations had already turned to voluntary donation and had organized national blood donation systems. After a scandal in which a U.S. Ambassador to Japan was stabbed and then given tainted blood (he developed hepatitis as a result), Japan also made the change to a nationalized system, with voluntary donation. The United States, however, remained unique in its reliance on paid donation and its lack of a central authority for blood banking. In the mid-1960s only seven states were licensing blood banks and only five states were inspecting them (Starr, 1998, 204). There was still no national policy. "The blood business boomed in the 1960s and '70s" (Starr, 1998, 207). The crazy-quilt fashion in which the U.S. blood industry was established meant that no one knew precisely how much blood was collected. Starr reported that most 10 By 1969, however, the Kansas blood bank at the center of the controversy was no longer in business 43

PAGE 57

estimates were well above six million pints a year in the U.S. The business was expanding, and also bifurcating into two separate systems: one for whole blood and another plasma. The development of plasmapheresis was the catalyst for this split. This procedure allowed the removal and then centrifuging of the patients' blood. The red blood cells were then re-infused back into the donor. The process was painful and time consuming but had the advantages of not causing anemia, and a shorter tum around time before the donor could donate again. 11 Donors could now donate 104 times per year, instead of six (Starr, 1998, 208). The plasma industry exploded, and with the explosion came corruption and unethical practices. Donor centers were opened in the high-risk parts of downtown areas, attracting those that were largely unfit for donation; e.g., donors were paid with vouchers redeemable at liquor stores (Starr, 1998). A particularly unethical practice was the gathering of gamma globulin from prisoners. Gamma globulin could be derived from plasma, but it was far more efficient to "find someone who had been exposed to a disease and had produced a high concentration of the antibodies in question" (Starr, 1998, 21 0). This could be accomplished by searching the population for individuals exposed to the "right" diseases. Or, you could expose people wait for their bodies to build up sufficient antigens, and then pay them to extract their plasma. According to Starr (1998), the latter was regularly done with 11 Plasma can be replenished in a few days as opposed to the weeks it takes the body to replenish red blood cells. 44

PAGE 58

the prison population. The finns running the plasma drives in prisons were exempt from federal safety inspections, since they only collected the plasma and did not process it. Not surprisingly, disease rates skyrocketed at the prisons where these sorts of practices were going on (Starr 1998, 211 ). Despite the well-known unsanitary conditions, according to Starr, the major phannaceutical companies remained loyal clients. Researchers had suspected that blood from "professional" donors was more likely to be tainted by disease than blood by voluntary donors. However, the difference in disease rates had never been detennined. In 1966, Dr. Garrott Allen published a study showing that professional blood was ten times more likely to be infected than was voluntarily donated blood (Starr, 1998, 220). Allen recommended the use of single donors (rather than pooling blood from multiple donors), rapidly moving away from professional donors, and a reduction in the amount of blood given for each transfusion. He also recommended the labeling of blood that came from high-risk populations (namely skid row and prisons). Allen's recommendations did not sit well with certain members of the blood banking establishment. In fact, a few months after he published his study, the California and Los Angeles Medical Associations denounced his suggestions as "impractical, unworkable, and cause for concern". (Starr, 1998, 220) The media soon became involved, "asserting that the blood-and-plasma industry was engaging in a game of 'transfusion roulette' with products that might 45

PAGE 59

transmit hepatitis" (Starr, 1998, 221, citing Altman, L., 1970). Allen began a massive letter-writing campaign, with one of the recipients being Elliot Richardson, thenSecretary of the U.S. Department ofHealth, Education and Welfare. Richardson presided over the Division of Biologics Standards, the lax regulator of the blood banking establishment (Starr, 1998, 225). In their correspondence, Allen presented statistics and exhorted Richardson to require the labeling ofblood as being from either paid or volunteer donors. In 1971, Allen sent Richardson a copy of a new book on the topic of blood donation, Richard Titmuss's The Gift Relationship: From Human Blood to Social Policy, which compared the blood donation systems between the United States and Britain. The final chapter ofTitmuss's book contains an often-cited tirade against payment for donation In it, he first attacked the commercialization of blood first for ethical reasons, arguing that the commercialization ofblood donation was odious because it "represses the expression of altruism, erodes the sense of community ... [and] places immense social costs on those least able to bear them" Titmuss, 1970 314). Titmuss went on to critique the commercialization ofblood on four other criteria: economic efficiency, administrative efficiency, price and quality He concluded that on these "four testable non-ethical criteria, the commercialized blood market is bad" (Titmuss, 1970, 314 ). He contended that commercialized blood systems are wasteful, administratively inefficient, costly and more likely to produce contaminated blood. 46

PAGE 60

Critics have contended that Titmuss' work was biased, comparing the worst ofthe U.S. to the best ofthe U.K., nonetheless, The Gift Relationship hit a responsive chord with the American public-and with Secretary of DHEW Richardson. Secretary Richardson established a task force to "look at new ways of managing the American blood supply." A few months later, President Nixon, declaring blood "a unique national resource," ordered DHEW to conduct an intensive study of better ways to manage it (Starr, 1998, 228). Several blood reform bills were also introduced into Congress. Several months later, Richardson gave the FDA regulatory oversight of the blood banks. The FDA had much more authority than the Division ofBiologics Standards, this meant that the blood banks would now be regulated and inspected. As Starr and Titmuss pointed out, commercialization ofthe U.S. blood banking industry had failed. The for-profit system was wasteful (29 percent of blood was going bad on the shelves) and was producing unsafe blood. A more centralized, bureaucratized system had some appeal, but was anathema to the free market ethos of the time. So, despite the serious failures of the blood banking system, it was not reorganized. Instead, "in 1974, the HEW proposed a National Blood Policy, an attempt to reform and unify the blood system without resorting to federal control" (Starr, 1998, 251 ). The industry was called upon to develop a plan. The blood bankers "founded the American Blood Commission, a nonprofit 47

PAGE 61

voluntary body, to serve as a forum for all blood-related debates, and to issue decisions that members would voluntarily follow" (Starr, 1998, 251 ). The continuing acrimony between the AABB and the Red Cross "balkanized" the proceedings. Their perennial, territorial disputes were augmented by fights over the use of the non-replacement fee -an issue that the organizations were using to characterize what they saw as differing philosophies on blood. The AABB and Red Cross both charged mo11:etary fees for the blood. The AABB would waive the fee if the blood used by the patient was replaced by blood donated in their name. The American Blood Commission sponsored a study of the replacement fees. The report, with little factual support and much interjected politics (Starr, 1998, 253), concluded that replacement fees were "tantamount to selling blood" (Starr, 1998, 253 citing American Blood Commission, 1977, Recommendation for Unified Donor Recruitment, July 26, 1977). Of course, the AABB members were outraged; the President, Bernice Hemphill issued a minority report, arguing "either doctrine could succeed, depending on the experience ofblood banks in their region" (Starr, 1998, 253, citing National Blood Policy papers, MS C 393) The Red Cross withdrew from the AABB, after 16 years of membership, partially as a result of continuing conflict regarding competition for donors and the controversy over replacement fees (Starr, 1998, 254). It left with a significant financial debt to the AABB, since the Red Cross had withdrawn more blood from the clearinghouse than it had replaced or paid for. Moreover, its departure disrupted 48

PAGE 62

the replacement credits many Americans had built up. Many donors had built up credits donating at Red Cross banks, assuming that this would waive their replacement fees in the eventuality that they needed blood. When the Red Cross withdrew, all AABB credits accrued at the Red Cross were eliminated. The public was understandably angry. A series ofheated letters, accusations and lawsuits followed. The Red Cross, in tum, accused the AABB of profiteering. The American Blood Commission voted to phase out the non-replacement fee but the Commission had no enforcement authority. The Commission disintegrated in the late 1970s, acrimony remained the rule and little changed. On the safety front, the world was already witnessing the explosion of communicable diseases amongst the hemophiliac population as a result of the large amounts ofblood required to treat their disease. A large portion of the world's plasma was now being imported from the poorest comers of Latin America and was then purchased by large U.S. based companies who cornered the international market in plasma (Starr, 1998, 241 ). European nations and Japan were almost entirely dependent on the United States for their plasma. Payment for whole blood donations in the United States was also still common, and a cause for concern. Then, in 1978, the FDA issued a regulation that effectively put an end to paid donation for whole blood. The regulation was a compromise position between the status quo and an outright prohibition against paid donation. Blood banks were now required to label each unit of blood as coming from either paid or voluntary 49

PAGE 63

donors. Regulators had been loath to outlaw paid donation, since plasma banks relied almost exclusively on remunerated donors. Paid blood disappeared since no hospital was willing to buy the implicitly more risky blood. Hepatitis rates declined rapidly, at least for type B. Whole blood donated in the U.S. was not being shipped overseas because of the shorter shelflife. However, it was being shipped around the country-and much profiteering was the result. A Pulitzer Prize-winning expose on the blood banking industry by Gilbert Gaul ( 1989) found that some non-profit blood banks were staging "blood crisis campaigns" and then selling the excess blood to areas in need at substantial mark-up. Indicative of American capitalism, blood was becoming big business. By the early 1980s, the HIV/AIDS epidemic had begun. Early researchers had made the connection between gay men and infected blood. However, representatives for the gay community did not want gay men to be labeled unacceptable donors. They argued for the screening of"blood, not people." The drug companies did not agree.12 There was no test for AIDS, but because ofthe high degree of correlation between AIDS and Hepatitis B, screening for hepatitis would catch most of the AIDS. The major blood banking organizations (AABB, ARC and others) issued a document "insisting that the case for blood borne transmission was inconclusive, and offering several 'reasonable' measures for blood 50

PAGE 64

banks and physicians to follow. These included educating donors about AIDS, allowing 'autologous donations' in which patients could set aside blood for their own future use, and discouraging donations among groups 'that may have a high incidence of AIDS"' (Starr, 1998, 275). They did not advocate additional testing of the blood, but instead recommended asking the donors questions about "risky" behavior. At this point, the medical establishment was confident that AIDS was transmitted via blood, that it was highly contagious and that there was a way to screen for it (the hepatitis test, which would cost about $5 per test). Nonetheless, blood was not tested-this was partially due to public resistance to confront the disease. Shilts ( 1987) argues that this reluctance stemmed from the public discomfort with discussing the risk factors for AIDS and because, at least in the beginning, it was marginalized populations that were mostly likely infected (homosexuals and intravenous drug users). The donation rates dropped, as people feared that they could become infected by donating. In 1983, the U.S. Public Health Service issued its first AIDS-related recommendations, including a call for "sexually active homosexual or bisexual men with multiple partners to refrain from donating plasma or blood" (Starr, 1998, 277). The FDA then issued guidelines for the blood industry recommending self-exclusion. That is, blood banks would now 12 For a very thorough analysis of these issues, see, And the Band Played on: Politics People and the AIDS Epedmic, by Randy Shilts, 1987. 51

PAGE 65

be required to ask more questions and have people sign a form indicting they understood the risk of AIDS;13 however, no additional testing would be conducted. There remained an enormous backlog of contaminated blood products. In 1984, researchers discovered that heating plasma proved effective in reducing infection (this did not work with whole blood). The FDA permitted the phasing in of new "heated" blood products (like factor VIII for hemophiliacs) and did not require the issuance of a recall. Also in 1984, Dr. Robert Gallo announced that he had identified the virus that caused AIDS, identifying AIDS in the blood would be much easier. By 1985, an inexpensive test was developed (ELISA). There were two concerns with the test: (1) the false positive rate, (2) and it could only detect AIDS once the donor had become symptomatic. There was also growing public fear about AIDS, as the death and disease tolls began to rise. "In 1987, a poll conducted by the American Association of Blood Banks found that 27 percent ofthe people they surveyed thought they could get AIDS merely by giving blood" (Starr, 1998, 317, emphasis in original). Government regulation and oversight of the blood banks were never extensive to begin with and under the Reagan Administration became more lax. A blood bank could expect to be visited by the FDA about once every two years (Starr, 1998, 317). As a result of the AIDS crisis, the FDA stepped up inspections and 13 The idea behind self-selection is that at-risk individuals would identify themselves as being high risk, and hence, would decide not to donate blood. 52

PAGE 66

found gross violations-especially in the Red Cross. "In almost every region, inspectors found that Red Cross blood banks had released blood that had failed the ELISA test or neglected to notify the recipients of bad blood" (Starr, 1998, 317). Blood was being purposely mislabeled (reactive blood being labeled as non reactive) and then shipped offto pharmaceutical companies. In 1988, the FDA secured an agreement with the Red Cross, in which it promised to correct its deficiencies. This did not occur but was eventually remedied by a court order and a tight to which the Red Cross was required to adhere (Starr, 1998). In 1993, the Subcommittee on Oversight and Investigations ofthe Committee on Energy and Commerce of the House of Representatives held hearings on the safety of the United States blood supply. Preceding these hearings, there had been growing perceptions and documentation regarding the safety of the blood supply. Not surprisingly, these hearings occurred in the early 1990s, when information about AIDS and the spread of other communicable diseases (namely hepatitis) via blood transfusion were becoming more widely recognized. Also, concerns over the persistent blood supply deficit were sparking additional interest in the issue. This subcommittee had met previously to discuss issues of blood safety and supply. At the committee hearing, policies-recommended and implemented based on previous committee hearing recommendations -were re-evaluated. Also, the entire issue of the safety of the blood supply was revisited. 53

PAGE 67

There was a great deal of acrimony in this subcommittee hearing, stemming from fundamental differences over the conceptualization of blood as well as differences in opinion on acceptable levels of risk. Furthennore, there were disagreements over the proper role of the FDA. FDA Commissioner David Kessler summarized the regulatory community's position in the following manner: FDA's belief that a voluntary agreement with the American Red Cross would be sufficient was, I think, emblematic of our collegial approach to regulated industry at the time Those days are behind us. We have shifted, from relying on voluntary agreements to the use of agreements involving court supervision and sanctions where necessary, from concentrating on jaw boning with industry to writing enforceable regulations (Kessler, 1993, 22). The FDA was rather severely criticized by Congress for the perceived shortcomings in enforcement of blood safety regulations The lack of success of the FDA can be viewed as a problem of poor regulatory performance or, as Representative John Dingell (DMl) presented it, it could be a case of insufficient authority. If the problem was the level of authority, then the solution was to give the FDA broader power over blood donation. The argument was that the enforcement problems were not caused by anything the FDA had or had not done; critics claimed that it was the "partnership" type approach that was the culprit. Thus, a stricter, more regulatory approach was necessary. They argued that the problem was that the FDA has treated the blood community differently than they had other industries. If you accept this metaphor, then there was no reason that a "partnership" was necessary--the clear solution to improving the safety of the blood 54

PAGE 68

supply was to strictly regulate the industry, as other similar industries were regulated. The FDA's conceptualization ofblood as a commodity rather a human product underpinned this important distinction about whether and how the industry should be regulated. Increased "official" concern over the blood quality and quantity was evidenced from the numerous hearings and meetings on the topic in the 1990s. For instance in 1990, 1991, and 1993, there was a series of Congressional hearings on blood safety, focused on the FDA's role in regulating the blood collection and distribution process. Then, in 1996, the National Academy of Sciences' Institute of Medicine released a series of monographs from the Forum on Blood Safety and Availability. As a result of these hearings and especially the AIDS crisis, the current blood supply is safer.14 The FDA policies have been effective, and the tests are becoming more accurate. Invasive questioning about risky behavior is required as a screen before donation. The basic structure of the blood banking industry has not changed, but there is more regulatory control now. The FDA and its Center for Biologics Evaluation and Research, is now responsible for the establishment of standards (this is done in collaboration with the Public Health Service) and for the regulation of blood products and blood banks The FDA has set more stringent 1 4 In fact, there has only been one identified case of AIDS resulting from blood donation in the last three years (Associated Press, 2002). 55

PAGE 69

guidelines with the help of its Blood Products Advisory Committee "stipulating that the committee must only deliberate over safety and efficacy-not over cost effectiveness, as in the past" (Starr, 1998, 349). The committee must also include consumers. The National Heart, Lung and Blood Institute of the National Institutes ofHealth's Division of Blood Diseases and Resources supports research related to blood transfusion issues, though they have no regulatory authority. Still, much remains the same, and new risks are apparent. The unfolding saga with CJD (aka "mad cow" disease) has given the blood industry another opportunity to be bold in its protection of public safety. The blood industry has been intrepid, severely restricting donations from individuals who have spent time in Europe. These new policies are particularly challenging for some metropolitan areas, such as New York, and for the armed forces. The armed services have adopted even more stringent restrictions than those recommended by the FDA. Given the mobile nature of their populations, these changes will have dramatic impacts. Another important trend has been an increase in the commercialization of non-profit blood centers. Blood centers sell voluntarily collected blood to hospitals, pharmaceutical companies and other customers. In the past, at least some blood centers saw themselves as performing a community service; and since the "environment" did not force it, the organizations' focus was not on sales. However, drastic changes in health care have made even the market for blood more 56

PAGE 70

competitive and have put blood centers under increasing financial pressure. Regulators and purchasers ofblood are putting great scrutiny on blood centers. Once-stable relations between blood centers and hospitals are being disrupted as hospitals seek to buy blood from the lowest bidder (an understandable position, to be sure, but something relatively new to blood banking/hospital relationships). Trends are now toward more competition between blood centers in a given locale, more adversarial relationships between the main organizational players, and consolidation ofmarkets.15 Continuing acrimony between the Red Cross and the AABB manifests itself in competition for both donors and purchasers of blood, and, battles over "lucrative" markets (Frantz, 1996; Clark, 1998). On Themes and Metaphors One of the recurrent themes through this history ofblood policy is the disagreement over the nature of blood. There are two classic views previously discussed (one more overtly than the other) in this historical review One view is that blood is a human product, given altruistically. The competing view is that blood is a manufactured product. The difference is that although the "raw material" (blood) is donated by humans, it is in every other way a manufactured good. The argument is that the process of preparing the blood products for transfusion means that it is no different from other pharmaceutical products, and hence should be 15 San Diego Blood Bank (1999) Personal communication. 57

PAGE 71

regulated like any other pharmaceutical. These competing views of blood greatly influence how the stakeholders in the blood industry believe that blood should be regulated, and the proper role for government in relation to securing a safe supply of blood. Below, we explore the importance of these divergent blood metaphors. FDA Commissioner Kessler was a vocal supporter for the blood as a manufactured commodity metaphor. In fact, he specifically encourages its adoption. His contention is that the adoption of this metaphor would force the recognition of the importance of regulating blood in a fashion similar to any other pharmaceutical or manufactured products. We are dealing with a product, the product that is going to get transfused (Kessler, 1993, 35). Like any regulated industry blood banks are responsible for ensuring the safety of their products (Kessler, 1993, 20). The product needs the same kind of rigorous quality control and adherence to good manufacturing practices that are in place for any of our regulated industries that manufacture products (Kessler, 1993, 35). Peter Tomasulo, M.D. (1995), in his discussion oftransfusion alternatives, mentions some of the difficulties associated with considering blood to be a manufactured product that can be regulated by the FDA just as any other drug. First of all, "drug manufacturers try to make absolutely pure compounds with a small number of components that have well understood effects. Compared to most drugs, blood is complicated and nonspecific. It has too many components and too many 58

PAGE 72

effects (desirable as well as potentially dangerous)" (Tomasulo, 1995, 3). He also points out that while there are substances in blood that are non-beneficial to the patient, these cannot be removed (given the current state of technology) as they often can be with commercially synthesized drugs. In addition, the processes for ensuring the quality of source material are very different. Tomasulo points out that "it is difficult to imagine a drug manufacturer would ever depend on interviews of thousands of volunteers over whom they have no control for the quality of their raw materials" (I 995, 93). He claims that blood products, because of their nature, cannot undergo the same end-product testing that other manufactured drugs can (1995, 93). Finally, Tomasulo points out that drug manufacturers are required to list all of the components of their drugs. With blood, this is particularly difficult to do, because of its many components and because knowing what is in the blood relies on individuals reporting information that they may not know (e.g. that they have been exposed to a particular communicable disease) or may be reluctant to report (e.g. that they have engaged in a risky behavior). Moreover, it might simply be the case that the relevant information was not requested for instance, "have you been to England in the last 2 years." The points that Tomasulo addresses is that blood is made by a human, is very complex, and the blood industry does not have a large degree of control over "the product." It is difficult for those responsible for collecting and distributing blood to meet the exact requirements of other drug manufacturers. 59

PAGE 73

Lori Andrews and Dorothy Nelkin, in Body Bazaar: The Market for Human Tissue in the Biotechnology Age, provide a third view ofblood (or in fact all body parts). They document the ways in which "demands for skin, blood, placenta, gametes, biopsies tissue and sources of genetic material are expanding" (200 1, 2) and discuss the many ways in which these samples are being taken, sometimes without consent, and used to produce lucrative biological products.16 Often the "donors" received no benefit from their biological samples, and may not even be aware of the uses to which their body samples have been put. Hence, "people's tissues, cells and genes are increasingly being perceived as natural resources to be harvested and transformed into value-added commodities" (Weiss, 2001). The blood as natural resource metaphor is a new one, and one likely to grow as the profits to be reaped from the genetic information contained in blood continues to grow. These competing metaphors for blood are, of course, nothing new. Titmuss remarks on them (except for the "blood as natural resource" metaphor) in his book The Gift Relationship. What is important is that the debate about which metaphor is most appropriate for framing the discussion is still a contentious one. The regulatory community seems to be currently holding sway and the blood as product metaphor seems to be quite strong. In The Gift Relationship, Titmuss concludes 16 This is happening both in the United States and abroad. Vivid international examples include the Chinese system where organs can be purchased from donors that may not have consented, and the Singaporian system where donor consent is presumed. 60

PAGE 74

that the commercial capitalistic nature of the U.S blood donation system jeopardized the safety of the blood supply Specifically, he notes that rates of infection were higher in blood donated by paid donors than by voluntary donors. He also noted safety differences by forand not-for-profit blood banks. His conclusion was that the United States needed fundamentally to rethink how it was to conceptualize blood donations (Titmuss, 1970). While critics have contended that Titmuss overstated his case, 17 many of Titmuss's recommendations have been embraced. Since the publication of his book, there have been fewer paid donations, and more not-for-profit blood banks. However, now, over 30 years later, the American blood community finds itself going back to the very metaphor and mindset that had been considered the bane of the blood banking industry in the past. Granted, there has not been much serious talk of paying donors (there is always some talk of this), but it is clear that some significant reconceptualization of blood as an industry is occurring. The irony is that both the shift away from and now back towards this commercial metaphor has been justified as an attempt to improve the safety of the United States blood supply. In summary this background on blood donation has thus far reviewed the nature ofblood, the history of blood policy and some competing metaphors for 17 Sapolsky and Finkelstein ( 1979), for instance critique Titmuss for overstating the deleterious impacts of capitalism 61

PAGE 75

blood. Missing from this analysis is a description of blood donors, and a consideration of previous research on why people donate blood. These two issues will be addressed, in tum, below. Blood Donor Motivations There has been a modicum of research into blood donation issues, including a large number of published surveys on the characteristics of blood donors. These reviews have considered a wide range of factors that can be roughly clustered under the categories of socio-demographics, personality and motivations. A brief review of the literature is presented, describing blood donors and why individuals donate blood, since this helps set the stage for the proposed methodology and study plan, which will then follow. In 1977, Robert D_swalt published a review of the literatUre on blood donation (much of the literature was his), identifying motivations for and against donation. He concluded that the results of previous studies were remarkably similar for the last 20 years and that nothing new could be added (meaning since the 1950s). He summarized the major motivations for donation as altruism, personal or family 62

PAGE 76

credit, 18 social pressure, and reward.19 He noted that reasons for not donating include fear, medical concerns, apathy, and inconvenience (Oswalt, 1977, 123). While Oswalt's work focused mostly on differentiating between the motives for donors and non-donors, he recommended that blood centers focus on the "retention and management" of existing donors rather than on the recruitment of new donors. However, much of the work reviewed by Oswalt may now be obsolete since it came from an era of payment for blood donation. Ten years later, Jane Piliavin wrote a review that covered relevant literature published after Oswalt's review (note, much of the work was hers). Piliavin focused her literature review on how first-time donors became regular or habitual donors (Piliavin, 1987).20 Her explanation was that the emergence of AIDS had encouraged blood centers to transition their efforts from recruiting new donors to maintaining previous donors who had "safe" blood. While her work provided many useful insights, reconsideration of the blood donor literature is warranted, once again, due to environmental circumstances. 18 Family credit refers to the system in which donors could repay use of blood with either money or blood donations made in their name. 19 It is interesting to note that in an era of payment for donation, Oswalt (1977) did not think that the payment was a major motivator. While there are no national data, past research indicated a very high turnover rate after the cessation of payments. A study in New Mexico found a nearly 100 percent turnover in the donor pool-indicating that payment might have been more important than Oswalt hypothesized. See Piliavin and Callero ( 1991) for a discussion of the turnover rate. 20 Piliavin (1989) claimed that there had been two important changes, as a result of AIDS, that justified a reconsideration of the blood donor literature. First, there was a fear of getting AIDS by donation. Second, there had been a precipitous rise in autologous blood donation. 63

PAGE 77

Blood centers have been finding that a strict reliance on repeat donors is no longer sufficient to maintain the current supply of blood (at least prior to the September 11th bombings). Apparently, the "older" generation of donors is not being replaced by younger donors. Once again, blood centers need to understand what differentiates and motivates donors and non-donors. Below, we compare briefly the findings from Oswalt's and Piliavin's studies as a foundation for what information to incorporate into the present inquiry. Both Oswalt and Piliavin found that the past literature considered a host of. demographic factors. Apparently, the early researchers thought that a demographic profile of donors might have been used in a predictive sense, while later researchers use demographic information either as simple descriptive information, or as control variables in the very few cases where modeling was conducted. Both Piliavin and Oswalt reviewed demographic factors including age, gender race and marital status. Piliavin found the average age of donors to be between 33 to 38.21 She found that donorship rates declined dramatically for those 2'J over age 50, even though 65 was the legal cutoff. Oswalt found that 70-75 percent of donors were male; Piliavin found that percent to be going down, in fact, "women now predominate among current beginning donors" (Piliavin, 1989, 445). Women, 21 This is based on studies that asked for the exact age Most studies, however, asked for ages in a range. 22 Piliavin also points out that there is currently little justification for the 65 cutoff 64

PAGE 78

however, were much less likely to be multi-gallon donors. Both researchers found that donors were more likely to be married than were non-donors. As to race, Piliavin found a higher proportion of Caucasians than other races donate blood. The difference in donation rate by race increased between Oswalt's and Piliavin's reviews. Piliavin questioned if this race disparity might be due to blood centers' recruitment practices. Ifblood recruitment personnel presumed that Caucasians were more likely to donate blood, they might have solicited donations from other races either less often or less earnestly, thereby creating a self-fulfilling prophecy. The differences in donation rate, by race, may also be a residual of the World War II controversy regarding race and blood. Oswalt considered neither occupation nor education in his review of the literature. Piliavin tried, but found that the data were not often comparable. However, a "study of 15 blood center areas found that frequent donors had incomes roughly 30% higher than nondonors" (Piliavin, 1990, 445). Also, she found that frequent donors were better educated than occasional or non-donors. Interestingly, very frequent donors reversed the trend. They were more likely to be high school or trade school educated than less frequent donors (Piliavin, 1990, 445). Both Piliavan and Oswalt reviewed the role of altruism in relation to blood donation. Oswalt noted that altruism was the most often cited reason for the donation decision. However, he wondered if this were an explanation or a rationalization. Sympathetic to the egoistical account of altruism, Oswalt cited 65

PAGE 79

research from Osborne and Bradley (1975) that stated, "77 per cent of donors reported feeling pride, virtue, worthwhile, or having a halo for what they did" (Osborne and Bradley, 1975, 125). Piliavin seemed to agree with Oswalt's assessment and called for more theory-based research in the area of altruism. While Piliavin seemed unimpressed with the literature on altruism, she seemed more impressed with research on motivations for donation.23 She considered extrinsic and intrinsic rewards and incentives for donation. Altruism, in its pure sense, for instance, was an intrinsic reward, while social pressure was an extrinsic incentive. Piliavin commented that the literature was contradictory on which rewards and incentives were most effective.24 A few studies looked at perceived community needs and support as motivation for donation (Piliavin, 1990). Building on such work, Piliavin found that communities that perceived strong support for donation were rewarded with higher donation levels ( 1990). In addition, social pressure had been considered as a motivator for donation (Drake, Finkelstein and Sapolsky, 1982). Finally, Piliavin suggested that there might be something like an addiction to donation ( 1987, 1990) In such instances, individuals donated blood because of the literal "adrenal-high" that they experienced after donation. This "high" became addictive, and hence, perpetuated the donation behavior. 2 Piliavin does not consider motivations for altruism 66

PAGE 80

Piliavin concluded her review of the literature on blood donation with the following critical summary: The quality of the research in this area is not too high. In 1977, Oswalt noted that few studies had begun to manipulate and experimentally control the variables involved in donating. In reviewing the studies since that time, we have noticed little change .... [D]etails of research design, sampling, questionnaire return rates, etc. are often missing When the details of design and analysis are available, the work is often of poor quality .... In addition, data are usually presented as simple numbers or percentages of donors who reported certain motivations or behaviors. Except in journals not easily accessible to donor recruitment professionals, it is rare to find studies that relate motivations to actual behavior. Also controlled analyses, such as multiple regressions, are almost never performed on data to determine which of many aspects of motivation or experiences are most important in affecting donation (Piliavian, 1990, 26). In review, there is a large body ofliterature on blood donor motivations. However, the literature has several important limiting characteristics. The early literature (pre-1975) focuses often on differences between paid and unpaid donors. While there is some literature that compares donors to non-donors, it is difficult to interpret its usefulness since some studies indicate a nearly 100 percent turnover rate when payment for donation is discontinued.25 Such a high turnover rate implies that the current donor pool of non-paid individuals is substantially different from the paid donor pool of the early 1970s. By the early 1980s, blood centers (as per Oswalt, 1977) began to focus their efforts on retaining current donors rather than on recruiting new donors This trend continues to the present. As a result, the bulk of 24 As in most literature on donor recruitment, incentives refer to those items designed to encourage donation whereas rewards are items designed to show appreciation for past donation. In practice, the difference is less than clear 67

PAGE 81

current literature focuses on differences between single-time and multiple time donors as did the vast majority ofPiliavin's work. Unfortunately, there are no published studies after 1980 that consider the reluctance of non-donors. Moreover, the research often lacks a sound theoretical base and there are very few cases that use multivariate techniques. Summary In summary the importance of human blood has long been recognized. For nearly 2500 years, practically across the globe, bloodletting was a common practice as an intervention against disease and in some cases as a form of prophylactic medicine. In the middle of the seventeenth century, early experimentation with transfusion occurred, with donors (and often recipients) being animals. It has only been in the last 125 years that human-to-human transfusions have been attempted. The advances in the short period of time are truly remarkable, and donating and receiving blood have become a relatively commonplace activity. Since World War II, there has been a dramatic transformation in the infrastructure surrounding the acquisition and distribution of blood for medical purposes. Blood has become big business. While still being reliant upon voluntary action, blood donation has become inseparable from the profit motive Until a viable substitute for blood in its many uses of blood is developed, we can expect the demand for blood to continually 25 See Piliavin and Callero, ( 1991) for a discussion of this. 68

PAGE 82

increase and hence we will have continued concerns about supply. Issues over the supply of blood can not be adequately addressed without a clearer understanding of the motivations for blood donation. The review of the literature on motivations for blood donation highlights several important gaps. For instance, there is little research into the motivational difference between donors and non-donors. Moreover, limiting assumptions about the nature ofhuman motivation plagues much of the literature on blood donation. In the next chapter, several literature areas are introduced that can provide insights into motivations for blood donation. 69

PAGE 83

CHAPTER3 LITERATURE REVIEW Introduction Traditional accounts of the blood donation decision presupposed the "rational actor" model of human behavior. Hence, payment for donations has traditionally been seen as the logical method for motivating the donation of whole blood, with market forces determining the price. In 1978, the FDA required that all blood and blood components must be labeled as coming from paid or volunteer donors. This, in effect, eliminated the market for paid blood because paid blood was presumed to be of poorer quality, and hospitals were concerned about the risks, both medical and legal (Starr, 1998). A few years later, the American Association of Blood issued a policy statement in support of a reliance on a totally voluntary blood supply.1 1 AABB position manual, 30 52, 1980. 70

PAGE 84

Despite the reliance on voluntary behavior, there has been little understanding of what factors motivate an individual to donate and keep non-donors from donating. The predominance of the rational actor model has eclipsed the consideration of alternative explanations of the motives driving an individual's behavior. Other possible explanations can be found in neighboring literature areas including: new literature in the public choice vein, altruism, and social capital. In addition, it is useful to review the policy design literature for what it says and does not say regarding the importance of understanding the judgment and decision making processes of the targets of public policy. Rational Choice Theory Rational choice theory is a model ofhuman behavior that has become hugely influential across the social sciences. In this section, the assumptions of rational choice theory are explicated, followed by a discussion of the application of this economic theory to the study of political behavior. Next, the major criticisms of rational choice theory are explored. Finally, extensions of rational choice theory are presented, including: research on social dilemmas and risk, both derived from the Judgment and Decision-Making body ofliterature. Rational choice theory first gained prominence in the 191h century in attempts to systematize understandings of economic activity on the basis of utility theory. 71

PAGE 85

The theory held that every economic agent could derive a certain amount of utility or satisfaction from any amount of any commodity. Furthennore, subject to the limitations on available resources and information, each agent acquired the bundle of commodities that maximized the agent's utility. Rational was thus defined as the maximization of available utility, and all agents were assumed to be rational (Rosenberg, 1994, 74). Utility maximization, according to James Buchanan (1972), is a meaningless postulate until "further restrictions are imposed on the definition of utility or, technically, on the utility function. Once this step is taken, once the "goods" that the individual (in some average or representative sense) values are identified, the way is open for the derivation of hypotheses that can be tested against observation" ( 1987, 16). Buchanan does not describe what restrictions need to be imposed on the utility function, but presumably, they would be rather similar to the assumptions of rational choice theory identified by Green and Shapiro (1994). Though Green and Shapiro are strong critics of rational choice theory, they nicely summarize the theories assumptions, and hence their work will form the basis of the discussion on rational choice theory, though the discussion will be augmented by the work of traditional proponents of the theory. Green and Shaprio (1994) present five generally accepted assumptions of rational choice theory. First, it is assumed that individuals maximize their own utility. Second, there is agreement that requirements of consistency are central to rational choice theory, and should include a rank ordering of preferences and these preference orderings should be transitive. Third, rational choice theorists generally 72

PAGE 86

assume that individuals maximize the expected (as opposed to actual) value of their payoffs. Fourth, there is agreement, according to Green and Shapiro ( 1994 ), that the relevant maximizing agents are individuals. "Finally, rational choice theorists generally assume that their models apply equally to all persons under study-that decisions, rules and tastes are 'stable over time and similar among people"' (Green and Shaprio, 1994, 17, citing Stigler and Becker, 1977, 76). The concept of a social optimum (or Pareto optimality) is also a core tenet of rational choice theory (Coleman, 1994 ). The presumption is that when individuals pursue personal ends, they inadvertently promote the ends of society. "[P]ursuit of one's interests in exchange leads to an improvement for all those involved in the exchange with no loss to others. When no more voluntary exchanges are possible, a social optimum has been achieved" (Coleman, 1994, 41). As Green and Shapiro (1994) summarize: In sum, rational choice theorists generally agree on an instrumental conception of individual rationality, by reference to which people are thought to maximize their expected utilities in formally predictable ways. In empirical applications, the further assumption is generally shared that rationality is homogeneous across the individuals under study (Green and Shaprio, 1994, 17). One recurrent theme by proponents of rational choice theory involves clarifying the presumption of rationality as used by rational choice theory. Herbert Simon ( 1987) distinguishes between "substantive" and "procedural" rationality. Substantive rationality is viewed in terms of the choices it produces. Procedural 73

PAGE 87

rationality is concerned with process employed in decision making. He claims that economics is based upon substantive rationality, and the other social sciences presume procedural rationality. This seems to be similar to the comment made by Buchanan, that the economic model of behavior is "almost entirely predictive in content rather than prescriptive" (Buchanan, 1987, 17) and argues that most of the criticisms of rational choice theory stem from a misunderstanding about this. Buchanan continues: Failure to understand the descriptive and predictive content of economic theory along with a proclivity to interpret all social "science" in prescriptive terms has caused many critics to deplore the "dismal science" and to rail against the "crass materialism" that economic behavior allegedly represents. The appropriate response of the economist to such criticism should be (but perhaps too rarely has been) that he is wholly unconcerned, as a professional scientist, about the ethically relevant characteristics of the behavior that he examines. To the extent that men behave as his model predicts, the economist can explain uniformities in social order. To the extent that men behave differently, his predictions are falsified. It is as simple as that. (Buchanan, 1987 17). In a further discussion of the criticism economists have received for the use of their models outside traditional market activities, Buchanan contends that the criticisms generally stem from a misunderstanding of the descriptive versus predictive nature of economic reasoning. Moreover he contends that criticisms are a result of inappropriately conflating the distinction between "is" and "ought" (Buchanan, 1987 18). 74

PAGE 88

Arrow ( 1987) argues for a contextual understanding of rational choice theory, noting that "rationality is not a property of the individual alone. Rather, it gathers not only its force but also its meaning from the social context in which it is embedded" (Arrow, 1987, 201). It should be noted that Arrow also contends that economic theory need not always be based on rationality "as a matter of principle" (Arrow, 1987, 201). "Not only is it possible to devise complete models ofthe economy on hypotheses other than rationality, but in fact, virtually every practical theory of macroeconomics is partly so based" (Arrow, 1987, 202). He argues that the use of rational theory in some economic models (e.g. models by Keyenes, Friedman and Tobin) is "ritualistic, not essential" (203). Green and Shapiro, citing Ferejohn (1991), distinguish between "thick" versus "thin" accounts of rationality. In the thin account, "agents are assumed to be rational only in the sense that 'they efficiently employ the means available to purse their ends"' (Green and Shapiro, 1994, 17, citing Ferejohn, 1991, 282). Thick accounts of rationality are exemplified by utilitarianism and classic economics, where the assumption has been that "agents in a wide variety of situations value the same sorts of things: for example, wealth, income, power, or the perquisites of office" (Green and Shapiro, 1994, 17, citing Ferejohn, 1991, 282). Green and Shapiro conclude that that much of the work in rational choice theory "rests on unambiguously thick-rational assumptions" (1994, 19)-though there is 75

PAGE 89

considerable variation in both what is explicitly stated and what is implicitly assumed about the nature of rationality. An area of dispute within the rational choice corpus "concerns the amount of relevant information that agents can normally be presumed to possess and act upon" (Green and Shapiro, 1994, 19). Zey (1998) discusses the central role of information in making rational choices, noting that traditionally, rational choice theory presumed perfect information. Green and Shapiro ( 1994) also note the traditional assumption of the '"consumers' ability to understand and use that information" (19). However, important research, beginning with Simon's concept ofbounded rationality (1979), has demonstrated that individuals make decisions with imperfect information, and may misuse information to which they have access. Rational choice theory was original a theory of the market, but it had gained significant prominence as a predictor of a wide variety of behaviors ranging from voting to substance abuse (see Green and Kagel, 1996). The argument for the use of rational choice theory beyond market analyses is that individuals make decisions in the same way in all these domains. Classic work applying economic reasoning to political issues includes Anthony Downs' An Economic Theory of Democracy (1957), and James Buchanan and Gordon Tullock's The Calculus of Consent (1969). Their general argument is that traditional political scientists have been studying the appropriate phenomena with inappropriate methods. The major flaw, according to rational choice advocates, has 76

PAGE 90

been ignoring the micro-foundations of political behavior (Green and Shapiro, 1994, 3). Green and Shapiro note that: [R]ational choice theorists have worked out the microfoundations of these behaviors with a parsimony and rigor that has not hitherto been attempted by political scientists. By working through the logic of the incentives confronting political actors in a variety of structural settings, rational choice theorists have sought to enrich our understanding of the nature of politics and ofthe possibilities and limits of political reform (1994, 4). Hence, Anthony Downs, in An Economic Theory of Democracy ( 1957), conceptualizes a homo politicus that is analogous to the homo economicus prominent in economic theory. In Downs' conception, homo politicus is faced with uncertainty about the future (and therefore does not have complete information), arid "intends to strike an accurate balance between costs and returns" (Downs, 1957, 9). Downs' book represents one of the earliest applications of rational choice theory outside of the market. It is important to note that Downs was not only considering the political actor to be rational, but also "political parties, interests groups, and governments" (Downs, 1957, 6). With this theoretical framework, Downs proposes two main hypotheses: "parties act to maximize votes and ... citizens behave rationally in politics" (Downs, 1957, 300). Similarly, James Buchanan and Gordon Tullock in The Calculus of Consent ( 1969) apply rational choice theory to the study of the state and develop what they call a theory of collective choice. Relying on rather restrictive assumptions about human nature (e.g. perfect information and rationality), they argue that although 77

PAGE 91

decision makers often do not have complete information and do not act "rationally" in the real world, these differences will tend to cancel out so that the overall outcome will be "rational." In the end, Buchanan and Tullock argue that the application of rational choice theory does advance the understanding of political phenomena in a way that is not possible relying on traditional social science theories. The major contribution of Downs ( 1957), and Buchanan and Tullock ( 1969) is this application of economic theory to what were traditionally considered non market activities (i.e., the political arena). Their work opened to the door to the application of the rational actor model to a wide array of problems in a wide variety of disciplines. Another important contribution of this sort of work includes attempts to consider apparently unselfish behavior in a systematic manner within the framework of the model. Hence, issues such as volunteering, charity and other forms of philanthropy received serious attention (see Buchanan, 1972, 19, for a listing of some attempts). The approach required turning prescriptive norms into testable hypotheses, which then could be subjected to the rigors of positive analysis. This approach to the study of non-selfish behavior will receive more attention later in the discussion of altruism. It is perhaps the critique and extensions of rational choice theory, however, that provide the most promise for improving our understanding of the motives for blood donation. Critiques of rational choice theory are of two main types. First, 78

PAGE 92

there are criticisms levied against rational choice theory's explanatory power. Green and Shapiro summarize: To date, a large portion ofthe theoretical conjectures of rational choice theorists have not been tested empirically. Those tests that have been undertaken have either failed on their own terms or garnered theoretical support for positions that, on reflection, can only be characterized as banal; they do little more than restate existing knowledge in rational choice terminology ( 1994, 6). Second, there are concerns with the behavioral assumptions of rational choice theory (see Zey, 1992). One concern is that if the theory does not adequately describe behavior, then it will be unable to predict action. For instance, rational choice theory's assumption of a single preference ordering (utility) has drawn a great deal of attention. A person is given one preference ordering, and as and when the need arises this is supposed to reflect his interests, represent his welfare, summarize his ideas of what should be done, and describe his actual choices and behavior. Can one preference ordering do all these things? (Sen, 1990, 3 7, emphasis in original). These concerns are perhaps best summarized by Sen's comment that "[t]he purely_ economic man is indeed close to being a social moron" (Sen, 1990, 3 7, emphasis in original). Equally important are the concerns that rational choice theory's presumption of self-interested behavior does not adequately explain human behavior. Rational choice theorists would argue that helping, giving gifts and the like are actually motivated by selfish motives. As Zey laments, "[t]he anomalies of 79

PAGE 93

) unwise value-laden, altruistic, emotion based decisions do not limit the theory of rational choice for its most committed adherents" ( 1998, 88). Another important limitation has been the assumption of value neutrality. The idea of value neutrality has appeal to the social sciences as they have long desired to become more "scientific" (Rosenberg 1995). For instance in A Primer for Policy Analysis, Edith Stokey and Richard Zeckhauser, ( 1978) simultaneously advocate the use of rational choice theory as a means of policy analysis, and for the value neutrality of policy analysis without discussing the incompatibility of these positions. Likewise, Buchanan ( 1972) presents a rational choice theory where the economist (or other users of economic tools) is seen as a neutral scientist applying a positive value free tool for policy analysis. In summary, rational choice theory has been found to have insufficient explanatory power, plagued by incorrect assumptions about human motivation (Green and Shapiro, 1994). Moreover, the researchers using rational choice theory have often and inappropriately assumed it to be value neutral. Despite these limitations, the theory has been profoundly influential both in the academic literature and in policy formation (Tullock, 1972). In relation to blood policy, adherents to rational choice theory would argue that the logical way to increase the blood supply is to pay (or in some other fashion properly "incentivize") donors. The idea is that an appeal to the individual's self interest represents the best way to motivate behavior. This is in contrast to an 80

PAGE 94

appeal to other motives that might encourage the desired behavior (e.g. altruism, duty, community responsibility). Consistent with the self-interested view is a large body of literature calling for the reinstatement of paid donations as a means of expanding the blood supply (e.g. Eckert and Wallace, 1985; von Schubert, 1994; Solow, 1971; Stewart, 1984; Cooper and Culyer, 1968). While substantial amounts of policy analysis and theorizing are still done within the rational choice framework, theoretical extensions abound. One of the most promising of these areas is judgment and decision-making. Judgment and Decision-Making The judgment and decision-making (JDM) literature is often seen as an extension of rational choice theory. 2 JDM looks at the ways in which the psychology of decisions differs from the "deliberative" model presumed by rational choice theory. These ideas gained prominence with the publication of the oft-cited work ofKahneman and Tversky (e.g. 1973, 1974; also Kahneman, Slavic and Tversky, 1982) who attempt to show the biases and heuristics common in human decision-making.3 Also, Slavic's work on judgments involving risk, and JDM in 2 However, Coleman sees the heuristics and biases work by Kahneman and Tversky ( 1981) as a caveat to the narrowly construed rational choice theory (Coleman, 1990, 14) 3 For a wonderful review of the criticisms of their theory see Hammond, Human Judgment and Social Policy ( 1996). 81

PAGE 95

economic/financial situations, has popularized these ideas to a growing and increasingly receptive audience (see, for example Slovic, 1972, 1980). There are two veins of research within JDM that warrant further consideration for this dissertation. The first is the insights into social dilemmas; the second is work on risk. The work on social dilemmas is pertinent, since it provides an alternative framework for the blood donation problem, as well as alternative resolutions. The work on risk is relevant because blood donation is viewed by many potential donors as being inherently risky. This literature area provides a theoretical framework for understanding and addressing the perceptions of risk associated with blood donation. Social Dilemmas. Voting, attending a political rally, or, for that matter, donating blood are actions that are not adequately explained by rational choice theory. Green and Shapiro (1994, 68), for instance, demonstrate the lack of insights gained from the application of rational choice theory to voting, despite the huge amount of empirical work on the subject. The individual vote is not likely to determine the collective outcome, yet people regularly engage in these activities. A given individual is not likely to receive any benefit from the act of donation (assuming that there is no market for blood in the foreseeable future) or individually reverse the shortage, yet the system depends on this very act. 82

PAGE 96

Hardin nicely introduced the original concept of a social dilemma into the literature in his description of problems with grazing rights-known as the "tragedy of the commons." In this classic example, Hardin describes a situation in which individuals share grazing rights on a common green. Picture a pasture open to all. It is to be expected that each herdsman will try to keep as many cattle as possible on the commons ... As a rational being each herdsman seeks to maximize his gain. Explicitly or implicitly, more or less consciously, he asks, 'what is the utility to me of adding one more animal to my herd?" (Hardin, 1968, 162 emphasis in original). It is in each individual's self-interest to add one more animal to herd, since the herdsman will receive all of the benefit of the sale of the extra cattle. The negative consequences (externalities), however, are shared by all herdsmen. If each individual acts without consideration of others, he or she would over-graze the commons. This, then, is the prototypical social dilemma. Hardin's solution is "mutual coercion: mutually agreed upon." But alternate solutions can be found. Theorists from a variety of disciplines have identified alternate methods for eliciting the cooperation necessary for resolving social dilemmas. Dawes, van de Kragt, and Orbell ( 1997) summarize what they consider to be four main strategies for resolving these sorts of social dilemmas. 1. Leviathan (Hobbes, 1651/1947). However it is established, a central state mandates cooperation by punishing defection ... 2. Reciprocal altruism (Axelrod, 1984). Through some mechanism-perhaps biological (Trivers, 1971) cooperation on the part of one individual in a dilemma situation enhances the probability that others will cooperate later 83

PAGE 97

in the same situation or a similar one. Thus, an individual's "enlightened self-interest" is to cooperate in hopes of eliciting reciprocity. 3. Mutual coercion mutually agreed upon (Hardin, 1968). Rather than being punished for defection by a (potentially arbitrary) central authority, freely choosing people agree to provide punishments to each other for choosing a dominating defecting choice. 4. Socially instilled conscience and self-esteem (Campbell, 1975). While externally provided payoffs may define a social dilemma, social training can lead to such a "bad" conscience for choosing a dominant defecting strategy-or to such heightened self-esteem for eschewing such strategies in favor of cooperation that the individual is better off cooperating, irrespective of external consequences (Dawes, Van De Kragt and Orbell, 1997, 379, emphasis in original). Dawes, Van De Kragt and Orbell ( 1997) summarized their presentation of the four solutions to social dilemmas with the following remarks: These four "solutions" have one characteristic in common; they turn an apparent dilemma into a non-dilemma by manipulation (conscious or automatic) of the consequences accruing to the individual for cooperation or defection. Manipulation of behavior through the egoistic payoffs resulting from such consequences is compatible with: (i) psychoanalytic beliefs in the preeminence of primitive drives, (ii) behaviorist beliefs in the automatic and omnipotent effects of rewards ad punishment, (iii) conservative economic theory, (iv) social theory, (v) the insistence of sociobiologists that altruism be compatible with "inclusive fitness," and (vi) the obvious success in U.S. current societyofappeals to personal payoffs (Dawes, Van De Kragt and Orbell, 1997,379, emphasis in original). This listing provides a useful starting point, but it is interesting to note what is missing from this list. The four "solutions" for resolving social conflicts assume that resolutions must in some way appeal to an individual's self-interest. This is a point, also made by Jane Mansbridge in "The Relation of Altruism and SelfInterest": 84

PAGE 98

Other solutions derive from love or duty. That is, they require one or more of the interacting parties either to make the other's good their own or to be committed to a principle or course of action that requires cooperation. These two distinct motivations which I together call the "unselfish" or "altruistic" motivations, have been variously labeled ''sympathy" and "commitment" (Sen), "love" and "duty" (Elster), "empathy" and "morality" (Jencks), "we feeling" and "conscience" (Dawes, van de Graft, and Ordbell), or ''affection" and "principle" (Hume) (Mansbridge, 1990, 135). Mansbridge points out that even this list is only illustrative and not exhaustive. She indicates that Rawls, for instance, suggests that prisoners' dilemmas can be overcome via commitment to a common purpose ( 1971, 260) and Taylor (1987) suggests altruism can be a solution. From a philosophical orientation, there are a number of possible resolutions: for instance, a rule-utilitarian approach would lead us to conclude that cooperation could be encouraged simply by asking those in the dilemma to formulate the rule or law that, if followed, would produce the greatest amount of happiness for the greatest number and follow it (see Mill, 1863/1993, or Brandt 1975). The writings of Immanuel Kant also provide an alternative strategy for resolving social dilemmas (1785/1993). Here, the decision rule we use to choose actions would be "act as though the maxim of our actions were to become a universal law." So, one ends up with non-egoistic motivations, not because they are only better for the individual, but because it is what morality demands. Alternate resolutions to social dilemmas can also be identified from a broad reading of Robert Axelrod's reciprocal altruism solution summarized by Dawes and 85

PAGE 99

colleagues (1997). This explanation from Axelrod's classic work on eliciting cooperation in games has decidedly evolutionary roots. However, evolutionary theorists, starting with Darwin, offer a much more nuanced understanding of cooperation in social dilemmas.4 Axelrod's approach to evolutionary theory considers the individual to be the adaptive unit. Alternatively, we can consider both the individual and the group to be adaptive units. The latter alternative is much less tidy, but is also much more realistic. Sober and Wilson argue that there is a long history of seeing groups as adaptive units, subject to the laws of evolution ( 1998). If the group is the adaptive unit, then resolutions of social dilemmas no longer look self-interested. The purpose of this discussion is not to identify the most suitable solution, but to recognize that alternatives to egoistic self-serving resolutions exist. Researchers within the JDM tradition have also identified factors, not easily identifiable as egoistic, that increase the chances of cooperation. Dawes, Van De Kragt and Orbell ( 1997) find, based on ten years of research, that group identity alone can radically alter individual's willingness to cooperate in social dilemmas. In their game-theory style research, the authors systematically alter the group identity of the subjects in relatively simple ways, including letting individuals chat or dividing subjects into groups before an experiment (of a 4 The arguments will be only briefly summarized here, as they wiii be covered more extensively in the discussion of altruism later in this chapter. 86

PAGE 100

modified prisoners' dilemma sort). Comparison to situations in which group identity is not manipulated reveals dramatic increases in cooperative behavior. The authors conclude that group identity, even in these simple games, significantly increases cooperation: this finding, they argue, fits with an evolutionary framework in which "sociality is primary for humans" (Dawes et al., 1997, 379 emphasis in original). They conclude, "[i[t is not just the successful group that prevails, but the individuals who have a propensity to form such groups" (Dawes et al., 1997, 390). The authors note that previous research has considered what factors have lead to sociality, but then point out that there is no indication that people were ever not social. Finally, they point out, as well they should, that sociality is not morality, as numerous examples of amoral and immoral groups can be readily identified. Interestingly, Aristotle, in his Nicomachean Ethics, is one of the first theorists to propose that humans are social animals ("zooa politica") (Aristotle, -350 B.C./1999). Other work from JDM provides us with insights into the factors that might enhance cooperation in social dilemmas. For instance, Axelrod notes that communication and expectations of future interactions increase the likelihood that an individual will adopt cooperative strategies in game situations (Axelrod, 1984). Contrarily, Frank, Gilovich and Regan find that studying economics decreases cooperation in games (2000). 87

PAGE 101

In summary, solutions to social dilemmas can be found from a variety of orientations. Research from judgment and decision-making has offered us interesting groundwork into these ideas-and can help us flesh out some of the particulars. This research, though, would be remiss if only solutions appealing to egoistic motives are considered "appropriate" resolutions to social dilemmas. Dawes and colleagues provide one example of a factor, group identity, that increases motivation for cooperation that is not easily explained by the egoistic model. This is likely not simply an anomaly, but an indication that solutions solely aimed at egoistic motivations may be overly narrow. This is a theme we will pick up again later in this literature review. Risk The second line of research within JDM that warrants consideration is perceptions of risk. Elster ( 1986) has acknowledged that choice situations involve some amount of incomplete information; as such, they entail risk. Previous research has revealed that many potential donors consider blood donation to be risky (Andaleeb and Basu, 1995). The discussion of risk, as relevant to the blood donation decision, can be bifurcated into two areas: perceptions of risk, and differences between experts and lay people in risk estimations. Slovic, Fischhoff and Lichtenstein ( 1982) discuss the judgmental biases that are likely to affect perceptions of risk (e.g., availability and overconfidence). The availability bias causes individuals to judge an event to be more likely if it is easier 88

PAGE 102

to recall. "One particularly important implication of the availability heuristics is that discussions of a low-probability hazard may increase its memorability and imaginability and hence its perceived riskiness, regardless of what the evidence indicates" (Slovic, Fischhoff and Lichtenstein, 1982, 465). Individuals, according to the research, are more likely to remember vivid or salient events. Slovic and colleagues discuss the importance of media coverage on risk perceptions. In one study of newspaper coverage of causes of death found that the newspapers: [S]tatistically frequent causes of death (e g., diabetes, emphysema, various forms of cancer) were rarely reported In addition, violent, often catastrophic, events such as tornadoes, fires, drownings, homicides, motor vehicle accidents and all accidents were reported much more frequently than less dramatic causes of death having similar (or even greater) statistical frequencies (Slovic, Fischhoff and Lichtenstein, 1982, 468). Interestingly, there was a high degree of correspondence between the newspaper coverage and individuals' judgments about the likelihood of causes of death. "Rare causes of death were overestimated and common causes of death were underestimated" (Slovic, Fischhoff and Lichtenstein, 1982, 467). The systematic errors in judgments of risk were also combined with overconfidence in one s judgments, a second judgmental bias. That is, subjects were often very confident of their own estimations of risk, event when their estimations differed dramatically from actual levels of risk. Gerd Gigerenzer and his research group ( 1999) argue it might appear that people are being irrational when they use these heuristics, but perhaps heuristics are actually the most optimal 89

PAGE 103

method to deal with everyday risk-they are "fast and frugal heuristics"5 that make us smart (Gigerenzer, 1999). McClelland, Schulze, and Hurd, (1990) also note the discrepancy between experts' and lay persons' judgments of risk. In their study, they find that lay persons' estimations of risk associated with a hazardous waste site were substantially different than the experts' estimations. The difference can be partially explained by the lay persons' sensitivity to "perceptual cues that provide people with continual reminders about the hazard" (McClelland, et al., 1990, 495). The authors conclude their discussion with the observation that there has been much work on the consequences of"under-estimation" of risk but not on "over estimation" of risk. Blood donation, however, may be a case of "over-estimated" risk, because reluctance to donate may partially reflect the public's inclusion of subjective risk assessments in their decision-making processes. Perhaps the most important work on perceptions of risk comes from the work of Slovic and his colleagues discussing the divergence between expert and public perceptions of risk, for such issues as the risk from nuclear power plants (Fischhoff, Slovic, and Lichtenstein, 1982, 1985). 6 They suggest that when there are important differences between how experts and lay-persons view risk in a particular circumstance, it might be useful to investigate the root of the public's 5 A term coined by Gigemezer, Todd and the ABC Research Group (1999) 90

PAGE 104

perceptions as this might help uncover additional information or perspectives that the public is using in their risk assessment process. This divergence could be occurring for three reasons. Perhaps either the laypersons or the technician are wrong. This could be occurring for instance, if as Fischoff and colleagues suggest-lay persons' perceptions of risk do not correspond well with "actual" levels of risk. So, perhaps the experts have it right, but lay people are simply no good at this task. Alternatively, the experts and the lay people might be considering different risks. Finally, there might be an intervening variable that explains the difference (e.g., lay-persons may distinguish between voluntarily and involuntary exposures to risk). Slovic (2000) provides a possible intervening variable: trust. In his article entitled "Risk, Trust and Democracy", he discusses four psychological tendencies that "create and reinforce distrust in risk situations" (Slovic, 2000, 507): 1. Negative (trust-destroying) events are more visible or noticeable than positive (trust-building) events. 2. When events do come to our attention, negative (trust-destroying) events carry much greater weight than positive events. 3. .. sources ofbad (trust-destroying) news tend to be seen as more credible than sources of good news. 4. .. distrust, once initiated, tends to reinforce and perpetuate the distrust. (Slovic, 2000, 506). He summarizes: 6 There is a vein of research that documents both lay-persons' and professionals' difficulties in incorporating risk into their decisions; see Arkes and Hammond (I 986), for a selection of articles on this theme. 91

PAGE 105

Conflicts and controversies surrounding risk management are not due to public irrationality or ignorance, but instead, can be seen as expected side effects of these psychological tendencies, interacting with our remarkable form of participatory democratic government, and amplified by certain powerful technological and social changes in our society (Slovic, 2000, 507). Slovic provides a particularly useful example to help illuminate the between risk and trust. He explains, "France leads the world in the percentage of electricity generated by nuclear power" (Slovic, 2000, 508). The perceptions of risk of nuclear power in the United States and France are similar-nuclear power is seen as very risky. The difference is that the French have a high degree of trust in their government and in the experts who run and operate the nuclear plants (Slovic, 2000, 509).7 The Americans, alternatively distrust government and science and industry. In the French system ''policy formation and implementation" are not accessible to public intervention. Slovic suggests that given the high levels of distrust in America, the French system would not work in the United States, he concludes: Given that the French approach is not likely to be acceptable in the United States, restoration of trust may require a degree of openness and involvement with the public that goes far beyond public relations and "two-way communication" to encompass levels of power sharing and public participation in decision making that have rarely been attempted. (Slovic, 2000, 509). 7 An alternate interpretation is that this is a result of the highly authoritative nature of the French bureaucracy. 92

PAGE 106

In summary, the literature on risk reveals that systematic heuristics and biases plague our estimations of risk, as they plague other forms of judgment. Moreover, lay-persons and experts may differ in their perceptions of risk. The literatures on risk and social dilemmas can be seen as augmenting an understanding of blood donation as derived from the rational actor model of human behavior. However, the rational actor model contains troubling implications-namely the supremacy of self-interested behavior. Other literature areas, including altruism and social capital, can broaden our discussion of motivations for donation as well as provide a framework for a consideration of appropriate policy options to address the disturbingly low levels ofblood donation that prevail (except, of course, when blood donations temporarily sky-rocket after national or regional crises). Altruism Altruism can be defined as an unselfish act that benefits others (Hardin, 1993). Beyond this seemingly uncontroversial starting point, there is considerable disagreement surrounding the concept. Generally, there are two schools of thought on altruism. The "egoists" argue that all apparently altruistic behavior is in fact self-interested. The egoists argue that individuals engage in altruistic acts not for the benefit of others but for the benefits they receive from engaging in the act. Hence they believe that altruism does not truly exist; it is merely self-interested behavior in another guise. 93

PAGE 107

Research adhering to egoistic views derives from many disciplines. For instance, economists have couched altruism in terms of utility (see Zey, 1998). The assumption is that the individual acts only to improve his/her interpersonal utility function. Bateson ( 1987), a psychologist, has identified a variety of egoistic accounts of altruism, which he terms "pseudo-altruistic." For instance, pro-social behavior, by its very name, is agnostic about altruistic intentions. All that can be observed is behavior, not motivation. Hence, this line of work looks into how individuals volunteer their time and resources, but does not consider the motivations behind the donations. Likewise, there is literature on what Bateson terms self-rewards for altruism (e.g., BarTal, 1997, Schwartz, 1977). This is a redefinition of altruism "to include seeking benefits for the self, so long as these benefits are internally rather than externally administered" (Bateson, 1987, 69). The final "psuedo-altruistic" approach comes from one of the most prolific writers on blood donor motives, Jane Piliavin, who holds that altruism can be seen as arousal-reduction (Piliavin and Piliavin, 1973; Piliavin, et al. 1982). By this, she and her co-authors mean that as the unpleasantness of a situation increases for a bystander, bystanders become motivated to reduce the unpleasantness. According to the arousal-reduction theory, individuals would, for instance, donate blood, because they would feel anxiety or other negative emotions if they did not. Such individuals are not motivated by 94

PAGE 108

helping others, but by their desire to mitigate the psychological burden of not helping. In contrast to the egoists' self-interested accounts of altruism, other researchers believe that altruism, in the classic sense, does indeed exist. These "pure-altruists" believe that individuals have the capacity for altruism and contend that individuals regularly act in ways that are against their own best interest for the benefit of others. Again, we see multi-disciplinary contributions, namely from philosophy and psychology. Adherents to the classic view of altruism have put much effort into defining altruism. For instance, Leeds defines altruism as an act that is: "(a) beneficial to at least one person, (b) emitted voluntarily, and (c) not motivated by the donor's expectation of achieving immediate reward for self' (Quigley, et al., 1989). Dovido takes a different analogous approach and provides four dimensions useful for identifying altruistic behavior: "(I) the consequences of the act for the recipient, (2) the locus of reinforcement (i.e., internal or external), (3) the intent of the benefactor, and (4) the motivation underlying helping" (Dovido, 1984, 363). More directly, Schmidtz defines altruism as behavior that is motivated by regard for others ( 1993, 52). Consistent in all of these accounts is a consideration of the motivation behind the act. The motivation for an altruistic act must be for the benefit of others, not for the self. Beyond definitional issues, philosophers are contributing to the understanding of the existence and nature of altruistic behavior. David Schmidtz 95

PAGE 109

(1993) dismisses the egoistic accounts of altruism that incorporates other-regarding behavior into a rational actor model. S. H. Schwartz ( 1993) contends that altruism is ubiquitous and is evident in a wide variety of human interactions in a vast array of settings. He argues, however, that the "cultural influences that regulate social relations and that contribute to establishing the boundaries between self and other can have profound effects on altruism" (Schwartz, 1993 314). Russell Hardin ( 1993) ties philosophical theories of altruism together with theories of mutual advantage and discusses the conflict between the two. The contributions from philosophers are bolstered by psychologists' methods for identifying and operationalizing altruism. For instance, there have been useful insights into an individual's perceptions ofwhat makes an act altruistic (Quigley, Gates and Tedeschi, 1989) and the development of scales for altruism (Rushton, Chrisjohn and Fekken, 1981 ). Rushton argues that a trait of altruism exists and seeks to identify '"consistent patterns of individual differences' in altruistic behavior" (Rushton, 1984, 271 ). Piliavin and Chang ( 1990) summarized the literature on altruism and draw the important conclusion that "there appears to be a 'paradigm shift' away from the earlier position that behavior that appears to be altruistic must, under closer scrutiny, be revealed as reflecting egoist motives" (1990, 27). It seems obvious to many that donating blood is in some sense an altruistic act. It invokes significant costs and offers little discernable benefit. Hence, a 96

PAGE 110

keener understanding of altruism and the conditions under which it is more likely to emerge will be useful for unraveling the mystery behind decreased donations. Egoistic accounts have a limited ability to explain the donation decision. Cookies and juice seem to be an insufficient reward or justification for blood donation. Likewise, reducing donation to self-motivated behavior severely limits the types of motivations that can be considered. Hence, this dissertation focuses on "purealtruism" as opposed to egoistic altruism. In addition, altruistic motives may have great importance for the emergence and maintenance of social capital. Egoistic and altruistic motives are posited to be influential for the blood donation decision. Social Capital There are at least two competing, or perhaps complimentary accounts of social capital. The original account is from James Coleman, in his book The Foundations of Social Theory ( 1990) 8 Social capital, as explained by Coleman, reflects the relations between individuals that enable them to get things done collectively that otherwise would not be possible alone Social capital is defined by its function. It is not a single entity, but a variety of different entities having two characteristics in common: They all consist of some aspect of a social structure, and they facilitate certain actions of 8 Actually, Coleman did not create the concept of social capital but he is the first theorist to popularize its use and to provide substantial theorizing on the concept. Coleman (1998 300) credits Loury (1977; 1987) with the first use of the term. According to Coleman "in Loury's usage social capital is the set of resources that inheres in family relations and in community social organization and that are useful for the cognitive or social development of a child or young person" (Coleman, 1998, 300). 97

PAGE 111

individuals who are within the structure. Like other fonns of capital social capital is productive, making possible the achievement of certain ends that would not be attainable in its absence. Like physicai capital and human capital, social capital is not completely fungible, but is fungible with respect to specific activities. A given form of social capital that is valuable in facilitating certain actions may be useless or even harmful for others. Unlike other forms of capital, social capital inheres in the structure of relations between persons and among persons. It is lodged neither in individuals nor in physical implements of production (Coleman, 1990, 302). For Coleman then, social capital is a neutral term simply describing the relations between individuals, but making no assumption about the results of these relations. Social capital can be acquired from any social relationship. This includes relationship from organizational involvements, familial and school contacts. Putnam (2000), citing Coleman as the originator of the concept of social capital, presents an appreciably different picture of social capital. Putnam, as Foley and Edwards point out, uses a definition "whereby social capital comes to denote attitudes and habits (or 'values') conducive to civic engagement" (1997, 553).9 There are two important differences between Coleman's and Putnam's conceptions of social capital. Putnam's conception like that current in contemporary popularized versions -both narrows the field of what might constitute or generate social capital and bestows on it a valorization that Coleman's conception cannot sustain by itself(Foley and Edwards, 1997, 553). For Putnam, then, social capital is a decidedly positive phenomenon, whereas for Wilson it is neutral. Putnam also places much more value on organizational 98

PAGE 112

involvement than Wilson does. This shift in focus about organizational involvements reflects "what Richard Wood calls the 'second face of culture,' whereby social capital comes to denote attitudes and habits (or 'values') conducive to civic engagement" (Foley and Edwards, 1997, 553). Specifically, Putnam identifies three basic forms that social capital can take trust, norms and networks. Putnam describes social capital as being a public good, and "like all public goods, social capital tends to be undervalued and oversupplied by private agents" (Putnam, 1993b, 170). Putnam summarizes: "[t]rust lubricates cooperation. The greater is the level of trust within a community, the greater the likelihood of cooperation. And cooperation itself breeds trust" (Putnam 1993b, 171 ) Highlighting the importance of trust, Putnam ( 1993b) and Levi ( 1996) both note"[ c ]ollective life in civic regions is eased by the expectation that others will probably follow the rules. Know that others will, you are more likely to go along, too, thus fulfilling their expectations" (Putnam, 1993b ). Putnam has received a fair amount of criticism for his work In particular, his conception of social capital has been criticized for placing undue emphasis on organizational involvements. For instance, Foley and Edwards ( 1997) critique Putnam's focus on "the neo-Tocqeuivillian argument" (553), that is, the question why "voluntary associations should be singularly adept at promoting the attitudes and habits necessary for engaged and 'civil' citizenry" (Foley and Edwards, 1997, 9 This is similar to Bellah's and others notion of Habits of the Heart (1985). 99

PAGE 113

553). Furthennore, Newton (1997) argues that other institutions, such as "family, school and the workplace might more reasonably be expected to generate the sorts of commitments proponents of 'civic renewal' have in mind" (as summarized nicely in Foley and Edwards, 1997, 553). Putnam has also been critiqued for developing a social capital divorced from its broader context (Berman, 1997), for not considering the importance of national organizations and for perverting and watering down Coleman's original use ofthe term (Wood, 1997; Greeley, 1997). Despite these criticisms, Putnam's work remains remarkably influential. For the purposes of this dissertation, there are three pertinent areas of social capital theory: the development of social capital; measurement of social capital; and the complex relationship between social capital and trust. This discussion of social capital concludes with a brief summary of current trends in social capital and an evaluation of the implications these trends have for blood donation and other types of voluntary behavior. The Development of Social Capital There has been some promising work on the conditions that are conducive to the emergence of social capital. Y ouniss McLelland and Yates ( 1997) argue that social capital develops in the young. They find that "students who participated in high school government or community service projects, meant in the broad sense, are more likely to vote and to join community organizations than are adults who 100

PAGE 114

were non-participants in high school" (Youniss, McLelland and Yates, 1997, 620). They offer two main explanations for why "early belonging" matters in terms of civic involvement. First, they argue, "on a practicalleve1 it introduces youth to the basic roles and processes (i.e., organizational practices) required for adult civic engagement" ( 1997, 623-624 ). Second: [I]t helps youth incorporate civic involvement into their identity during an opportune moment in its formative stages. Participation promotes the inclusion of a civic character into the construction of identity that, in tum, persists and mediates civic engagement in adulthood. The fonnation of civic identity, then, is the hypothesized developmental link across time and the factor that differentiates adults in the degree of their civic engagement" (Younnis, et al., 1997, 624). In addition, they assert "[m]any youth organizations typically provide direct exposure to explicit ideological orientations and world views" (Y ounnis, et al, 1997, 624 ). They conclude, "individuals acquire practices that are constitutive of civic identities" ( 1997, 629). Youniss and colleagues, like Putnam, adopt a neoTocquevillian stance. In a Coleman version of social capital development, there would likely be much less emphasis on civic involvement andmore emphasis on other sorts of less formal relationships that could help build social capital. Nevertheless, what they did find is important; early involvement in civic life appears to be an excellent predictor of later involvement as adults. 101

PAGE 115

Building on the work of Putnam and Coleman, Newton (1997) identifies three variants of social capital in the literature. These are norms and values, networks, and consequences. The main contribution from Newton's work, however, is his exploration of trust as a central and recurrent theme in the discussions of social capital. He fears that trust and social capital form a chickenand-egg problem. Do high levels of social capital lead to trust? Or, is trust a precondition of social capital? Francis Fukuyama, in Trust ( 1995), argues that trust is requisite for the smooth running of both the economy and society. Fukuyama ( 1995) demonstrates that there has been a dramatic decline in trust and contemplates the consequences of this decline. Putnam (2000) as well as Fukuyama ( 1995) identifies trust as the linchpin of smooth-running social interactions (and, in fact, of a smooth-running society). Putnam identifies two sources for trust-norms of reciprocity and networks ofsocial civic engagement (Putnam, 1996, 171-4 as quoted in Levi, 1996, 47). Levi explains: Participation in dense networks ofhorizontal interactions of relative equals produces norms of reciprocity, provides sanctions for defectors, offers some information about others, and creates a "culturally-defined template for future collaboration". Drawing on findings from non-cooperative game theory, he argues that a tit-for-tat strategy, particularly in the fonn of what Robert Sugden labels the "brave reciprocity of the 'mutual aid game' is a self-sustaining equilibrium. Thus, if people act trustfully, they tend to 102

PAGE 116

cooperate and invite cooperation in return'' (Levi, 1996, 46 quoting Putnam, 1996).10 Russell Hardin makes some important contributions to the discussion of trust, clarifying that trust in government is not analogous to trust in other people"most of the time" (2000, 31 ). "A personal relationship involving trust is far richer and more directly reciprocal than a citizen's relationship to government" (Hardin, 2000, 33 ). Hardin is arguing for a distinction between trust and confidence, where "trust" refers to personal relationships and "confidence" refers to governmental relationships. Finally, Hardin makes some observations about the relationship between social capital and trust. He points out that while the decline in social capital appears to be particularly pronounced in the United States, the decline in trust (confidence) in government is not unique to the United States. As a result, he suggests that perhaps in the United States, the link between trust and social capital (as posited by both Putnam and Fukiyama) might be real, but it is likely only part of the explanation. Alas: he does not suggest other potential explanations. There seems to be a consensus that trust is an important part of the discussion of social capital. There are, however, concerns about the direction of 10 Although more recent research has discovered that in a "noisy" environment where feedback is imperfect and misidentification of intentions occurs, tit for-tat as a general strategy loses to a tit-for tat-with-forgiveness approach. I 03

PAGE 117

causality. For the purposes of this dissertation, it will be presented that trust is an important component of social capital. More on this can be found in Chapter Four. Operationalizing Social Capital Admittedly, with the controversy surrounding definitions of social capital (Glasser, Laibson and Soutter, 1999; Newton, 1997; Greeley, 1997), it is little wonder that there is broad disagreement on how to best operationalize the concept. Previous research provides us with numerous examples of how social capital has been defined. Putnam recognizes that the forms of social capital are "multiple and complex" ( 1995a). He laments that no individual measure is "perfect for testing the hypothesized decline in social connectedness, although the consistency across different measuring rods is striking" (1995a, 664). As such, he adopts multiple measures of social capital in his work (for instance, group membership, political participation, time spent volunteering) and encourages other social capital researchers to do the same (Putnam, 1995a, 76). The same approach will be taken in the present dissertation. There is one final challenge for operationalizating social capital. This is the controversy over the appropriate level of analysis for social capital: individual or group/community. Coleman, as Foley and Edwards (1997b) note, makes it clear that social capital resides in relations or structures between individuals and groups 104

PAGE 118

and not within the individuals themselves (1997b, 670). However, it "cannot be conceived in purely structural terms because even in its structural sense it carries a cultural weight ('expectations', 'obligations', 'trust') that is nested in structure but not simply reducible to structure" (Foley and Edwards, 1997b, 670). Trends in Social Capital and Implications for Blood Donation and Other Types of Voluntary Behavior Social capital is important because it helps society "overcome the 'freerider' problem that bedevils organizers and recruiters trying to mobilize individual effort to achieve collective goals. Social capital makes it easier for people to link their own identity and interests with those of their community" (Wilson and Musick, 1998, 799). In Putnam's well-known work on social capital (1994), he documents a disturbing decline in social capital and attributes it to a decrease in factors such as social and organizational membership. He explains that social capital theory "presumes that, generally speaking, the more we connect with other people, the more we trust them, and vice versa" (Putnam, 1994, 6). Hence, he attributes the decline in social capital to a variety of factors ranging from pressures of time and money, mobility, suburbanization and the changing role of women (see Putnam, 1995b and 2000, for in-depth discussions of these issues). This posited decline in social capital might pose an alternate explanation for the recent and disturbing reduction in new blood donors, i.e., people are less 105

PAGE 119

inclined to civic involvement, and hence blood donation rates also decline. If so, it may be that social capital had, in part, been encouraging donation. And conversely, as individuals evince less social capital, they feel less connected to their community, and hence are less likely to donate blood, or engage in any sort of voluntary behavior. Moreover, social capital might help explain the dramatic increase in blood donations after the September 11th bombings in New York and Washington D.C. While the effect of major crisis on social capital has not been addressed (yet) in the literature, there are some discussions that can help explain the shift in donation. First, Putnam (2000) describes the impact of war on social capital, noting "membership in civic associations has spurted after both major wars in the twentieth century, and political scientist Theda Skocpol has extended this argument to the whole of American history" (Putnam, 2000, 267). In addition, union membership had grown "during and immediately after major wars" (Putnam, 2000, 267). Putnam provides numerous examples of the upsurge in patriotic zeal that accompanies and then follows war. One particularly pertinent example is from historian Richard Lingeman, noting: American flags were displayed everywhere-in front of homes, public buildings, fraternal lodges. Elks, Lions, Kiwanis, Rotary, even trailer camps, gas stations, and motor courts had them. The war reinforced solidarity even among strangers: 'You just felt that the stranger sitting next to you in a restaurant, or someplace, felt the same way you did about the basic issues' (Lingeman, 1970, 71 as cited by Putnam, 2000, 268). 106

PAGE 120

As a result ofthe war: voluntary activity increased dramatically; important voluntary associations were founded (including the American Red Cross); and other organizations greatly expanded their membership (Putnam, 2000). These examples do not tell us why increases in civic behavior occur, but they provide historical precedents for understanding the increase in voluntary civic behavior that has followed the September 11th bombings. Putnam provides several possible explanations of the high level of social capital after World War II. He proposes "the possibility that a wartime Zeitgeist of national unity and patriotism that culminated in 1945 reinforced civic mindedness" (Putnam,2000, 267 emphasis added). Further, he quotes William Graham Summer as describing how external conflict increases internal cohesion: A differentiation arises between ourselves, the we-group, or in-group, and everybody else, or the others-group, out-groups ... The relation of comradeship and peace in the we-group and that of hostility and war towards others-groups are correlative to each other. The exigencies of war with outsiders are what make peace inside ... Loyalty to the group, sacrifice for it, hatred and contempt for outsiders, brotherhood within, war likeness without all grow together, common products of the same situation (Putnam, 2000 267 citing Summer, 1911, 12-13). The description of increased patriotism --of a greater feeling of "we" --is used by Putnam to describe the exceptionally high level of civic mindedness in the "civic generation" (born in the 1925-1930, according to Putnam, 2000, 254) as compared to those born later, the baby-boomer generation (born between 1946 and 1964, according to Putnam, 2000, 257) and Generation Xers (born between 1965 107

PAGE 121

and 1980 according to Putnam, 2000, 259). Putnam concludes that the generational differences can be partially explained by having a generation's "values and civic habits [formed] during a period of heightened civic obligation" (Putnam, 2000, 272). He also describes the transitory nature of the patriotic zeal and notes that, by the middle of the war, as compared to the beginning, voluntary activities were declining, as were organizational involvements. The relationship of transitory patriotic zeal to blood donation in the post bombing era is clear. We should not be surprised by the dramatic increases in donation that occurred directly after the bombings-blood banks were almost on every corner -nor should we be surprised when these high levels of donation taper off. The impact ofthe bombings on blood donation is a theme that will be revisited later in this dissertation. In summary, the theories of rational choice, altruism and social capital have been explored for the purpose of improving our understanding of the motivations for blood donation. An adequate understanding of the donation decision is logically a function of individual motivations to donate blood as well as the donation context. The policy design literature will be reviewed for insights into blood policy. 108

PAGE 122

Policy Design "Policy design refers to the content or substance of public policy-the blueprints, architecture, discourses, and aesthetics of policy in both its instrumental and symbolic forms" (Ingram and Schneider, 1997, 2). Ingram and Schneider describe policy designs as both dynamic and contextual; they describe the process of policy design as a purposeful and normative enterprise ( 1997, 3 ). They also elucidate the core elements of a policy design: From an empirical perspective, policy designs contain specific observable elements such as target populations, (the recipients of policy benefits and burdens), rules, (that guide or constrain action), rationales (that explain or legitimate the policy), and assumptions (logical connections that tie the other elements together) (Ingram and Schneider, 1997, 2). The manner in which these policy elements are "chosen and linked," according to Ingram and Schneider (1997, 101) will ultimately determine the success or failure of public policy. In fact, Ingram and Schneider implicate policy design in the numerous public policy failures that have occurred over the past few decades (Ingram and Schneider, 1997, 66). In particular, they blame the deleterious impact of assumptions embedded within policy design elements. For instance, citing Petraea, they explain that [i]nstitutions that presume self-interest as a motivation are likely to produce such motivations as the institution's legitimate self-seeking behavior and, furthermore, seriously disadvantage anyone who attempts to engage in more cooperative collaborative efforts (Petraea, 1991, as summarized by Ingram and Schnieder, 1997, 50). 109

PAGE 123

The implication is that if a given policy was designed with embedded assumptions about self-interested behavior, then self-interested behavior may have inadvertently been encouraged. In some sense, this is a reformulation ofTitmuss's early concerns about U.S. blood policy ( 1971 ). He argues that a blood policy based on market concepts promotes self-interested behavior that is consistent with the market model at the expense of promoting other, perhaps more socially efficacious motives such as altruism. Another element of policy design is emphasis on understanding the judgment and decision-making processes of the target population. Bobrow and Dryzek in Policy Analysis by Design (1987), discuss the importance of"having sensitivity to the proclivities" of the audience of a policy analysis ( 1987, 19), where the "audience" is defined as "those in a position to further or hinder a policy, or those the policy will ostensibly serve" ( 1987, 19). Hence, Bobrow and Dryzek (1987, 20) point out that "the idea here is not necessarily to generate analysis acceptable to the audience but simply to take its position as one starting point-of only for critical dialogue" (Bobrow and Dryzek, 1987, 20). Schneider and Ingram (1997), devoting nearly 50 pages to their discussion of policy targets, recognize that for a policy to be effective, target populations must "behave in ways that are needed to achieve policy goals or solve problems" (Schneider, 1987, cited in Schneider and Ingram, 1997, 84 ). They go on to note "targets and their choices are also central to the way citizens construct their role I I 0

PAGE 124

and that of government" ( 1997, 85). Despite recognition of the importance of the choices of policy targets, they go no further in an exploration of the importance of considering how the target populations perceive the policy, or how they are likely to interact with the policy.11 A post-positivistic conception of the policy sciences can provide the logic for incorporating the views of policy targets into a more prominent role within policy design. Post-positivistic views of the policy sciences rest on three basic tenets. First, there is the assumption that the strictly positivistic methodologies are wanting. Bobrow and Dryzek ( 1987) make the argument that the choice of an analytic frame can ha:ve three important results. First, it makes some topics of inquiry more central and salient than others. Second, it makes some kinds of policy instruments look markedly more attractive than others. Third, it makes some social consequences more legitimate than others and thus affects the likelihood that public sector resources will be harnessed to their pursuit (Bobrow and Dryzek, 1989, 8). As a result, post-positivists argue for careful choice of an analytic frame and strongly embrace adopting a multi-disciplinary approach to policy problems (Bobrow and Dryzek, 1989; deLeon, 1997). This approach is adopted in the present inquiry. The final tenet of post-positivism is the belief that it is imperative to democratize the policy analysis. Discussions of deliberative democracy stress the 11 But ultimately, this is an aside in Schneider and Ingram's main concern about target populations that they are socially constructed in sometimes negative, degenerative ways. Ill

PAGE 125

importance of including citizen voices in the design and implementation of public policy and in the administering of a democracy (see, for instance, deLeon, 1994, 1997). Fischer notes that under this deliberative model, citizenry preferences are not presumed. As he says: "[u]nlike most contemporary policy analysis, the postpositivistic approach would not take individual preferences as 'given ... but would instead have to account for where people get their images of the world and how they shape their preferences"' (Fischer, 1998, quoting Stone, 1988, vii). A deeper understanding of the judgment and decisionmaking (JDM) process of policy targets is consistent with the post-positivistic view ofthe policy sciences, but the JDM process is an often overlooked element of contemporary policy design. Blood policy development and implementation reveal a lack of consideration of the JDM processes employed by the individuals on whom the policy is dependent potential donors. Summary Rational choice theory, the philosophical bedrock of much of our current public policy, offers a starting point to understand the assumptions about human behavior embedded in our current blood banking system. Despite the widespread reliance on rational choice theory, it has been found wanting as a descriptor of human motivation. Hence, we have looked to other literature areas for alternative 112

PAGE 126

understandings of why an individual would engage in an act that provides them with little obvious benefit and some significant costs. Research from the JDM tradition highlights the manners in which individual decision making diverges from the rational choice model. Work on social dilemmas is relevant, as it provides alternate understandings of the blood donation problem, as well as alternate resolutions. The major finding is that there are resolutions to social dilemmas that emerge once other than self-interested motivations are considered. Moreover, group identity, communication, and expectation of future interactions are factors that have been found to increase the chances of cooperation in social dilemmas. These factors do not sit neatly with a traditional conception of rational choice theory. In summary, resolutions to social dilemmas that appeal only to egoistic motivations are likely to be overly narrow. The literature on perceptions of risk also provides insights into the blood donation decision. From this literature review, we found that perceptions of risk are subject to the same sorts of heuristics and biases that plague other types of decision making. Moreover, we found that there is commonly a divergence between expert and lay persons' perceptions of risk. The difference in risk estimations highlights the importance of acquiring an in-depth of understanding of perceptions of risk, since lay person's risk estimations might be constructed differently than the estimations of risk calculated by "the experts." 113

PAGE 127

The literature on altruism also augments our understanding of the motivation for blood donation. Though some theorists have attempted to incorporate theories of altruism into the rational actor model, egoistic accounts of behavior have a limited ability to account for the decision to donate blood. Hence, insights into the nature of altruism, and the conditions under which it is more likely to emerge, will assist in understanding the variety of motivations for the blood donation decision. Theories of altruism are augmented by theories of social capital as possible explanations for the donation decision. Social capital has been found to be a powerful motivator for a variety of voluntary behaviors. There are some important findings from the literature on social capital. First, social capital is developmental that is, early belonging and volunteering predicts those same behaviors later in life. Second there are ties between social capital and trust, though the directionality of the relationship is far from clear. Third, social capital is a notoriously sticky concept to operationalize and more work needs to be conducted on this front. Finally, social capital has been in decline, though events like the September II th terrorist attacks generally result in temporary upswings. Finally, the analysis of the policy design literature demonstrates the inattention within policy design to the judgment and decision-making processes of policy targets, the importance of adopting a multidisciplinary perspective, and highlights the danger of assumptions embedded within policy designs. Il4

PAGE 128

This theoretical background, coupled with the historical analysis of blood donation, sets the stage for a series of surveys into blood donor motivations, presented in the next chapter. 115

PAGE 129

CHAPTER4 METHODOLOGICAL FOUNDATIONS Introduction Beginning this section, hypotheses and their rationales are presented. Derived from a multi-disciplinary literature review the hypotheses are designed to structure a cumulative light on the research question, "What is the structure of motives that drives the blood donation decision?" The hypotheses are followed by an analytic plan, including study design and statistical methods. The methodology is comprised of a two-part survey administered at two different time periods: before and after the September 11th bombings of Washington, D.C. and New York. The first component ofboth surveys is a policy-capturing exercise in which 125 subjects are asked to evaluate their likelihood of donating blood at 30 different hypothetical blood drives. The purpose of this component is to "capture" the subjects' judgmental policy for making blood donation decisions; and to cluster subjects by their decision making policy. 116

PAGE 130

The second component of the survey is comprised of mostly categorical questions on a range oftopics including, the subjects' donation history, rationales for and against donating blood, socio-demographics, and other voluntary behaviors in which they engage. These variables will be used to help describe the clusters formulated from the policy capturing exercise. The goal of this study design is to use a multi-disciplinary, multi-method perspective to improve understanding of the motivations for and against blood donation. The survey design permits analyses of subjects prior to the September 11th bombings, after the bombings, and a comparison between the two groups. Hypotheses Before beginning, a brief word on terminology is in order. In the following hypotheses, a variety of factors or conditions that may increase a person's likelihood of donating blood are discussed. However, given the nature of the study design described below, what will be measured is the "perceived likelihood to donate blood" rather than "likelihood to donate blood." There is strong evidence from previous research "that the intention to commit an act is the most direct predictor of performance" (Lee, Piliavin, and Call, 1999, 379). Lee, Piliavin and Call summarize: Two meta-analyses (Randall and Wolf, 1994; Sheppard, Hartwick and Warshaw, 1988), respectively involving samples of 87 and 98 studies, found 117

PAGE 131

average correlations of .53 and .45 between intentions across a wide range of behaviors. Randall and Wolff also found no significant effect of the time lag between intention and behavior, over a period ranging from a few hours to more than a year. In regard to donation behavior specifically Chang et al. (1988) and Piliavin and Callero (1991) have shown in longitudinal research that intention to give blood is an excellent predictor of future blood donation. In addition, the factors that predict intentions, generally also predict actual donation (1999, 379). Consequently, the behavioral intention to donate blood serves as a reasonable proxy for blood donation in this dissertation, recognizing, of course, that the correspondence will not be perfect. Individuals may have a tendency to overstate their willingness to engage in an act that is generally considered to be meritorious. Likewise, they may underestimate the influence of pressures (i.e., peer) that may propel them to donate. The hypotheses for the dissertation are divided into three main categories: altruism versus self-interest, incentives, and social capital. The rationale for each hypothesis is followed by a statement of the hypothesis. Altruism Versus Self-Interest Previous accounts of the blood donor decision presume the rational actor model ofhuman behavior (e g., Eckert and Wallace, 1985 ; von Schubert, 1994; Solow 1971; Stewart, 1984; Cooper and Culyer 1968). The assumption is that to increase donor turnout it is necessary to "incentivize" properly the potential donors. 118

PAGE 132

Prior to the 1970s, the standard incentive had been cash payment for donation. When the FDA required the labeling of blood as being either paid or unpaid, many blood centers turned to non-monetary incentives to encourage donation. Now, these types of non-monetary incentives (of the hat/mug/lottery variety) are very common.1 In a recent survey, nearly 80 percent of blood centers indicated that they provide some type of incentive to their donors (NBDRC, 1998). As some blood banks are still stymied by the persistently low donation rates, they are considering now a return to a paid system (Hernandez, 2002). This is particularly worrisome because, as mentioned previously, there are well-known links between payment for donation for blood and poor quality blood (Shilts, 1987) (e.g., blood with high rates of communicable diseases). Nonetheless, the persistently low donation rate is compelling blood centers to reconsider this problematic practice. It is interesting to note that the consideration of reintroducing payment for donation has even occurred since the September 111 h bombings (Hernandez, 2002). The hugely popular provision of incentives (be they pecuniary or nonpecuniary) overly simplifies the decision-making processes of potential donors. It assumes that individuals are donating primarily (if not solely) for their benefit and not for the benefit of others. It implies that blood centers must appeal primarily to 1 Technically, blood banks can label their blood as "voluntary" if the incentive given can not be easily exchanged for cash. This, of course, is a loose definition, and many questions are being raised as blood drives offer incentives that are more easily redeemable for cash, such as raffle tickets. 119

PAGE 133

an individual's self-interest to motivate donation; likewise, it denies that individuals can be motivated solely by altruism, i.e., a desire to help others. The assumption here is that the motives of individuals are more complex than current recruitment policies acknowledge; that is, an individual's decision to donate blood reflects a combination of selfish and altruistic motives. Specifically, the opportunity costs of the donation function similar to a gate (this concept is discussed in more detail below). Ifthe cost, in whatever manner perceived by individuals, is very high, then donation is less likely. The conception of opportunity costs presented here is of two basic types convenience and risk. The convenience element of a blood drive would include such factors as location, visibility, waiting time and other factors that would increase the time and effort ofblood donation. The second element of the opportunity cost of donation is perceived risk. While there is much research on cases where individuals undervalue risk (e.g., a skier not wearing a helmet), there is little research on cases of over-estimation of risk.2 3 We argue that blood donation is likely to be the latter case because previous research indicates that fear of contracting a contagious disease is common for potential donors as is fear of pain 2 This is an issue discussed by McClelland, Schulze and Hurd ( 1990) 3 See Aaron Wildavsky's Searching for Safety, (1991) for a particularly interesting discussion of the dangers of overestimating risk. 120

PAGE 134

and discomfort (Andaleeb and Basu, 1995). The argument is that perceptions of risk (as opposed to the actual level or risk) as well as the convenience costs thus affect the donation decisions. If donation is seen to be inconvenient or painful or risky, then the perceived benefit (to oneself or others) must be sufficiently higher than the costs or the donation will not occur. In some sense, then, donors may be functioning as intuitive (or unwitting) economists, roughly considering both the perceived costs and the benefits of donation before reaching a final decision. This does not imply any formal calculus on the part of the agent for the decision-making processes: that exercise will be left to colleagues more firmly rooted in the economic and cognitive sciences. However, for our purposes, the convenience element of a blood drive as well as the perceived risk element of donation can be considered the "cost" of a drive; and, donors, are likely to consider both this "cost" to themselves as well the "good" that can be generated by their action when determining if they will donate.4 Moreover, donors will be less likely to donate if the "cost" is exorbitantly high or if the benefit to others is perceived as being dismally low. This dissertation focuses on what happens in the decision making process when the cost is not exorbitant, that is, the blood center has basically done the job of 4 The high rate of blood donation after the September 11 '" tragedy is a prime example. The lines of people waiting to give blood were long and the wait was substantial (i.e., the "costs" were high). 121

PAGE 135

making donating as convenient as possible and minimizing the apparent risk. Then, what motivates individuals to donate blood? As mentioned previously, the opportunity cost of a blood drive can be conceptualized as being a function of the convenience cost (e.g. perceived convenience of the drive location, parking, information costs) and perceived risk of donation (e.g. of pain or acquiring AIDS or other communicable diseases on a Likert scale). The benefit of the drive can be conceptualized as how beneficial subjects perceive their action to be (on a Likert scale), both to themselves and to others. These various factors will be experimentally manipulated using policy-capturing. techniques, the details of which will be examined in much greater detail below. To test these concepts, two sets of hypotheses on altruism are presented. The first set is designed to test the hypotheses that the decision to donate blood is a function of mixed motives (altruistic and self-interested). The second set of hypotheses on altruism is designed to investigate the effect that incentives may or may not have the on an individual's willingness to donate blood. Based on these arguments, we can derive a 2x2 table of possible combinations relating the opportunity cost of the blood drive and the perceived benefits of the drive. The lettering of the box indicates the likelihood of donation with "4" being the highest and "1" being the lowest. This yields the following hypotheses. 122

PAGE 136

Table 4.1: Opportunity Cost Versus Benefit to Others in the Donation Context Low Opportunity High Opportunity Cost Cost Low Benefit to Others 3 1 High Benefit to Others 4 2 H1A: Individuals are more likely to donate blood if the perceived opportunity cost of blood donation is low (Cells 3 and 4). H18: Individuals are more likely to donate blood if the perceived benefit of blood donation is high (given HtA) (Cells 2 and 4). These two hypotheses conflict in the cases when both the benefits and costs are high and conversely when both are low. In such cases, we can hypothesize that individuals adjudicate their response based on the cost (convenience and risk) to themselves, when the benefit is low. In these cases, cost is functioning as a "gate." If the cost is perceived as being too high, individuals will not donate in any case. But, if the cost is negligible (or low), then individuals will donate even if the 123

PAGE 137

perceived benefits to others are also low.5 If the benefit is sufficiently high, cost does not function as a gate. A different set of motives may function in cases of highly salient emergencies (e.g., the September 11th bombings or the Oklahoma City bombing). When the benefit is very high, potential donors are less sensitive to cost. This is something that will be addressed again below in the section on social capital and in the results section as well. Hence: H1c: When the perceived benefit of blood donation is low, potential donors adjudicate based on cost (Cells 1 and 3). When the perceived benefit is high, potential donors do not adjudicate based on cost (Cells 2 and 4). Incentives This dissertation also tests the hypothesis that the provision of traditional incentives, consistent with the rational actor model of human behavior, does not adequately capture the complexity of motives .for blood donation. Instead, assuming a potential donor has cleared the convenience gate, an appeal to an individual's altruistic impulses (i.e., perceived benefits to others) is more likely to encourage a non-donor to donate and more likely to keep current donors donating as opposed to the non-donating status quo. 5 There is likely to be a baseline benefit under which individuals will not donate-even if the cost is very low. 124

PAGE 138

To test this hypothesis fully, multiple full-scale interventions would have to be conducted where, for instance, some donors would be told they were getting a mug, while others were told that a donation would be made to a charity (anonymously) on their behalf. Since this type of intervention is beyond the scope ofthis dissertation, we will rely here on individuals' stated willingness to donate.6 An appeal towards altruism for the policy-capturing portion of this study will be operationalized as an anonymous donation to a charitable organization. The donation will be $7, an amount presumed to be of value equivalent to the hat/mug variety of incentives that are so commonly used. 7 This will be in contrast to the self interested scenarios where an individual is offered the standard hat/mug variety of incentives or a lottery ticket for a CD-player for their donation. Consistent with this line of reasoning, the second altruism hypothesis is: H2A: An incentive structure that appeals to altruism alone will increase individual's perceived likelihood of donating blood more than an incentive structure that appeals to self-interest alone. 6 Based on a standard seven point Likert scale. 7 In pilot-testing of the survey, subjects were asked if they believe the shirts etc. are worth $7 125

PAGE 139

Classic theories of altruism suggest that individuals donate blood not for the benefits they personally receive, but because the donors recognize that others will benefit from their actions-even if, as donors, they might potentially be harmed (or at least inconvenienced). It seems obvious to many that donating blood is, in some sense, an altruistic act. However, there is little understanding of what factors are triggering the altruistic act of donation, or what factors might hinder the emergence of the donation behavior. In the next section, addressing "social capital," some potential triggers for altruism are considered including trust, organizational involvements and social connections-factors that are all intimately linked to the general concept of social capital. Social Capital As noted previously, Putnam and others claim to have observed appreciable declines in social capital (Putnam, 1995b, 2000 and Fukuyama, 1995a). Moreover, researchers have discovered that social capital has links to other types of voluntary and philanthropic behaviors (Putnam, 2000). We argue here that social capital effectively works as a lubricant for altruism. If true, the observed decline in social capital in the United States is partially responsible for the decrease in blood donations. As a result, individuals with greater proclivity towards social capital will be more likely to donate blood than individuals displaying less access to social capital. Specifically, in situations where individuals perceive large stocks of social 126

PAGE 140

capital, they are more likely to donate blood; conversely, where social capital is perceived to be low, individuals will be less likely to donate blood. The rationale is that when people feel connected to a group, recognize the potential for reciprocation, perceive the existence of social norms, and/or have a sense of belonging, they are more likely to engage in behavior that can be considered altruistic. There is disagreement about the most appropriate way to operationalize social capital. However, there are several important factors that multiple theorists argue are intimately connected to the concept. These inclucie: level of trust in others; feelings of community connectedness (a perception); and associational involvement (see Fukuyama, 1995 and Putnam, 1995b). Putnam (1995b) recognizing that the forms of social capital are "multiple and complex," laments that no individual measure is "perfect for testing the hypothesized decline in social connectedness, although the consistency across different measuring rods is striking" (1995b, 664). As such, he adopts multiple measure of social capital in his work (for instance, group membership, political participation, time spent volunteering) and encourages other social capital researchers to do the same (Putnam, 1996, 76). In this dissertation, we consider how trust, community connectedness, and associational involvement affect the blood donation decision and how they interact with the altruism motive. 127

PAGE 141

One remaining difficulty is that social capital is in some sense a community variable and the unit of analysis for this dissertation is the individual (see Putnam, 2000). Hence, instead of analyzing the level of social capital residing in different communities, we will instead assess the level of connection individuals have with group experiences, like organizations and communities and their perceptions of the amount of social capital to which they have access. 8 Trust. Trust is an interesting variable in and of itself and because of its connections with social capital (Putnam, 2000; Fukuyama, 1995). Researchers have found connections between an individual's trust of others and a variety of other behaviors, including charitable donations and organizational involvements (Newton, 1997). Given the variety of contexts in which trust has been found to be an important predictor of voluntary type behaviors, it seems reasonable that level of trust and blood donation may also be related. By far the most common question wording for the concept oftrust comes from the General Social Survey (GSS), in which subjects are asked how trustworthy they think other people are (In general, can others be trusted?). This measure has become widely used because of the questions' high correspondence with other measures of trust and because its' 8 Please see Chapter 3 for more discussion on this topic 128

PAGE 142

continued use allows comparison across studies. Hence, the GSS standard question on trust will also be used in this study. The specific hypothesis relating trust to blood donation is as follows : H3A: Individuals who reflect higher levels of trust are more likely to donate blood than individuals with lower levels of trust. Feelings of Community. Research has demonstrated that feelings of community are important in an individual's donation decision. For instance, Piliavin ( 1984, 4 7 4) presents community responsibility as a motivation for monetary donations. She also discusses how this feeling of community responsibility may be transformed into an internalized moral standard of being a responsible member of the community (1984; 1990). However, this account (and other similar accounts) presupposes that feelings of community exist. Individuals are unlikely to feel social pressure or feel community obligation if they do not perceive themselves to be (somehow) part of a community. The importance of feeling connected with a community has not been adequately explored in the literature. Putnam (2000) has noted that community connection is an important component of social capital. However, there is little work that assesses how feelings of community change as a result of a highly salient national crisis (such as the September II th bombings). I29

PAGE 143

Hence, feelings of community are the second aspect of social capital considered in this dissertation. Community connectedness is a difficult concept to define, let alone operationalize.9 However, here we are proposing a fairly straightforward approach. Subjects will be asked to assess (on a Likert scale) how connected they feel to their community. This is appropriate since it is their perception that is of interest (as opposed to a more network type measure of the number of community involvements).10 With this in mind, the following hypothesis will be tested: H3s: Individuals who consider themselves to be connected to their communities are more likely to donate blood than individuals who do not consider themselves to be connected to their communities. Associational Involvement. Another traditional way of operationalizing social capital is by consideration of associational involvements. The idea is that 9 Berry, Portney and Thomson (1993), building on the work of Cole (1974) develop a complex index for measuring community connectedness. Unfortunately, the questions from which the index was derived were rather extensive, relying in part on open-ended questions, and knowledge about each of the groups to which individuals belonged. The amount of information required for building the index made it impractical for this dissertation. 10 Self-report of community of associational involvements may be subject to bias as individuals may want to present themselves in the best possible light. We do not, however, anticipate any systematic bias. 130

PAGE 144

individuals who belong to (and are active in and committed to) voluntary organizations are likely to have access to greater stores of social capital. The general argument is similar to those described above for trust. Organizational involvement creates trust (i.e., norms ofbehavior, expectations of reciprocal behavior), nurtures a sense ofbelonging, and helps people identify with a "we." The basic argument is that organizational involvements are capturing an important aspect of social capital and are useful in predicting the decision to donate blood. This will be ascertained via a series of questions about the number of associational involvements the individuals have, and the amount of time subjects spend each week with each association. Hence, the following hypothesis will be investigated in this dissertation: H3c: Individuals with high levels of associational involvement are more likely to donate blood than individuals with low levels of associational involvement. Analytic Techniques A two-part study has been designed to address the above hypotheses. The goal is to use multiple analytic techniques with different populations to triangulate 131

PAGE 145

on an understanding of motivations for blood donation. Primarily, this study relies on analytic methods derived from the Judgment and Decision-Making (JDM) tradition of research.11 This section builds extensively on Cooksey's Judgment Analysis (I 996), as it is an exceptionally thorough guide to approaches and issues associated with the use of policy capturing. Some general background on JDM research techniques is provided to address issues of ecological validity and explore the appropriateness of using JDM techniques for the proposed study. Brunswick ( 1952), considered the father of modem judgment analysis, focused on understanding the relationship between an organism and its environment. His ideas were in reaction to what he considered to be the "wrong" path that psychology researchers were pursuing. His concern was that the prevailing methodology of the natural sciences would allow inappropriate generalization to other research subjects in similarly contrived laboratory settings; further, Brunswick wrote that prevailing methodology said "little about other conditions or contexts for the phenomena in question" (Cooksey, 1996, 1 ). Brunswick thought that psychology researchers were amassing much information that could not be generalized that is, results that would not be supported outside the specific experimental setting. Therefore, he introduced the 11 For purposes of this dissertation JDM and "judgment analysis" are interchangeable 132

PAGE 146

concept of representative design as opposed to representative sample. A representative design would focus on the sampling of situations or contexts as opposed to subjects His arguments were premised on two main ideas: first, there was uncertainty in the environment; and, second, there was uncertainty within the individual about how cue information should be used to guide decisions (Cooksey 1996, 3). These premises led Brunswick to suggest the importance ofthe representative design of experiments, and pioneer efforts "to study behavior under the naturally occurring entangled conditions in the ecology" (Cooksey, 1996, 4). Also, Brunswick argued that good research should first describe individual behavior, then generalize across individuals. This was termed the idiographic statistical approach (Cooksey, 1996, 7) Brunswick's ideas would prove very influential in the field that was to become JDM. Hammond was the first to apply Brunswick's ideas outside of Brunswick's field of perception, in particular to clinical judgments (Cooksey, 1996 8). Physicians were required to judge the IQ of patients based on cue information. Hammond's view was that clinical judgment could not be understood using conventional psychological methods and that the use of representative design was the only way to leave the entanglements among environmental variables intact for a proper study of clinical judgment under the conditions it naturally occurred in (Cooksey, 1996, 9). These early studies "set the stage" for a more widespread application of Brunswick's methods and concepts to human judgment issues (Cooksey 1996, 9). 133

PAGE 147

One of the most interesting research approaches to develop from this methodological framework was "policy capturing". Cooksey observed that "Bottenberg and Cristal ( 1961) are generally credited with the first use of the term policy capturing to refer to the use of multiple regression equations to idiographically analyze judgments" and that "[t]hey coupled this process with policy clustering which referred to the process of aggregating together judges having similar predictive policies" (Cooksey, 1996, 57, emphasis in original). The clustering of policies allowed the discernment of different typologies of judges; this would be advantageous if, for instance, we wanted to know if there were different groups of potential donors that made their decisions in similar fashion. Policy capturing requires subjects to review a series of scenarios and reach a judgment based on the information given. For instance, a panel might be asked to review student admission packages to a university and decide which students to enroll. The records would differ in important ways. For example, there might be systematic variation in the grades, test-scores, and socio-demographics of the applicants. Policy capturing offers the advantage that it allows researchers to "capture" the policy of the subjects and lessen the reliance on asking subjects what factors the subjects consider important in making their judgment. In the admissions example, analysis of the policy capturing exercise would yield information on which factors are being considered in admission decisions and which ways those cues are 134

PAGE 148

being combined. This is important because numerous researchers have found that subjects are neither terribly good at identifying which items are used for making a decision nor how items are combined (Hammond et al., 1986) Policy capturing allows the derivation of the decision-makers' policy from a series of judgments they were required to make. The term ''judgment analysis" was first used by Hoffman ( 1960) to describe what he termed the "paramorphic representation of clinical judgment" (Cooksey, 1996, 57). Hoffman required a panel of physicians to make a clinical judgment (diagnosis) based on the cues provided in hypothetical scenarios. He constructed a linear model where the cues provided to the physicians were the independent variables and the physician's response was the dependent variable. He recognized the linear model as an "as if' model ofthe decision made by the physician, and hence he termed it "paramorphic." It is important to realize that linear models are not designed to actually explain how people make decisions (see .Brehmer and Brehmer 1988) However, as Carroll and Johnson strongly argue, "one of the reasons that the linear models are such good predictors is that they can mimic the decision made by a wide variety of processes" ( 1990, 7 6). While the models do not mimic the actual processes humans 135

PAGE 149

use to make decisions, they produce analogous output (Carroll and Johnson, 1990, 67, emphasis in original). Policy capturing has many applications, ranging from weather forecasting to police judgments to clinical judgments. Policy capturing is particularly advantageous in situations where: ecological criteria are unavailable; when there is a problem of non-feasibility (for example, using policy capturing when considering alternate futures with which the subject has no experience); and when issues of confidentiality, ethics, legality and other potential bias are present. Cooksey ( 1996), however, is very critical about what he considers to be th_e proliferation of inappropriate uses. To this end, he developed a taxonomy for analyzing different situations where judgment analysis should and should not be used. Cooksey developed a two-by-two grid where one axis was the familiarity of judgment tasks (familiar versus unfamiliar) and the other axis was task congruence (abstract versus concrete, with high concreteness being most congruent). Cooksey concluded that in situations where there was both high task congruence and familiarity, policy capturing would be most representative (A). When neither high task congruence or task familiarity was present, then policy capturing was generally inappropriate according to Cooksey (D). In Cooksey's analysis of the other two cases (Band C) he suggested that policy capturing could be a useful analytic device. 136

PAGE 150

However, he suggested that there were specific design considerations that could be adopted to improve the reliability ofthe technique. These design considerations will be discussed in more detail below. Table 4.2: Broad Categorization of Judgment Analysis Research Contexts: Using the Dimensions of Task Familiarity and Task Congruence.12 Task Congruence Concrete Abstract Task Familiar -Judge has made these -Judge has made Familiarity sorts of judgments in these sorts of real life. judgments before in -Task information is real life represented and/or -Task information is obtained in original represented and/or units of measurement obtained using encountered in the abstract conceptual ecology. variables. A B Unfamiliar -Judge has seldom, if -Judge has seldom, if ever, made these sorts ever, made these of judgments before in sorts of judgments real life. before in real life. -Task information is -Task information is represented and/or represented and/or obtained in original obtained using units of measurement abstract conceptual encountered in the variables. ecology. c D The results from the policy-capturing portion of the survey will be analyzed using a weighted-additive model. This common JDM tool provides a powerful approach for understanding decisions. As Carroll and Johnson explain, weighted-12 Adapted from Cooksey 1996 (page 89) 137

PAGE 151

additive models "represent the basic structure of a decision program as the relationship between inputs (a set of alternatives described by attributes) and output (evaluation of alternatives made by the decision-maker)" ( 1990, 4 7). They argue "weighted-additive models are the most common and useful technology we have for predicting choices" (Carroll and Johnson, 1990, 67). The model can be represented simply as Yi=f (Xij), where Y is the overall evaluation (with i referring to the alternatives), and X are the attributes of the decision (with j multiple attributes). Finally, there is a substantial body ofliterature that indicates policy capturing, as a tool, reaches acceptable levels of external validity. This literature addresses two separate external validity questions. The first of these is whether the modeling component of judgnient analysis can accurately predict future decisions in similar settings. A large body of literature has either directly or indirectly investigated how models perform against subsequent decisions by the decisionmaker (in similar settings). The finding, from a variety of sources (see below) and with a substantial body of empirical support, is that policy-capturing methods can capture policies of decision-makers in a way that can predict future decisions. As Schmitt et al. (1991, 93) observe: This method of studying decision behavior (i.e., asking participants to indicate their intended decision) is supported by previous research showing a close correspondence between actual behavior and behavior predicted on the basis of a subject's decision-making strategy (Dawes & Carrigan, 1974; Eraker & Polister, 1982; Hammond, 1977; Hammond, McClelland & Mumpower, 1980; Wigton, Hoellerich, & Patril, 1986). 138

PAGE 152

The second external validity question is whether responses to the hypothetical scenarios in any way reflect the decisions that individuals would make outside of the controlled environment of the experiment. Once again, there has been some research that addresses this question.13 This research, mostly from the medical field, finds that policy-capturing tasks, in fact, have a fair amount of correspondence with similar judgments made outside of the experimental setting. While there is concern that only supportive studies that verify the utility of policy capturing have been published, it seems unlikely, since debunking a respected analytic tool is probably as publishable (if not more so) as work supporting the appropriateness of its application. In summary, policy capturing has been used in a variety of contexts for over 30 years. Research on the external validity of policy capturing suggests that policy capturing is able to represent the types of decisions that subjects make outside of the survey setting. The Surveys To improve our understanding of the reasons why some individuals donate and others do not, a two-part study design will be employed. A two-part survey will 13 See Cooksey (1996, 315) for a review of research addressing the correspondence between decisions in hypothetical scenarios and decisions outside of the experimental setting. 139

PAGE 153

be administered at two different time periods. The original plan for this research endeavor was to administer two different surveys to two distinct populations (a sample of undergraduates; and then a group of donors at a blood drive, and a group of non-donors). After the September 11th bombing, the study design was altered to take advantage of the unique opportunity to investigate the impact of a national crisis on the structure of decision making of potential blood donors. Hence, the first survey was administered several months prior to the September 1 I th bombings; a second survey was then administered about one month after the bombings. An additional reason for two surveys is that consistent results across the two studies using different populations should increase our confidence that the results are not spunous. The first portion of the survey will be a policy-capturing exercise. Subjects will review descriptions of30 different hypothetical blood drives. For each blood drive, they will be asked to indicate, on a Likert scale, how likely they believe they will be to donate blood. This will allow the discernment of the policy by which subjects reach their decision about blood donation. The second portion of both surveys will have a more traditional style with questions regarding the respondents' motivations for and against donation, familiarity with blood donation, socio demographics, and other factors to be elaborated below. 140

PAGE 154

The in-depth questionnaires will be administered to undergraduates at a four year university. The rationale for the chosen study population as well as the proposed study design follows. The Study Populations College students have been chosen as the subjects for these studies because they offer some distinct advantages. First, they are the next generation of donors and a segment of the population that the blood donation community is very interested in understanding (and of course engaging). Piliavin ( 1991) has noted that the younger generation is much less likely to either donate blood or to engage in a wide range of other voluntary behaviors than their parents. These observations coincide with Putnam's observations regarding the generational shifts in voluntary behavior (2000). Many in the blood banking industry, justifiably, find this low level of participation particularly troubling. Second, the proposed study design requires an amount of thought and concentration (the survey is expected to take at least 30 minutes) that exceeds what can be expected from a mail survey or from most survey populations. Verification of these results with other populations (potentially a general representative sample) will be the logical next step, but it is beyond the scope of this dissertation. 141

PAGE 155

Section I: Policy Capturing The first portion of the survey employs policy-capturing techniques. Subjects will be presented with a series ofhypothetical blood drives in which they are asked to volunteer to donate blood. In each scenario, a portion of the message soliciting donation will be experimentally manipulated.14 For instance, the subjects may be told that as an incentive either an anonymous donation to a charity will be made on their behalf or they will receive a commemorative t-shirt. Consistent with the within-subject design, each subject will be required to review multiple scenarios. I The purpose of this technique is to "capture" the decision-making rule that the individuals are using, rather than relying on self-report of their decision-making rule Unfortunately, this procedure does not address an important limitation of any survey assessing motives -stated or inferred motivation may not correspond to actual behavior. The only way to address this limitation would be to conduct a fullscale intervention in which the incentives offered at blood drives and the message soliciting donations at blood drives were experimentally manipulated. However, such an intervention is beyond the scope of this dissertation Task Familiarity. Before considering the specific design elements of the judgment analysis, it is necessary to determine where in Cooksey's taxonomy of task 14 A s discussed earlier, willingness to donate is a reasonable pro x y for actual acts of donation 142

PAGE 156

familiarity and task congruence the proposed study falls (see Table 3.2). Cooksey (1996, 99) provides some guidance for determining how familiar the task would be. He recommends considering judges to be one of three types. 1. Familiar and experienced 11. Familiar but inexperienced iii. Unfamiliar and inexperienced. The subjects are college students. It is assumed that most of them will have encountered a high school blood drive and very likely a university blood drive. Hence, they are very likely to be familiar with the specific judgment domain and experienced at making this type of judgment. Cooksey defines experience as "knowing what such judgments involve" and having actually made similar judgments in the past (Cooksey, 1996, 99). While the subjects (college students) may not have consciously considered their past decisions in any depth, at least they are likely to have encountered a blood drive and have consequently either donated or not donated. Hence, the subjects appear to be both familiar and experienced with the issues surrounding blood donation, implying that task familiarity for this survey will be high. To verify the task familiarity of the students, a series of questions will be included in the surveys. Questions will include: "Did your high school have a blood 143

PAGE 157

drive?", "Do you remember specifically making a decision about whether or not you would donate blood?", "Do you remember the decision?" and "Have you noticed other college blood drives?". Subjects will be stratified by their responses to these questions. At the initial stage of analysis ofthe results, the stratification's distribution allows an evaluation of the appropriateness of the use of policy capturing for this population. Task Congruence. Cooksey explains that task congruence: reflects the extent to which the judgment task utilizes either abstract representations of cue information and judgment assessments which would not ordinarily be encountered in the judgment ecology or more correctly measured representations of cue values and judgment assessments using the measurement units with which they would naturally observed within the ecology." (Cooksey, 1996, 88, emphasis in original). The task congruence employed in the survey will be abstract since information will be presented in ways that are novel as there are no natural units for some of the themes to be explored (e.g., incentives). Therefore, while the subjects are likely to be somewhat familiar with the cue information (e.g., t-shirt versus a donation to a charity); they may not have seen the cue information specifically identified in the survey format. Cooksey makes several recommendations for cases in which abstract concepts are being considered ( 1996). One recommendation is to tum all cues into a series of anchored scales (Likert style). A second recommendation is to use fewer 144

PAGE 158

than nine cues since previous research shows that subjects generally do not use many cues when making decisions (Brehmer and Brehmer, 1998). Finally, Cooksey recommends using information from real cases. If this is not possible, he recommends that simulated profiles with either orthogonal or near orthogonal design should be used (Cooksey, 1996, 101-106). In review of Table 4.2, this study falls into Cell B of Cooksey's taxonomy for the appropriateness of policy analysis. The task familiarity is presumed to be high (this will be verified), but the congruence is not, given the abstract nature of some of the cues. Hence, policy capturing can be considered an appropriate methodological tool for this dissertation. Refinement of Cues and Determination of Cue Quantity. Before proceeding, two final questions regarding the construction of the policy-capturing instrument need to be answered: "How should the cues be generated?" and "How should they be presented?" Cooksey noted that identification and selection of the cues were often the most subjective portion of the analysis (Cooksey, 1996, 107). Therefore, he recommended conducting pilot surveys or informal interviews to assist in the generation and streamlining of the cue list. In addition, researchers need to make some distinct tradeoffs when deciding how to query individuals about which cues should be considered. If open-ended querying techniques were used, then the focus should be on specific problems of interest. In doing so, he held that one should 145

PAGE 159

obtain consistent results. However, the results could be driven by the researcher's presuppositions, which might not be apparent. Cooksey's recommendation was to allow some open-ended options. This meant that the researcher should identify some cues she considered to be important, and then use informal interviews and pilot tests to finalize the list. This approach is adopted for this study and outlined next. A selection of cues was identified from previous research and derived from theories presented in the literature review chapter. Informal interviewing was then conducted with undergraduates to improve the cue selection. In addition, the entire survey instrument was pilot-tested by five compensated undergraduates. Initially, subjects were asked to complete the survey. Then they were instructed to go through the survey, question-by-question, to identify issues that were ambiguous, confusing or unclear. Also, they were asked to consider carefully the cue items presented and to suggest additional cues they thought would be important in their blood donation decision. Their feedback was used to alter the survey instrument prior to the full study. Upon completion of a review of the literature, informal interviews and pilot testing, a policy-capturing instrument was finalized, consisting of six main cues, corresponding to the dimensions of: location, need/salience, good-deed, donation to charity, lottery for a CD player, and the sponsor of the blood drive. (Please see 146

PAGE 160

Appendix A for the exact wording of the cues.) A near orthogonal design was used for the construction of the cases. 15 The final design decision for this portion of the analysis was the determination of how many cases each subject should view. Cooksey recommends using five to ten cases for each cue per subject.16 Hence, assuming five cues, then 25 to 50 paper cases should be included. The lower range is appropriate when intercorrelations between the cues are not expected. Hence, 30 cases were presented to each subject. Sample Size. The procedure used for determining the sample size for this type of analysis differs slightly from the techniques used in most nomothetic research 17 because "[ t ]he power of idiographic judgment analysis lies in its emphasis upon judgments of many cases by each judge, not in the number of judges" (Cooksey, 1996, 133). Cooksey continues: A properly designed and executed Judgment Analysis study, where judges are able to execute their policies with reasonable consistency, will therefore yield data for nomothetic comparisons (e.g. R2s or regression beta weights) which are of much higher quality than is traditionally available for such tests, 15 This means, "all of the correlations among cues are unifonnly forced to be zero" (Cooksey, 1996, 371) 16 This recommendation is based on research conducted by Tabachnick and Fidell ( 1989) and Hair, Anderson, Tatham, and Black (1992) Note that some unvalidated research (Cook, 1976) indicates that a 5 to 1 ratio is more than sufficienthence the lower end of the 5 to 10 range will be considered sufficient for this dissertation. 17 Nomothetic research is research where average responses are the outcome of interest, whereas idiographic research focuses on individuals within a particular environment. 147

PAGE 161

precisely because the Judgment Analysis measures are based upon many observations (Cooksey, 1996, 133). Since the subjects are not only considered independently, but also in aggregate, attention must be paid to issues of sample size. Both Howell (1992) and Cohen (1988) provide information on how to determine the number of subjects necessary for various statistical tests for the results of judgment analysis. Cooksey summarizes their work by stating that with a properly conducted judgment analysis, where clustering techniques will not be used the sample size can generally be 50 subjects (assuming a large effect size). If clustering techniques are to be used (as they will be in this dissertation, see below) with the same assumptions about effect sizes, then approximately 100 subjects should be sufficient for most purposes. The estimate of 100 subjects assumes that there will be between three to six clusters. This size sample provides some margin since it is undesirable to conduct analysis on clusters of vastly uneven sizes and on clusters with fewer than ten members. To be even more conservative, we will use 125 subjects for each survey Section 2: A Query into Motivations for Blood Donation The survey's final portion is designed to query individuals about their motivations for donation. It will consist of mostly close-ended questions.18 A sample survey can be found in Appendix A and will be described in detail below. 18 Several open-ended responses are also included. 148

PAGE 162

For the primary hypotheses, the dependent variable will be a simple dichotomous response to the question "Has the individual donated blood in the past?" Previous research has demonstrated that the motivations for single time and multiple time donors often differ substantially (Piliavin, 1989). Consequently, the number of times that an individual has donated is collected also. Note an important difference between Section 1 (Policy Capturing) of the surveys and Section 2 (Query into Motivations). In Section 1, the dependent variable will be the median response to the Likert scale of an individual's stated willingness to donate blood, 19 whereas in Section 2, the dependent variable will be a dichotomous response indicating whether the individual has or has not donated blood previously. In Section 2, subjects will be asked to review a series of statements that will reflect reasons for and against donating blood (e.g., "I want to help others."). They will be asked to use a seven-point Likert scale, ranging from "not very important" to "very important," to evaluate how important the statements are in their decision to donate or not donate blood. In addition to the statements provided, subjects will be encouraged to include additional items as they see fit.20 The inclusion ofboth encouragers and inhibitors to the blood donation statements reflects the finding that subjects often respond differently to positively and negatively worded concepts. 19 Based on the distribution of the seven-point Likert scale responses, it may be necessary to dichotomize the variable at a threshold or transform the variable. 149

PAGE 163

The second component of this portion of the survey will consist of a series of questions, generally of the categorical nature, regarding the subjects' socio economic status and blood donation history. These two categories of variables, as well as the specific variables used to operationalize the hypotheses are introduced in more detail below. Socio-demographics. The literature on philanthropic behavior reveals that measures of socio-economic status correlate positively with actions such as volunteering and donating money (Piliavin and Callero, 1991). Since our subjects. are college students, this will not probably apply. Information on students' income level is not likely to be very informative about their actual socio-economic status, as most students receive forms of financial assistance (e.g., loans, support from parents, scholarships). For survey simplicity, questions will be asked about the subjects' parents' income level (categorical variable) as well as their parents' education level (categorical variable). Also, the students' educational attainment will be queried to ascertain their year in college and scholastic major (categorical variables). Basic descriptors such as gender, race, ethnicity, and religion will be included, and will be categorical. Age (year born) will also be included in the socio demographic portion of the survey. 20 Note that the items will be generated from the literature review. 150

PAGE 164

Donation History Subjects will be asked questions regarding their familiarity with blood donation. They will be surveyed as to how often, if ever, .they have donated blood and if family members or friends have. In addition, there will be questions about the prior transfusion history of their family and friends as well as whether they have ever required a transfusion. These variables will serve as controls as they have been found to be important by previous researchers (e.g., Piliavin and Callero, 1991 ). Another valuable piece of data is the subject's history of blood donation eligibility. It will be assessed because Piliavin (1987) has found that once labeled ineligible, that label can have a long-term effect on an individual's willingness to attempt future blood donations. Hence, information about donator eligibility will be collected as a control variable. Hypothesis Related Variables The variables created to test the hypotheses proposed in this study are of two types those used in the policy-capturing portion and those used in the traditional portion of the survey. In the interest of clarity, an explanation will be presented below that describes the specific manner in which each hypothesis (category) will be operationalized. Altruism Versus Self-interest Hypotheses. The first main category of hypotheses (HIA, His and Hie), is designed to query subjects about their motives for 151

PAGE 165

donations and calls for the operationalization of the perceived costs and benefits of donation. The perceived benefits of donation will be investigated in both the policy capturing and traditional portion of the survey. In the policy-capturing section, a series of statements slightly alters the reasons donors are sought for blood donation. Subjects, for instance, will be told that the reason for the blood drive is the occurrence of a local natural disaster or subjects might read a more generic statement about the continued need for blood nationwide. In the main portion of the survey, the perceived benefit of the blood drive will be investigated via questions regarding perceptions of the need for blood. Subjects will record their responses on a Likert scale. In addition, the benefit of donation is presumed to vary between the two surveys as a result of the September 11th bombings. Hence, the dichotomous variable indicating survey will also be used in the comparison of benefit. The perceived cost of donation is a function of the risk of donation and its convenience cost. The risk of donation will be assessed only in the traditional portion of the survey through a series of questions about anticipated pain or discomfort, fear of needles and fear of contagion as a result of donation. The risk questions will not be included in the policy-capturing portion of the survey, because it is assumed that perceptions of risk are constant for the individual and would not vary for different (hypothetical) blood drives. 152

PAGE 166

To capture information on the convenience cost ofblood donation, subjects will be asked to respond to a series of questions. Questions will include such items as the last time they saw a blood drive; if the location and time were convenient; or if they noticed it advertised prior to the blood drive itself. Also, subjects will be asked if they remember a high school blood drive and if they have been aware of previous university blood drives. Additional questions regarding the convenience cost will be included in the policy-capturing portion of the survey. Respondents will be told alternatively that the blood drive is on the way to class, requires a walk across campus or requires a 15-minute bus ride. Incentive Hypothesis. The next hypothesis (H2A) is designed to adjudicate between altruism and self-interest as motivators for donation. The main vehicle for doing so will be the offering of various incentives for donation. The hypothesis regarding incentives (H2A) will be assessed in both the policy capturing and traditional surveys. In the policy-capturing section, subjects will be presented with statements about blood drives where there will be an explicit incentive for donation such as entering a drawing, or receiving a t-shirt or an anonymous benefit to an charity (of equivalent monetary value to the other incentives) In addition some subjects will be reminded of the meritorious nature of donation, and other subjects will be reminded of the benefit to themselves (the receipt of the t-shirt). 153

PAGE 167

During the traditional portion of the survey, subjects will respond to other questions related to motive. For instance, while using a Likert scale, they will assess how important helping others was in their decision to donate. All these questions will help us determine whether altruism self-interest, or a combination of the two were the driving force behind the subjects' blood donation. Social Capital Hypotheses. The final set ofhypotheses (H 3A, H38, and H3c) is designed to investigate the relationship between social capital and donation. In the traditional portion of the survey, social capital variables will include questions on level oftrust, organizational involvement, and social connections. Specifically, for trust, subjects will be asked if, in general, they believe that others can be trusted (the standard GSS question wording). To assess organizational involvements, subjects will be asked to list the organizations they are regularly involved in and to indicate how many hours per week they spend with these organizations. In addition, subjects will be asked to indicate whether they have donated money, food, clothes and/or time in the last year. Also, the subject's involvement in student government will be assessed as previous research has found this to be positively correlated with voluntary behavior (Y ounnis, McLelland and Yates, 1997). The last set of social capital questions is about social connections. Subjects will be asked to indicate how "connected" they feel to their community using a 154

PAGE 168

Likert scale. In the policy-capturing portion of the survey, respondents will be presented with scenarios in which they are told that an organization of which they are a member is running a blood drive, that their university is running the blood drive, or that the local blood bank is conducting the drive. The assumption is that the closer the institutional proximity of the connection with the blood drive, the greater the likelihood of donation. Analytic Plan: Portion I, Policy-Capturing Exercise The analytic plan in this dissertation will build on the work of Schmitt et al. ( 1991) and Cooksey ( 1996). Schmitt et al., in their study of women's judgment policies for hormone replacement therapy, proposed a multi-stage analytic plan to both capture and cluster individual's decision making strategies. Specifically, Schmitt et al. offered five successive steps: descriptive statistics, policy capturing, clustering, discriminant analysis and a final regression analysis. Cooksey ( 1996) recommended an additional step that should be conducted prior to the analysis of any policy capturing data, namely, looking at the responses for each subject individually. These analytic steps will altered to meet the needs of this dissertation and are described below. First, standard descriptive statistics will be computed. Categorical variables will be tested using a chi-square test or Fisher's exact test, where appropriate. Normally distributed continuous variables will be analyzed using t-tests; non-155

PAGE 169

normally distributed variables (e.g., age) will be analyzed using the Wilcoxin ranksum test. Second, the subjects' estimate of their likeliness to donate blood will be regressed on the explanatory variables described above as well as some control variables for the subjects' socio-demographic and blood donation history. Missing data will be imputed using the median respons'e for that subject. This will be the policy-capturing portion of the analysis. Third, a factor analysis will be conducted on the beta-weights from the regression analyses described above (Sclunitt et al., 1991 ). The goal of the factor analysis is to examine the structure of the relationship among the variables (Norman and Steiner, 2000). Separate factor analyses will be conducted for both surveys independently, and then for the two surveys combined. Between-group analyses using the Wilcoxin Rank Sum test will be conducted to determine whether similar decision making policies were used by both survey populations. This analysis will help determine if the surveys can be pooled for all subsequent analyses, or if they need to be considered independently. The fourth step of the analysis plan is the grouping (or clustering) of subjects. This will be done by assigning each subject to the factor on which they loaded most strongly. Differences in the demographic, blood history or motivations for donation of the clusters will then be compared. 156

PAGE 170

One additional step will include the use of cluster analysis on the beta weights derived from the regression analyses described above. Cluster analysis is a technique where judges are clustered by their similarity. Cluster analysis provides convenient groupings of the respondents, but does not result in a predictive model, as factor analysis does. The use of cluster analysis can be seen as an alternative to factor analysis, and the results from the two methods will be compared. In the end, this analytic structure will allow the discernment of the structure of decision-making at two distinct time periods; before the bombing, after the bombing and a comparison of the structure between the two time periods Analytic Plan: Portion 2, Core Survey For the second portion ofboth surveys, descriptive statistics will be followed by bivariate analysis (as described above). Categorical variables will be tested using a chi-square test or Fisher's exact, where appropriate. Normally distributed continuous variables will be analyzed using t-tests; non-normally distributed variables will be analyzed using the Wilcoxin rank-sum test. Results for the survey will be computed using SAS version 6.8. Validity and Reliability Upon completion of the analytic plan described above, steps will be taken to verify the internal validity, external validity and reliability of the survey. Internal 157

PAGE 171

validity will be assessed through a comparison of responses between the two surveys --a technique termed known group validity (Cooksey, 1996, 313). Cooksey explains that "a measurement process may be considered valid if it differentiates between members of different known groups in particular ways which can be specified a priori" (Cooskey, 1996, 313, emphasis added). In this dissertation, we can anticipate that subjects from Survey I would have lower median likelihood of donating blood than would subjects from Survey II. An alternate measure of internal validity relates to the appropriateness of using policy capturing methods. Cooksey identified task congruence (complexity) and task familiarity as two crucial factors for determining if policy capturing is an appropriate research tool for a given population. Task congruence (see page 141) for this dissertation is low (meaning that the task is not represented in natural units similar to those outside of the decision setting). Task familiarity will be assessed through analysis of a series of questions regarding subjects' awareness of and familiarity with blood drives. There are three separate external validity questions relevant to studies that employ policy capturing. Traditionally, external validity indicates how well our research holds up outside the experimental setting -usually this reflects whether results from a sample are generalizable to the larger population. There are additional issues of external validity that arise when within-individual (idiographic) 158

PAGE 172

analysis has been conducted. These relate to how generalizable the findings are to decis!ons made by the subject, outside of the experimental setting. Hence, the first external validity question addresses whether the modeling component of policy capturing accurately reflects future decisions (this is a question about the use of the linear model in policy capturing). The second question addresses whether decisions made in the experimental setting reflect those that would be made in a natural environment. These two questions, as mentioned previously, have been assessed in previous research, and the general finding has been that the linear model is a good predictor of future decisions, and that decisions made in policy capturing exercises do reflect real world decision making. In this dissertation, as in most policy capturing research, it is nearly impossible to adequately assess external validity (Cooksey, 1996). We have assumed that willingness to donate blood is a reliable predictor of future behavior. Previous research lends support to this assumption (Lee, Piliavin, and Call, 1999, 379). We would expect some correspondence between willingness to donate and previous donation, although they are different constructs. Individuals who have not donated previously, may decide to donate in the future. Likewise, previous donors may not donate again, for reasons of ineligibility or for other reasons (e.g., discomfort). Hence, these first two external validity questions can not be assessed in this dissertation. 159

PAGE 173

The third external validity question addresses whether these results can be generalized to a broader population. The subjects for this study are undergraduates at a prestigious four-year university in southern California and are in no way a representative sample. However, it should be mentioned that purpose of this study was not to gamer a representative sample, but to come to know, in a meaningful way, the decision processes of a group of potential donors. Moreover, the subjects of this study are members of the next generation of donors, and hence their decision making policies are of great interest. In sum, previous research indicates that external validity is very difficult to ascertain without follow-up outside of the experimental setting. Hence, we will not be able to assess external validity of the surveys used in this dissertation. Several methods will be used in this dissertation to ascertain the reliability of responses. A few duplicate items were included in the survey, and their correlations will be determined. In addition a series of internal consistency checks will be employed (for instance, if subjects indicated that they donated at their high school blood drive, did they indicate that they had donated previously?). Human Subjects Approval The study received Human Subjects Review Board (HSRB) approval under the auspices of the University of California at San Diego, and seconded by the 160

PAGE 174

Human Research Committee at the University of Colorado at Denver. Subjects will be provided with copies of the consent form after they sign a copy. Also, in accordance with the HSRB policies, subjects will receive information about whom they will be able to contact in case of harm caused by participation in this study. Conclusion One survey, consisting of two components each, will be administered at two different time periods. The first portion of both surveys will use policy-capturing techniques. These techniques will be used to explore the decision policies for a group of potential donors both before and after the September 11th bombings. The second portion of the survey will be designed to allow a comparison of motivational differences between the groups. This bifurcated design is intended to fill crucial gaps in our current understanding ofblood donation practices. The results from these studies are presented in the next chapter. 161

PAGE 175

CHAPTERS RESULTS Introduction A multi-component survey was administered to two distinct groups of undergraduates, before and after the September 11th bombings. The purpose of this design is to explore the factors that affect the blood donation decision and to access the decision-making structure of potential donors. Moreover, with the second survey having been administered after the September 11th bombings it is possible, to some extent, to explore the impact of the bombings on blood donation decisions. The chapter begins with basic descriptive information about the populations from both surveys. Next socio-demographic characteristics are compared across the surveys to assess the basic comparability of the two populations. This is a necessary step that will permit further between-group comparisons. Third, bivariate analyses of predictor and control variables by both "previous donation" and "willingness to donate" are presented. Fourth, the results of the policy capturing 162

PAGE 176

exercise are explored. Several distinct populations of decision makers are identified from the factor analysis. In the next section, the socio-demographic characteristics, familiarity with blood donation and motivations for blood donation of these various behavioral clusters are compared. This is followed by the presentation of a cluster analysis, as an alternative to clustering via the factor analysis. Next, the results of a betweengroup-analysis, before and after September 11th, are presented. A more thoroughgoing analysis of the implications of these results, as well as an evaluation of each of the study hypotheses, can be found in the final chapter of this dissertation. In all subsequent analyses, significance levels are set at .05.1 Results for non-donors are always presented before the results for donors in the following format (non-donors, donors, P-value =X). The same format is followed for analysis of "unlikely" and "likely" donors, as well as Survey I and Survey II respondents. The Study Populations The first survey was administered to 125 undergraduates approximately three months prior to the September 11th bombings. The second survey was administered to a group of 127 undergraduates approximately one month after the September bombings If the same respondents had been surveyed at the two 1 Norman and Steiner (2000) note "5% is only an agreed upon convention and not some absolu t e criterion of truth." Nonetheless, a cut-point is required, and hence the indu s try standard of .05 was adopted. 163

PAGE 177

different time periods, we could have been more confident that behavioral and outcome differences were a result of reactions to the bombings, and not inherent differences in the population. In the present inquiry, the populations were unique; hence a comparison of the two populations to identify a priori differences was necessary. To this end, a comparison of the two populations is presented below. Comparability of Survey I and Survey II Populations Basic socio-demographics, previous blood donation rates and familiarity with blood donation were compared between the surveys to determine if there were significant differences between the groups that would impede comparability (see Table 5.1 ). Socio-demographically, the two populations were very similar with respect to sex, parental income, parental education and race. The two populations did differ, however, with respect to age. The median age of respondents to the first survey was 21, while the median age of respondents to the second survey was 20 (P = .000 I). This difference in age was captured also by median years in college being somewhat higher in Survey I than in Survey II (median, 3 vs. 2, P = .0005). It should be noted that this difference might have been partially explained by the first survey being administered at the end of one academic year, while the second survey was administered at the beginning of the following academic year. 164

PAGE 178

Table 5 1 A Comparison of Survey I and Survey II ( selected variables) Survey I Survey II P-value Socio-Demographics Female gender, % Age Race2 White, not Hispanic Hispanic Asian/ Pacific Islander Other Years in college Donation History Donate previously, % Likelihood of donating, % Previously ineligible, % Currently ineligible, % Age at first donation High school blood drive Motivators Feelings of guilt T -shirt incentive Test for STDs Need for blood Want to help others Inhibitors Convenience Fear of needles Amount of time Fear disease Fear of getting AIDS Too many questions 55.74 21 (20, 22) 39.34 10.66 41.8 8.2 3 45.6 35.09 21.6 76.8 17 (16, 18) 71.9 2(1,4) 3(1,3) 2 (1' 5) 5 (4, 6) 5 (4, 6) 5 (3, 5) 3 (1' 5) 5 (3, 6) 5 (4, 5) 5 (4, 5) 1.75(1,3) Median (1st quartile, 3rd quartile) 65.6 20(19,21) 39.2 8 46.4 6.4 2 40. 8 54.05 34.68 73.39 17 (17, 18) 86.29 2 (1' 3) 2 (1, 4) 2(1,4) 6 (5, 6) 6 (5, 6) 4 (3, 5) 3 (1' 5) 4 (3, 5) 4 (3, 5) 5 (3, 5) 2 (1 t 3) 2 Races with low N's were bundled into the "other" category. 165 0.1126 0.0001 0.9209 0 .0005 0.4436 0.0042 0.0217 0.5335 0.054 0.0006 0 884 0.1995 0.0338 0.0001 0.0002 0.1546 0.7736 0.0012 0.1741 0.7441 0.9658

PAGE 179

Importantly, there was not a significant difference between the two groups in the rate of previous blood donation. However, the median responses to the 30 iterations of the Likert scale on "likeliness to donate blood" was significantly higher for Survey li respondents than it was for Survey I respondents (median, 5, vs. 6, P= .0007).3 There was no significant difference in other types of voluntary donations (e.g., money, time, clothes and food). More respondents in the second survey reported that their high school had a blood drive (71.90 % vs. 86.29%, P = .006), although the difference in the rate of high school blood donation was not significant. Respondents to the second survey were more likely to have been previously ineligible to give blood (21.36% vs. 36.48%, P = .0217). Although there was no difference in the number of blood drives reported by the two populations, there was a difference in the location where they saw blood drives. Survey I respondents were more likely to have seen a blood drive near school (88.52% vs. 69.67%, P = .0003), while Survey II respondents were more likely to have seen a blood drive near home (6.56% vs. 16.39%, P = .0159).4 This again, might have been an artifact of the younger age of the second population. 3 This finding will receive greater attention later in Chapter 6. 4 Survey II respondents were also more likely to report that they saw a blood drive at another location (7.38% vs. 22.95%, P = .0007), however, the write-in responses, for "other" locations where blood drives have been seen, show no discernable pattern 166

PAGE 180

Of the factors that either encourage or discourage donation, two differentiated the populations. Helping others was a more important motivator for Survey II respondents than for Survey I respondents (median, 5 vs. 6, P = .0002). Survey II respondents reported also that knowing that there was a need for blood was a larger motivator in their blood donation decision than did Survey I respondents (median, 5 vs. 6, P = .001). The greater importance ofhelping and understanding the need for blood in the second survey could have been related to the second group's recent exposure to the September 11th bombings. This notion will be explored later in the chapter. In summary, despite differences in age, the two survey populations seem socio-demographically comparable. The similarity provided us with sufficient confidence to continue with between-group analyses. There were, however, some notable differences in stated motivators for donation between Survey I and Survey II respondents, with Survey II respondents being more concerned about a desire to help as well as a greater sensitivity to the need for blood. Likewise, it is notable that Survey respondents indicated that they were more likely to donate blood. Though it was probable that this difference in motivators and willingness to donate was related to the September 11th bombings, we could not determine what portion of the differences between the two populations was related to the timing of survey administration relative to the bombings, as opposed to unexplained variation 167

PAGE 181

between the two groups. Between-group-analyses, presented at the end of this chapter, sheds more light on this question. Survey-by-Survey Analysis Survey responses were analyzed bivariately against two indicators ofblood donation. The first indicator was previous donation (coded 0, 1) as reported by the respondents. The second indicator was median response to the questions on the policy capturing exercise that asked subjects, on a seven-point Likert scale, how likely they were to donate blood at the 30 hypothetical blood drives. Based on the distribution ofthe median response (see Table 5.2), the variable was dichotomized at six and above as an indicator of respondents who were either "likely" or "unlikely" to donate blood. This conservative cut-point seemed reasonable, as willingness to donate blood was positively skewed. Both surveys were analyzed independently using the two measures of blood donation. The results will be presented below. 168

PAGE 182

J!J 70 CJ 60 C1) -50 .c ::::s U) 40 It-0 30 ... C1) 20 .c E 10 ::::s z 0 Figure 5.1: Distribution of Median Likelihood of Donation t---------------------------------------------1 2 3 4 5 6 7 Median Likelihood of Donation Results from Survey I: by Previous Donation Below, Survey I respondents were analyzed by their history of previous blood donation. The results from this analysis are summarized in Appendix C. Of Survey I respondents, 36 percent indicated that they donated blood previously. The median response on the policy capturing exercise was five for both previous donors and non-donors. Once this variable was dichotomized at six and above 35.48 percent of Survey I respondents were classified as likely donors. Part of the discrepancy between these variables can be better understood by further probing of the individuals who donated previously but were not highly likely to donate based on the dichotomization of their median response. Of the 30 previous donors, who were unlikely, 26 believed that they were currently ineligible to donate 169

PAGE 183

blood. Once they were removed from the analysis, the correspondence between the variables was, logically, much stronger. Survey I donors and non-donors seemed similar in many areas. They did not differ significantly with respect to age, gender, race, family income, parental education or year in school. Interestingly, there were no significant differences between the two populations on the factors that respondents had reported motivated (either encouraged or discouraged) their donation decision (i.e., need for blood, fear of needles). Survey I donors and non-donors had other similarities. In terms of familiarity with blood donation, previous donors and non-donors did not differ significantly with respect to current or previous ineligibility. While there was no significant difference in the rate of being currently or previously ineligible,5 there was a surprising percentage of both donors and non-donors who indicated that they believed that they were currently ineligible to donate blood (70.9% vs. 84.2 %, P = Further comparison of donors to non-donors' familiarity with blood donation showed little variation. On a percentage basis, donors and non-donors did not differ with either family or friends that had donated previously, nor on having a family member or themselves need blood in the last year. There were no differences in the 5 Though non-donors more often reported being ineligible, the difference was not significant. 170

PAGE 184

number of blood drives that donors or non-donors reported seeing, nor in the location of these blood drives (e.g., near school, work, home or other). Previous donors were much more likely to remember their high school as having had a blood drive (63.1% vs. 82.1%, P = .02). Impressively, 72 percent of Survey I respondents reported a high school blood drive! Measures associated with social capital also showed little variation between donors and non-donors. There was no difference in donors' and non-donors' reported participation in high school or college government. In terms of organizational involvements, donors belonged to more organizations than did non donors (median, 0 vs. 1, P = .0719). Likewise, donors reported spending more time with the organizations to which they belong (mean, 2.5 vs. 4.6, P = .0 168). There were no differences in levels of generalized trust or in how connected they felt to their communities. Also, the percentage of donors and non-donors that had donated time, money, clothes or food within the last year was not significantly different. Perceptions of risk were ascertained using nine variables borrowed from a study ofblood donation by Andaleeb and Basu (1995).6 These authors broke the nine variables into three risk scales. A general fear of donation scale was comprised of two items, evaluated on a Likert scale: "needles make me nervous" and "the sight ofblood makes me uncomfortable." A blood donation health-risk scale was 6 For the exact wording of the nine survey items assessing risk, please see Appendix A. 171

PAGE 185

comprised of three separate items: "when donating blood, it is likely that I could acquire health-related problems;" "a person who donates blood runs the risk of. contracting AIDS;" and, "it is risky to donate blood these days." Finally, a general risk-taking scale was comprised of three items: "I quite enjoy taking risks;" "I am an adventurous person;" and, "I sometimes like doing things that are a bit frightening." To verify the use of these risk-scales for this dissertation, Cronbach's alpha was computed for each scale. The alpha-levels are reasonable, .73, .73 and .56 (for medical fear of donation). Further investigation into the robustness of the risk scales was accomplished by entering the nine risk variables into a factor analysis with orthogonal rotation. Three factors emerged from the analysis, each one corresponding to one of the scales hypothesized by Andaleeb and Basu (1995). Only the hypothesized variables loaded on each scale. Hence, the composite risk scales will be used in the remaining analyses. It is important to note that the fear and health risk scales are specific to blood donation, while the risk-taking scale is attempting to capture a more stable characteristic of a propensity to accept risk. Analysis of the risk-scales revealed that there was variation between donor and non-donor risk-seeking, with non-donors being more risk averse (P = .0001). Also, non-donors had higher scores on the general-fear of donation scale (P = .0034), there was no difference in the health-risk scale. 172

PAGE 186

In summary, our analysis of Survey I, by previous donation, indicated that while previous donors appeared to be rather similar when compared to non-donors in terms of socio-demographics, familiarity with blood drives and the factors that encouraged and discouraged blood donation, the groups were differentiated by their acceptance of risk and with one measure of social capital organizational involvements. Results from Survey I: Likely Versus Unlikely Donors Analysis of likely and unlikely donor populations by the dichotomized median response revealed these groups had a slightly different pattern of results; the similarities and differences are summarized below, and presented in Appendix C. Significance level is .05, and statistics for unlikely donors are always presented before those for likely donors. Socio-demographically, the populations did not differ with respect to age, race, sex, family income, parental income, parent's formal education and average hours worked per week. Also, they had similar familiarity with blood donation. There were no differences in previous or current ineligibility, between likely and unlikely donors They were not differentiated by the rate of friends who donated or by family members who needed blood in the last year, although likely donors more often had family members who had donated blood than did unlikely donors (21.5% vs. 173

PAGE 187

40.91%, P = .0223). Unlikely donors reported seeing more blood drives than did likely donors (median, 4 vs. 3, P= .0118), although they reported similar patterns of where they had seen previous blood drives. More often, likely donors reported that their high school had a blood drive, though the result was not significant (67.1% vs. 81.8%, p = .082). Likely and unlikely donors on Survey I showed some important differences in the factors they believed motivated or inhibited their decision to donate blood. Likely donors were more inclined to report that "helping others" was important in their decision to donate blood (median, 5 vs. 6, P = .0058). Likely donors were also more motivated by HIV tests being offered at the blood drive than were unlikely donors (median, 2 vs. 4, P =.014).8 Unlikely donors were more bothered by the screening questions asked (median, 2 vs. 1, P = .0012). Likewise, unlikely donors were more concerned that blood banks were making a profit than were likely donors (median, 2 vs.l, P = .0034). Analysis of the measures of social capital revealed no significant differences between likely and unlikely donors on Survey II. Likely and unlikely donors showed no differences in volunteering food, clothes, money or time. There were no significant differences in student government involvement in either college or high school, although donors tended to be more involved. Likely donors tended to 7 This result has some troubling implications that will be discussed in more depth in the concluding chapter. 174

PAGE 188

belong to more organizations than unlikely donors, and spent more hours with those organizations, though the differences were not significant. The two populations presented different risk profiles.9 Likely donors were less comfortable with risk than were unlikely donors (median, 3.37 vs. 3.00, P = .0007). They also differed in their fear of donation (median, 3.5 vs. 4.25, P = .0161) with donors reporting lower levels of fear. They did not differ in their concern with health risks associated with donation (P= .. 6216). In summary, Survey I respondents, when dichotomized by likelihood of donation, showed remarkably similar socio-demographic profiles, familiarity and exposure to blood drives. However, these same respondents were differentiated by their assessments of risk and by the factors that motivated or discouraged blood donation. Actual donors and non-donors from Survey I were not differentiated by these motivational variables, while likely donors and non-donors were. Results from Survey II: by Previous Donation Below, Survey II (post-September II th) respondents were analyzed by their history of previous donation. The results for this analysis are summarized in Appendix C. 8 Lower scores on the risk scales indicate less concern about risk. 9 The same result was found later on in the survey on a validation question, asked in the opposite direction. 175

PAGE 189

Of Survey II respondents, 40.8 percent indicated that they had donated blood previously. The median response on the policy capturing exercise was six for prior donors and 5.5 for non-donors (P = .0711). When dichotomized at 6, this translated into 56.4 percent of Survey II respondents being classified as likely donors. As with Survey I, a large portion of the discrepancy between the two measures could be explained by consideration of the subjects who indicated that they had previously donated blood, but were now considered to be unlikely to donate. All 18 of these individuals indicated that they believed they were currently ineligible to donate blood Once these 18 subjects are removed the relation between actual donation history and perceived likelihood of donation is stronger. Analysis of Survey II by previous donation revealed that though there were no significant differences in the population in terms of race family income, or parent's formal education, there were differences in gender, with 54.9 percent of donors being female, as opposed to 72.97 percent of non-donors (P = .0366). Likewise, there were differences with respect to age. Donors, were, on average older than non-donors (median, 19 vs. 20, P = .0425). In terms of previous experience with donation as found in Survey I there were few differences on Survey II, between donors and non-donors. Donors were much more likely to have had a family member that donated blood (20.6% vs. 46 %, P = .0027) The difference in the percentage of friends that had needed blood was not significant. Likewise there were no significant differences between 176

PAGE 190

donors and non-donors in the percentage of family members that had needed blood in the last year, in the number of blood drives that the respondents had seen, or in the blood drives' reported locations (work, home, school or other). Of the motivators for their donation decision, donors and non-donors differed on several. On the importance of receiving a and the importance of being asked to donate, 12 the importance for donors was less than for non-donors. Donors indicated more often than non-donors that feelings of guilt were an important motivator, although the difference was not significant. Other differences among these two populations included several inhibitors of donation. Donors, unlike non-donors, reported being less affected by the types of questions asked, by the fear of needles, and by concerns that blood centers were making profits. Analysis of the predictors of social capital revealed no differences between donors and non-donors on Survey II. The respondents reported similarly high rates of high-school blood drives (the over-all rate was over 86 percent). They reported similar rates of government participation in either high school or college, and in the rates of other types of voluntary donations (food, clothes, money and time). There were no differences in other variables, including generalized level of trust and organizational involvements (whether defined as number of organizations belonged to, or hours spent with those organizations). 177

PAGE 191

In terms of risk, there were no differences between donors and non-donors on the overall risk-taking scale or on the health risk scale. Donors reported higher levels of fear of donation than did non-donors (median, 3 vs. 4.5, P=.0006). In a brief review of Survey II analysis, by previous donation, some salient results emerged. Socio-demographically, donors and non-donors differed with respect to both age and gender. Donors were both more likely to be female and older than non-donors. Donors and non-donors on Survey II did not differ appreciably with respect to familiarity with blood donation, or in measures of social capital. There were some differences in the factors that motivated the donation decision. Non-donors reported being more influenced by the offering oft-shirts, by being asked to donate, by fear of needles, by the personal questions being asked and by the concerns over blood banks making profits than were donors. Results from Survey II: Likely Versus Unlikely Donors Analysis of the second survey by likeliness to donate blood revealed that socio-demographically the two groups were not significantly different with respect to family income, their parent's formal education, years in college, average weekly hours worked and age. However, "likely donors" on Survey II were more often female (54.6% vs. 73.9 %, P = .0244). These results are summarized in tabular form in Appendix D. 178

PAGE 192

Individuals currently or previously ineligible to donate indicated that they were more likely to donate than those individuals that were eligible to donate blood (P = .00 15, and P = .0 179). Likely donors and unlikely donors did not significantly differ in having family or friends that had donated, although they did differ in having had a family member who needed blood in the last year. Specifically, 5.45 percent of unlikely donors had a family member need blood as opposed to 24.64 percent of the likely donors (P = .0034). Likely donors were more apt to report that their high school had a blood drive than unlikely donors (83.33% vs. 88.41%, P = .4186). There was no difference in the number ofblood drives seen or in where they reported having seen blood drives. Of the factors that motivated or discouraged donation, likely donors were more influenced by helping others than were unlikely donors (median, 6 vs. 6, P = .0388); none of the other motivators or discouragers differentiated the groups. Also, perceptions of risk were not significantly different between the likely and unlikely to donate groups. Analysis of the measures of social capital revealed that while there was no difference in involvement in high school government, likely donors were much more prone to be involved in college government (1.79% vs 17.65%, P = .0045). There were also no differences in trust or community connectedness. Likely donors belonged to more organizations than did unlikely donors (mean, .37 vs .. 75, .0691), though the differences was not significant. 179

PAGE 193

In summary, Survey II respondents, when dichotomized into likely and unlikely donors, show remarkable consistency in terms of demographics the measures of social capital as well as perceptions of risk. Likely donors, on Survey II, were more likely to be female, involved in college government, have had a highschool blood drive, and most surprising, to be ineligible to donate blood! Another interesting finding is that donors and non-donors on II showed many motivational differences, while likely and unlikely donors on Survey II only differed in the importance of helping others. On Survey I, the opposite pattern of results emerged; there were motivational differences when considered by perceived likelihood, but not by previous donation. This rather laborious bivariate comparison of the survey respondents by previous and likely donation provides us ample familiarity with the populations to proceed with a factor analysis designed to reveal the structure of decision making for blood donation These bivariate analyses are revisited when we evaluate the dissertation hypotheses at the beginning of the concluding chapter. Structure of Decision Making: As Revealed by the Policy Capturing Exercise For each of the survey respondents, a regression analysis was run in order to derive each individual's weights on the six hypothesized dimensions. The dependent variable was the "likelihood of donating blood" as recorded on a seven180

PAGE 194

point Likert scale for each of the 30 hypothetical blood drives. The independent variables were the six predicted dimensions with the responses randomly ordered (e.g. need, location).1 From each individual's regression analysis, the beta weights on each of the six factors were retained. Next, a series of factor analyses on these beta weights were conducted to provide insight into the structure of decision making regarding blood donation. This section begins with an evaluation of the appropriateness of using policy capturing with this study population; this entails an evaluation of the "task familiarity" of the subjects. There is also a brief description of the included in the regression analyses. This is followed by a review of the findings from the survey-by-survey factor analyses. Finally, the results of a factor analysis on both surveys pooled are discussed. Task Congruence and Task Familiaritv The use of policy capturing is predicated on the assumptions that the task familiarity was high and that the task congruence was abstract (as opposed to complex).11 Several simple cross-tabulations were helpful to evaluate the appropriateness of the assumptions regarding task familiarity. The assumption that 10 Please see page 169 for a description of the six hypothesized dimensions. 11 See Chapter Four for a discussion of this issue. 181

PAGE 195

task familiarity would be high was based on the assumption that subjects would be familiar with seeing blood drives and with making decisions about whether or not they would donate blood. The task familiarity was evaluated for both survey populations.12 A substantial portion of the populations from each survey reported that their high school had a blood drive (71.9% and 86.3%, P = .0006). In addition, most respondents reported that they had seen numerous blood-drives within the last year (median, 3 vs. 4, P = .2054), while only three survey respondents reported that they had not seen any in the last year. The next logical question was whether people remembered why they did or did not donate blood. The results of the surveys indicated that roughly 61 percent of Survey I respondents and 55 percent of Survey II respondents reported that they could remember why (P = .3250) they decided to either donate or not donate at the last blood drive they had observed. In addition, 57.4 percent of Survey I respondents and 64.7 percent of Survey II respondents could remember their rationale for donating or not donating at their high school blood drive (P = .2661 ). Another logical question was whether donors, as compared to non-donors, were differentially familiar with making decisions about donating blood. To evaluate this question for the surveys pooled, familiarity by previous donation was 12 The comparisons in this paragraph are between Survey I and Survey II respondents and are always presented in that order. 182

PAGE 196

considered. It was found that at the last blood drive, nearly 65 percent of donors and 53 percent of non-donors reported remembering their rationale for their donation decision (P = .0582). Similarly, donors tended to remember their rationale for donating or not donating at their high school blood drive (67.6%) more often than did non-donors (56%), although the difference was not significant (P = .075). While donors were slightly more inclined to remember making blood donation decisions (for or against), significant portions of both surveys reported remembering the basis for their blood donation decisions. This analysis provides sufficient justification for concluding that task familiarity was high, and hence, that policy capturing was an appropriate research tool. In review, the data revealed that subjects were rather familiar with blood donation; and, that a vast majority both recalled their high school blood drive and remembered their rationale for either donating or not donating at that drive. Significantly, the absolute numbers were high and the differences did not vary greatly by survey though they did vary (though not significantly) by previous donation. The Dimensions Six dimensions were hypothesized to affect the blood donation decision and were experimentally manipulated in the policy capturing exercise. Respondents reviewed six statements for each hypothetical blood drive, one for each of the six 183

PAGE 197

dimensions identified in Table 5.2. For more detail, see Chapter Four for a description of the dimensions and Appendix A (page 243) for the precise wording for each level of the dimensions. Table 5.2: Summary of the Dimensions Used in the Policy Capturing Exercise Dimension Name Description of the Dimension Coding DISTANCE Distance to the blood drive 1 = on the way to class 2 = 1 0 minutes walk 3 = 15 minute bus ride NEED The salience of the need for I 4, with 4 being less blood salient CHARITY A $7 donation to charity I =donation, 0 = t-shirt LOTTERY Entry into a lottery for a CD I = lottery, 0 = t-shirt player DEED A reminder that blood I = reminder about good donation is a good deed deed, 0 =reminder about tshirt SPONSOR Sponsor of the blood drive I-4, from local to national The Regressions An ordinary least squares regressions was conducted on each subject. The dependent variable was the subjects' stated willingness to donate blood. The independent variables were orthogonal orderings of the six dimensions experimentally manipulated in the 30 hypothetical blood drives. The purpose of these regressions was to determine the weight each subject placed on the six hypothesized dimensions. The results for each regression were reviewed independently and were then combined. 184

PAGE 198

Subjects who had either incomplete responses (N= 4) or no variation in their responses (N = 21) were eliminated from subsequent analysis. Schmitt et al. ( 1991) suggested eliminating subjects with low R2 as this might have been an indication of the lack of a policy. In this study, however, it was possible that exposure to the September 11th bombings might have been an endogenous factor that could have caused a portion of the variance in R2 likely in a non-constant, multiplicative fashion. Hence, no R2 cutoff was incorporated. The final dataset consisted of227 total subjects, with 115 from Survey I and 112 from Survey II. To determine the relative importance of each variable for each subject, a standardized coefficient was produced. This was done by multiplying the semi partial correlation coefficient by 100 Zedeck and Kafry ( 1997) refer to these as "objective implicit weights." These implicit weights describe "the relative contribution of each of the criterion elements as a proportion of the predictable linear variance" (Zedeck and Kafry, 1977). The implicit weights were reviewed on an individual basis. Average parameter estimates, as well as summary statistics, by survey, are presented in Table 5.3 185

PAGE 199

Table 5.3: A Brief Summary ofthe Regression Analyses Average Parameter Estimate Survey Ia Survey Ila P-valueb Sponsor .03 (-.13, .17) .04 ( -.03, .15) .2899 Distance .70(.17, 1.17) .66 (.26 1.1) .6543 Need .18 (.03, .48) .23 (.08, .41) .6154 Good deed .12 (-.20, .42) .10 (-.10, .28) .8564 Donation to Charity .15 (-.21, .98) .17 (-.1, .47) .7687 Lottery for CD -.14 (.42' .07) -.16 (-.43, -.02) .3519 st rd a. Medtan ( 1 3 quartile) b. As determined by the Wilcoxon Ranks Sum Test The Factor Analyses A series of factor analyses were conducted on the beta-weights from the regression analyses described above and summarized in Table 5.3. The analytic methods and their corresponding results are described in the following text. First, the results from each survey are discussed independently, and then the findings of the two surveys are compared. Finally, a series of analyses performed on the pooled surveys are presented. A Factor Analysis of Survey I For the first survey, there was remarkable consistency between the factor loadings on a variety of rotation methods. Using parsimony and interpretability as 186

PAGE 200

the guide, the Orthomax rotation method was selected for the analysis of Survey 1.13 Factor analysis was an appropriate analytic procedure for this data, given the low correlations between the factors in Survey I. The results from the factor analysis are summarized in Table 5.4. The scree plot14 of the eigenvalues revealed that the first three factors should be retained. The factors for this survey were not complex. That is, we did not have multiple variables loading on each factor. Hence, in Table 5.4, the variable loading most strongly in each factor is included next to thefactor ordering. The first factor, with an eignvalue of 1.9 accounted for 32 percent of the variation. The remaining two factors accounted for 22 percent and 21 percent respectively. Table 5.4: Results from a Factor Analysis: Survey I Factor Name Eigenvalue Proportion Variance Explained 15 Factor 1: Good-deed 1.945 .3242 1.02 Factor 2: Distance 1.292 .2153 1.01 Factor 3: Lottery 1.24 .2069 1.01 Table 5.5 presents the rotated factor pattern for Survey I. On this survey, administered prior to the September 11th bombings, the variable loading most 13 An orthogonal manipulation is one in which the factors are uncorrelated after their initial extraction and remain uncorrelated after rotation. This is in contrast to oblique rotations in which the factors are allowed to correlate upon rotation. 14 Derived from the geological term for the rubble at the bottom of a mountain, the scree refers to a plot of eigenvalues. The last factor retained is the one where there is a sharp break in the slope (Norman and Steiner, 2000). 187

PAGE 201

strongly on the first factor corresponded to donation being a good-deed, loading at .93. On the second factor, the distance to the blood drive loaded strongly at .99. On the third factor the lottery for the CD-player loaded strongly (.97). Table 5.5: Rotated Factor Pattern: Survey I Factor 1 Factor 2 Factor 3 Distance .11 .99 .05 Charity .04 -.03 .14 Lottery .05 .06 .97 Need .04 .07 -.15 Good-deed .93 -.13 .06 Sponsor -.37 .10 .09 A Factor Analysis of Survey II The second survey, like Survey I, had low levels of correlation between the factors. Considerations of interpretability and parsimony led to the selection of Orthomax as the most suitable rotation method. The results from the factor analysis are summarized in Table 5.6. After analysis of the scree plot and consideration of the eigenvalues, three factors were retained. Once again, the factors were very simple, with one variable strongly loading on each factor. The first factor, with an eignvalue of2.0 accounted for 34 percent of the variation in the model. The second and third factors accounted for 23 percent and 18 percent respectively. 15 This is variation after rotation. 188

PAGE 202

Table 5.6: Results from a Factor Analysis: Survey II Factor Name Eigenvalue Proportion Variance Explained16 Factor 1: Charity 2.020 .3367 1.01 Factor 2: Lottery 1.353 .2255 1.009 Factor 3: Distance 1.08 .1799 1.008 Table 5.7 contains the rotated factor pattern for Survey II. The variable loading most strongly on the first retained factor was charity (.98). For the second factor, the lottery for a CD-player was most important (.98). Distance to be the blood drive was the variable that loaded most strongly on the third factor (.99). Table 5.7: Rotated Factor Pattern: Survey II Factor 1 Factor 2 Factor 3 Distance -.01 -.02 .99 Charity .98 .04 .02 Lottery .04 .98 .02 Need .13 .21 .05 Good-deed .14 .08 -.13 Sponsor -.08 .07 .10 The variable loading most strongly on the first retained factor was charity (.98). For the second factor, the lottery for a CD-player was most important (.98). Distance to be the blood drive was the variable that loaded most strongly on the third factor (.99). 16 This is the explained variance after rotation. 189

PAGE 203

The structure of factors between the two surveys has some differences. The reminder that donating blood is a good deed, the most important factor (see Table 5.5) on the first survey did not load strongly on any of the retained factors in the second survey. The distance to the blood drive as well as both variables indicating incentives for donation ( charity17 and lottery) were important for both surveys, although the ordering was slightly different. The implications of these results will be discussed in Chapter 6. A Factor Analysis of the Surveys Combined The next research step -to compare the structure of decision-making between the two surveys -was a challenge as there was no obviously satisfactory method for making the comparison. However, in recently published literature, Darlington (2001) suggested that when comparing factor analyses between two different populations, one can include a dichotomous variable to indicate the population (in this case, the survey number). Hence, we conducted one additional factor analysis. In that analysis, all subjects were included as was a survey indicator variable. As before, the levels of correlation between the variables included in the factor analysis were low, indicating that the data were appropriate for factor 17 Charity was the most significant variable in the fourth but unretained factor for the first survey 190

PAGE 204

analysis. For simplicity, the orthomax rotation method was selected. After analysis of the scree plot, and consideration of the eigenvalues, four factors were retained The results are summarized in Table 5.8. Table 5.8: Results From a Factor Analysis ofBoth Surveys Combined18 Factor Name Eigenvalue Proportion Variance Explained Factor 1: Good-deed 1.969 .3281 1.008 Factor 2: Distance 1.287 .2146 1.008 Factor 3: Lottery 1.186 .1976 1.006 Factor 4: Charity .82 .1367 .1.006 We found that the significant variables loading in each of the first four factors were the same as in Survey I, while in the fifth, un-retained factor, the variable indicating the survey loaded most strongly. The first four factors combined accounted for 87.7 percent of the variation in the factor analysis. Hence, it appeared that there was a unique contribution from the survey indicator variable, although its contribution to the factor-analytic model was not significant. Table 5.9 presents the rotated factor pattern for the combined analysis. The first factor is predominantly explained by the reminder of donation being a good deed (.92). Factor two is mostly explained by the distance to the blood drive (.99); while factor three is explained by the lottery for CDplayer. The factor loading strongly on the final retained factor is the donation to charity (.98). 18 This is the explained variance after rotation 191

PAGE 205

Table 5.9: Rotated Factor Pattern: Surveys Combined Factor 1 Factor 2 Factor 3 Factor 4 Distance -.11 .99 .04 .01 Charity .07 -.01 .10 .98 Lottery .05 .04 .98 .10 Need .04 06 -.17 .17 Good-deed .92 -.13 .07 .08 Sponsor .35 -.I -.08 .05 Survey .01 -.01 -.03 .01 The results from these structure-of-decision-making analyses demonstrated some consistency across surveys. Median likelihood to donate blood was on average higher on the second survey which led us to surmise that respondents exposed to the September 11th bombings manifested the impact of the bombings as a multiplier of sorts. Certain dimensions did change in importance or more appropriately, several factors gained importance, while others did not. Hence, once controlling for each survey's population's exposure to the bombings, we found four dimensions that influenced donation: the distance to the blood drive; a lottery ticket for a CO-player; an offer of a donation to charity as an incentive for donating blood; and, a reminder that blood is a good deed. Having gained the insight that four factors affected the blood donation decision, the next step was to identify socio-demographic, familiarity or other differences between individuals who were most strongly influenced by these same four factors. This amounts to a clustering or grouping of respondents by their blooddonation decision making strategies and an analysis of these groups. 192

PAGE 206

Clustering of the Decision Makers by Factor Analytic Results To group subjects by their decision policies, subjects were assigned to the factor for which they had the highest loading. Hence, a new variable was created that indicated the subjects' "group." This variable was analyzed by survey. The results are presented in Table 5.1 0. The results indicate that there were no differences in group by survey. This suggests that there was consistency in decision making patterns across surveys; hence, for the remainder of this analysis, the surveys are pooled. Table 5.10: Clusters, as Determined by the Factor Analysis, by Survey Factor 1 Factor 2 Factor 3 Factor 4 P-value Distance Lottery Charity Good-deed Survey 1 53.7 45.9 43.33 61.54 0.212 Survey 2 46.3 54.1 56.67 38.46 Socio-demographic and predictor variables were then analyzed by the variable indicating group for both surveys combined. The results are summarized below and presented graphically in Table 5.11. Factor 1 (distance) is not differentiated by socio-demographics. This group has the highest percentage oflikely donors, and 45 percent of Factor 1 members-have donated previously. They are the least likely to believe they are ineligible to give blood. They are not motivated by feelings of guilt or by tests for STDs. The distance to the blood drive, as well as the good-deed of giving blood, are important 193

PAGE 207

factors in their decision making structure. Over 20 percent indicate that a family member has needed blood in the last year, and they donated food and money more often than the other groups. 19 This cluster appears uniquely "civic-minded" as compared to the other clusters, given their high rate of other charitable donations, and is referred to as the civic-minded cluster. Factor 2 (lottery) is marked by their preference for the lottery and t-shirt incentives and for their concern over the distance to the blood drive. This is the group most likely to have donated blood previously, and by their older age at first donation than the other groups. The lottery group contains the greatest percentage of Caucasians (47.5 %) and the fewest Asians (35.6%) of any of the clusters. They are more likely to have had friends, but not family, who have donated. They are influenced by the convenience of the blood drive. They are the least likely to believe you can get AIDS via donation. This group appears to be responding predominantly to self-interested motives and logically will be termed self-interested. 19 This group has some similarities to what Putnam describes as the civic-minded generation of World War II. The connection will be substantiated in the final chapter. 194

PAGE 208

Table 5.11: Analysis by Cluster (selected variables) Factor 1 Distance Socio-Demographics Female gender,% 60.38 Age1 21 (19, 22) Donation History Donate previously, % 45.28 Likelihood of donating, % 51.85 Currently ineligible, % 66.04 Age at first donation 1 17 (17, 18) Motivators 1 Guilt 2 (1, 3) T-shirt 2 (1, 4) Test for STDs 2 (1, 5) Need 6 (5, 6) lnhibitors1 Convenience 4 (3, 5) Needles 3 (2, 5) Time required 4 (3, 5) Disease 4 (3, 5) Get AIDS 3 (5, 5) -st rd 1 Medtan (1 quartile, 3 quarttle) Factor 2 Factor 3 Factor 4 Lottery Charity Good-deed 61.67 66.1 49.02 20 (19, 21) 19 (19, 21) 22 (20, 24) 52.46 33.33 40.38 49.18 35.59 41.18 78.69 75 82.69 18 (17, 18) 17 (17, 17) 17(16,18) 3 (1, 4) 3(1.5,4) 2 (1, 3) 2(1,4) 2(1,3) l (1, 3) 3 (1, 5) 2 (1, 4) 2 (l, 4) 6 (5, 6) 6 (5, 6) 5 (5, 6) 5 (5, 6) 4 (3, 5) 4 (3, 5) 3 (1, 5) 4 (2, 5.5). 3 (1.25, 5) 5 (4, 6) 4 (2.5, 5) 4 (3, 5.5) 4 (4, 5) 4 (3.25, 5) 4 (4, 5) 5 (4, 5) 5 (3.25, 5) 5 (4, 5) Factor 3 (charity) is more likely to be female, and tends to be slightly younger. 54 percent of the charity group is Asian-the highest representation of Asians in any of the groupings They are the least likely to have donated previously, to have donated at their high school blood drive, and indicate the lowest 195

PAGE 209

average levels of likelihood to donate. They are also the most likely to believe they are presently ineligible to donate blood (35%). They have lower levels of family and friends who have donated. They are more likely to be afraid of needles and have higher levels of concern over risks associated with donation. They are tenned the worriers. Factor 4 (Good-deed) are more often male, they are the oldest of the factors, 40 percent have donated previously. They are also the least likely to report their parents income as being "upper-class" (17%). They are the least likely to be motivated by the t-shirt. The donation to charity has great appeal for this groupthe lottery does not. They are also influenced by the perceived need for blood. They have the lowest levels of college or high school government involvement. Hence, they will be called the principled donors In summary, four clusters of decision makers emerge from our factor analyses. The clusters have unique personalities. Policy recommendations targeted to these various groupings will be elaborated up in the concluding chapter. Next, an alternate method for clustering decision-making policies for blood donation from the policy capturing exercise is explored. Cluster Analysis Cluster analysis is a technique that groups subjects into "clusters" based on either their similarities or differences. Cluster analyses have often been used for the 196

PAGE 210

classification of subjects from policy capturing exercises. Hence, for comparability, a cluster analysis was also conducted. Only subjects with an R2 of over .60 were used in this analysis, as is recommended by Schmitt et al. ( 1991 ). Hence, 66 subjects are included, 31 from Survey I and 35 from Survey II. The Ward Method20 was used for clustering the subjects. The number of clusters to be retained was determined by an analysis ofR2 from the clustering, in an inverse scree plot (Cooksey, 1996). Hence, four clusters were retained. A variable indicating cluster was used in bi-variate analyses of the weights from the regression analyses as well as socio-demographics, blood donation familiarity and motivational differences. The clusters did not differ significantly by survey. Cluster One was the most likely to report that they would donate blood; they were also the least affected by distance of the blood drive. They were the most likely to report that their high school had a blood drive (85%) and were also most likely to have donated blood in high school (37%). This group has the highest levels of trust and community connection. Cluster Two was more greatly influenced by the donation to charity, the need for blood and the reminder that donating is a good deed. They were the least likely to have donated previously and did not indicate that they were very likely to 20 "In Ward's minimum-variance method, the distance between two clusters is the ANOVA sum of squares between the two clusters added up over all the variables. At each generation, the within cluster sum of squares is minimized over all partitions obtainable merging two clusters from the previous generation" (SAS, 1990, 536) 197

PAGE 211

donate at the blood drives described. They were the most likely to have donated food and clothes previously. Cluster Three was most influenced by the need for blood, the reminder that donating is a good deed, and the distance to the blood drive. This group was disproportionately female. Nearly 69 percent indicated that they had donated previously, and roughly 62 percent were "likely" donors. Cluster Four was the most likely to have donated previously, and were influenced strongly by the distance to the blood drive as well as the offering of a lottery ticket for a CD player. They had the lowest median response to the questions regarding likelihood of donating blood. They were also the group most likely to be previously ineligible. In summary, the results of the cluster analysis revealed that there were surprisingly few, demographic, familiarity, or motivational differences between the clusters. They were differentiated by the dimensions entered into the cluster analysis, as well as donation history and perceived likelihood of donating blood. To do this analysis justice, additional thorough consideration of the clusters would be necessary. In addition, there appears to be some correspondence between the clusters derived from the factor analysis and those derived from the cluster analysis; although the correspondence is not clear. Further analysis into the similarities and differences of these groupings is worthy of consideration. However, given that 198

PAGE 212

cluster analysis, unlike factor analysis, does not result in a predictive model, the factor analytic groupings will be used in the duration of this dissertation. Between Group Analysis Survey I Versus Survey II Thus far, our inquiry had adopted either a within individual or within group level of analysis. The final critical piece of analysis is a between-survey-group comparison, because this helps address both the issue of stability of the factors across surveys and the question of the effect of the September 11th terrorist attacks on the Survey II population. To accomplish the comparison, three different types of between-group analyses were conducted. The methods, as well as the results from these methods, are described below. A Wilcoxon Rank Sum Test of the regression weights derived from the individual-level regression analyses revealed no significant differences between the surveys. However, it should be noted that there was a limitation to this comparison; namely that the factors that influenced low-likelihood donors and high likelihood donors were given equal weight. Moreover, as was noted previously, the "likelihood to donate" question elicited a significantly higher response by Survey II respondents. Hence, to adjust for this difference, the betas were weighted by the median "likelihood to donate" response. The Wilcoxon Rank Sum Test of these weighted-betas again revealed no significant differences. Hence, we surmised that although after the bombings 199

PAGE 213

(affecting only Survey II respondents) an individual's stated likelihood to donate blood was, (on average) higher, the relative importance of the six dimensions was not significantly different. A second between-group analysis also added insight into the question over comparability ofthe results between the two surveys. To identify if there were differences in the structure of decision-making between the two populations, a pooled factor analysis was constructed.21 This was accomplished by running a separate factor analysis that included all respondents from both surveys. The factor analysis included a dummy variable indicating to which survey group the respondents belonged. Four strong factors emerged (good deed, distance, lottery and charity), accounting for 77.8 A fifth factor, indicating the survey, had an eigenvalue of .82 but was not retained because it was and on the flat portion of the scree plot. A third and final between-group analysis was conducted in recognition of the on average higher responses to the second survey. A new "likelihood to donate blood" variable was constructed. Each individual's median likelihood to donate blood (this is the median from their 30 responses to the Likert scale) was assigned to the appropriate quartile (1-4) for their survey (survey 1 or 2). Median responses for the six policy capturing were analyzed for each quartile, by survey. That is, unlikely donors (quartile 1) were compared between the two surveys. 200

PAGE 214

When the beta-weights for the six policy capturing dimensions were compared once again for each quartile, by survey (via Wilcoxon Rank Sum Test), no major differences were found. That is the least likely donors (quartile 1) on both surveys did not differ in their use of the cues-likewise quartiles 3 and 4 The only difference found was for the dimension "need" in quartile 2. Respondents from the second survey had weighted the need for blood more strongly than had respondents from the second survey. In summary, three different types ofbetween-group-analyses were conducted. From these analyses, we surmised that the two survey populations had similar decision-making policies regarding blood donation. Hence, it appeared that the September 11th attacks increased average likelihood of donating but did not appreciably alter the overall structure of decision making. Therefore, we can have increased confidence in the decision to pool the data between the two surveys. Conclusion In conclusion, the two surveys contained a total of 252 undergraduates who, besides their age, did not differ a priori in any observably significant ways. It was possible, however, that the two populations differed in significant ways cannot be identified. Nonetheless, the pattern of results (aside from a few noteworthy differences) revealed some consistency in decision-making styles between the 21 Described above. 201

PAGE 215

respondents on the two surveys: the distance to the blood drive and the lottery were important for both surveys. The noteworthy differences included the following findings. On average, respondents on Survey II reported that they were more likely to donate blood than were respondents to Survey I. The donation to charity was important for the second survey, whereas the reminder that donation was a good deed was important for the first survey. The analysis of the combined surveys revealed that the differences between the two surveys were an artifact of an omitted variable. When a dummy variable indicating the survey was included in the analysis, the results for the entire population were the same as for Survey I. The implication, by and large, was that the structure of decision-making was strikingly similar between the survey groups. From the main factor analysis, consideration of the most important factor for each individual permitted the identification of several "types" of donors who had different characteristics and decision-making policies. Policy recommendations for each policy type will be discussed in the concluding chapter. 202

PAGE 216

CHAPTER SIX SYNTHESIS AND CONCLUSION Introduction This concluding chapter offers a synthesis of findings and observations from the history ofblood donation as described in Chapter Two, the review of both theoretical and applied literature surveyed in Chapter Three, as well as the multi method investigation into the motivations for and against blood donation discussed in Chapter Four. This chapter begins with an evaluation of the study hypotheses presented in Chapter Four, based on the analysis described in Chapter Five. Second, the theoretical implications derived from these findings are explored. Third, recommendations for blood donation policy are presented. Fourth, the limitations of the study are reviewed. This is followed by a discussion of the role of the judgmentand-decision-making (JDM) style of research for other public policy 203

PAGE 217

issues, addressing the importance of understanding the psychology of targets of public policy. Fifth, areas for further research will be outlined. Next, the applicability of the dissertation's findings to other voluntary behaviors, including other voluntary medical donations, is discussed. The chapter concludes with a review of what we have learned about voluntary donations in the wake of a national cnsts. Evaluation of the Hypotheses Seven hypotheses were proposed in Chapter Four. In this section, all seven hypotheses are evaluated in light of the analysis. Where possible, P-values from the bivariate analysis described in Chapter Five are included. Unless otherwise stated, the P-values are derived from a Wilcoxon Rank Sum test of the variables in question. Evaluating the Altruism vs. Self-Interest Hypotheses A series of hypotheses were proposed, investigating the relationship between the costs and benefits of donation. The opportunity cost of blood donation was posited to be a function of distance, perceived risk and discomfort. The benefit of blood donation was considered in terms ofbenefit to others as well as benefit to oneself. The first hypothesis proposed is: 204

PAGE 218

H1A: Individuals are more likely to donate blood if the perceived opportunity cost of blood donation is low. There were several measures by which H1 A could be evaluated, from both portions of the survey. From the policy-capturing exercise, the distance to the blood drive was included as a measure of cost. No clear relationship between likelihood of donating blood and distance to the blood drive emerged when the surveys were pooled (P =.2212). However, an inverse relationship between likelihood of donation and distance was identified on Survey I, with the importance of distance decreasing as likelihood of donation increases (P=.0136). There is no clear relationship between likelihood of donation and distance in Survey II (P=.3925). Another measure of cost evaluated in the second section of the surveys was the perceived risk of donation. Individuals who perceived the risk of donation to be high reported that they were less likely to donate blood (P= .0134, .0106 and 0.476 for the three risk scales). Fear of needles differentiated previous donors from non donors (.001) but not likely versus unlikely donors (P=.01). In addition, we measured cost in terms of reaction to intrusive personal questions asked at blood drives. Again, we saw significant differences between donors and non-donors (P = .01) and likely and unlikely donors (P=.001) with non-donors and unlikely donors being more bothered by questioning than donors. 205

PAGE 219

In summary, "cost," when conceptualized as distance, perceived risk and discomfort differentiated likely and unlikely donors. These results support the idea that there is an inverse relationship between the perceived cost of donation and likelihood of donation. That is, H1 A is accepted as posited; individuals are more likely to donate blood if the perceived cost of donation is low. It should be noted that for this hypothesis cost was considered in the absence of benefit. The next hypothesis concerns the relationship between perceived benefits and donation, in the absence of cost. H18: Individuals are more likely to donate blood if the perceived benefit of blood donation is high (given H1A) The benefit ofblood donation was measured through two different manipulations in the policy-capturing exercise. One manipulation involved the benefits to others, while the second involved benefits to oneself. In the exploration of the perceived benefits to others, a statement of the need for blood in the hypothetical blood drives was varied. The need for blood (benefit to others) was found to be highly important in the judgmental policies of the survey respondents of both surveys. Likely donors were more influenced by the statement of need than were unlikely donors (P = .001). Hence, when considering benefits to others, it appears that there is a positive relationship between perception of benefit and 206

PAGE 220

likelihood of donating blood. Thus, we can accept Hypothesis H 18 as posited; donation is more likely when perceived benefits are high. We also investigated benefits to one's self. In this context, benefits to self were the various incentives offered for blood donation (e.g., the t-shirt, a lottery ticket, or a donation to charity). The comparison of the effect of various incentive types are discussed in more detail under the discussion of Hypothesis HzA The previous two hypotheses have been rather intuitive. Individuals are more likely to donate when the cost is low, and, they are also more likely to donate when the benefit is high. The next hypothesis is designed to consider the interaction between costs and benefits for blood donation decisions. H1c: When the perceived benefit of blood donation is low, potential donors adjudicate based on cost. When the perceived benefit is high, potential donors do not adjudicate based on cost. This hypothesis was designed to test if cost functions as a gate to donation when benefit is low, but not when benefit is high. One way to evaluate the relationship between cost and benefit was to consider the importance of the distance to the blood drive between the two surveys. Many have suggested that the perceived need for blood directly after the September 11th bombings was rather high, and the data supported the idea. Hence, we expected 207

PAGE 221

that there would be a greater relationship between cost and likelihood of donation on the first survey than on the second survey. If the benefit of donation for the first survey were lower, we anticipate that the distance to the blood drive would be more salient. This, in fact is what we found. An analysis of the importance of distance to the blood drive, by likelihood of donation for each survey, revealed that Survey I respondents are differentiated by the importance of distance to the blood drive (P = .01): Survey II respondents are not (P = .39). Figure 6.1: The Relationship Between Perceived Benefits and Importance of Cost1 Perception of Benefit Point A, represents the case where the perceived benefit was low and hence the importance of cost is high. This can be exemplified by the individuals on 1 Only the endpoints of this hypothesized relationship can be verified with the current data. The shape of the curve is conjectural and should be verified in future research. 208

PAGE 222

Survey I, when the benefit of donation is not unusually high. Alternatively, Point B represents the case in which the perceived benefit is high (Survey II, after the bombing), and the importance of cost is low. Hence, we have support for the hypothesized relationship between cost and benefit in Hypothesis H 1c. When the perceived benefit of donation is low, potential donors adjudicate their decision based on cost. When the benefit is perceived to be high, individuals do not adjudicate based _on cost. This relationship is depicted graphically in Figure 6.1. This relationship between perceived costs and benefits and likelihood of donation provides useful insights for those involved in recruiting new blood donors. Individuals in these two surveys seem to be responding to their environment, that is, the decision calculus appears to be context dependent. This context dependence is not surprising, but it is in contrast to the non-contextual presentation of blood donors in recent publications regarding recruiting blood donors (Kohr and Sayers, 1998). Hence, some individuals may be, in general, less likely to donate blood for intrinsic reasons, such as fear of needles. However, there are extrinsic, environmental factors that can affect the decision process. Different recruitment strategies are likely to be more effective at some times rather than at others (depending on context), and for some individuals rather than others (for intrinsic reasons). This hints that a multiplicity or menu of recruitment strategies might be 209

PAGE 223

more effective than a single unifying recruitment strategy, a theme that will be revisited later in this chapter. Another implication of the relationship between perceived costs and benefits is that recruitment strategies should be aimed at moving potential donors along the curve hypothesized in Figure 6.1. Namely, efforts aimed at reducing perceived cost as well as increasing perceived benefit should improve the probability that a given individual will donate blood. Moreover, individuals were willing to disregard cost when the perceived benefit was high. This suggests that strategies should focus on increasing perceptions of the benefit of donation, since high benefits situations diminish consideration of costs. The challenge is to increase perceptions of benefit without appearing to "cry wolf," as this will likely result in a loss of credibility. Specific methods for accomplishing this shift in perceptions will be discussed in more detail in a later discussion of practical implications. Evaluating the Incentive Hypothesis The next hypothesis is designed to probe the comparative effectiveness of incentives designed to appeal to altruism as opposed to incentives designed to appeal to self-interest -pecuniary or otherwise. 210

PAGE 224

HzA: An incentive structure that appeals to altruism alone will increase an individual's perceived likelihood of donating blood more than an incentive structure that appeals to self-interest alone. If this hypothesis were true, then we should find that individuals were more likely to donate blood when they were offered an anonymous donation to a charity as opposed to either at-shirt or entry into a raffle. From the policy-capturing exercise, we found mixed results. The comparison category for the charitable donation was the t-shirt. We found that the donation to charity was a significant factor in the second survey (explaining 34 percent of the variation in the analysis), though not in the first surveys In the pooled analysis, the charitable donation loads strongly on a retained factor and explains 14 percent of the variation in the final factor analysis. Hence, the incentive designed to appeal to altruism was more influential than one of the self-interested incentives (the t-shirt) but hot the other (the lottery). This likely indicates that the relative benefit (to self or others) of these incentives was not equivalent. It is possible that subjects were responding to the "absolute benefit" of the incentive, and not the benefit that they would receive personally. This idea is explored in more detail below, as will the implication of this mixed result regarding the effectiveness of the motives behind incentives It should be noted that two ofthe groups from the clustering of subjects responded quite strongly to incentives that appealed to altruism (the civic-minded 211

PAGE 225

and the worriers). In fact, the principled group found the incentives that appealed to self-interest to be a disincentive -lowering the probably of donation. It appears, then, that the mixed support for Hypothesis H2 A is masking the differences in decision structure by the various donor groups. The groups of potential donors are responding to the incentives in different ways. This reflects pivotal differences in underlying motivational structures between the groups and suggests that incentives that appeal to altruism as well as incentives that appeal to egoism, can both be effective, but for different segments of the population. Evaluating the Social Capital Hypotheses A series of three hypotheses were included to investigate the importance of social capital in the blood donation decision. The hypotheses will be briefly evaluated individually and then discussed summarily. H3A: Individuals who reflect higher levels of trust are more likely to donate blood than individuals with lower levels of trust. Individuals' level of trust was evaluated with a standard question regarding how trustworthy survey respondents thought others were. This admittedly crude measure of social capital did not differentiate between likely and unlikely donors (P = .16) for either survey; nor did it differentiate between the two surveys (P=.62). 212

PAGE 226

The comparison of donors' and non-donors' responses revealed no significant differences either. Hence, we reject Hypothesis HJA for lack of support. H3s: Individuals who consider themselves to be connected to their communities are more likely to donate blood than individuals who do not consider themselves to be connected to their communities. This hypothesis can be evaluated by consideration of a single response on the surveys a question asking how much connection individuals reported feeling to their communities (responses indicated on a Likert scale). Unlikely donors reported higher levels of community connectedness than did likely donors (P=.03). No difference was found by previous donation. Hence Hypothesis H3s is rejected. There is also an intriguing difference in connectedness by survey. Survey II respondents reported feeling less connected to their community than did Survey I respondents (.02). We anticipated that social capital would have been higher for Survey II respondents than for Survey I respondents, as a result of the feeling of "we" generated by the September 11th bombings. The distinction could reflect an inherent difference between the groups that is unrelated to the September 11th bombing. Perhaps it represents a timing effect as Survey I was administered at the end of an academic year, while Survey II was administered at the start of the academic year. Unfortunately we are not able to discern the cause of the difference 213

PAGE 227

in reported community connectedness, and hence, will look to our last measure of social capital. H3c: Individuals with high levels of associational involvement are more likely to donate blood than individuals with low levels of associational involvement. The final measure of social capital is associational involvements. Involvements were operationalized both as the number of organizations that an individual belongs to as well as the total number of hours spent with all of the organizations on an average week. The analysis revealed no differences between the surveys. Likely donors tended to belong to more organizations, though the difference was not significant (P=.06) and to spend more hours with those organizations (P=.04). Likewise, previous donors were more likely to belong to membership organizations (P = .06) and spent more time with these organizations (P=.02). This corresponds with Putnam's observation that "what really matters from the point ofview of social capital and civic engagement is not merely nominal membership, but active and involved membership" (Putnam, 2000, 58). Hence, we have mixed support for Hypothesis H3c; individuals with more substantial community involvements (measured as hours spent with organizations) are more likely to donate blood. 214

PAGE 228

A measure of social capital was also included in the policy capturing portion of the survey. The variable indicating the "sponsor" of the blood drive was not found to be influential in the structure of decision making for either survey. The relative importance of the sponsor of the blood drive was compared between surveys, and by previous and likelihood of donation. Although we did not find differences in the importance of the sponsor of the blood drive by survey, or by previous donation, we did find a difference by likelihood of donation on Survey II (P=.03). Unlikely donors on Survey II were more highly affected by social capital, when operationalized as the sponsor of the blood drive, than were likely donors. This is a perplexing finding. The median response for likely donors on Survey II was similar to the median responses for Survey I respondents. So, there is something unique about how unlikely donors on Survey II were viewing the sponsor of the blood drive. Until more research can be conducted, this result will be regarded as spurious Our multiple measures of social capital present a murky picture of the relationship between social capital and likelihood of donation. Trust and perceptions of community connectedness were not found to be indicative of donation. However, organizational involvements, operationalized as hours spent with membership organizations, were found to be predictive oflikelihood to donate. The finding that more organizationally-involved individuals are more likely to donate fits squarely with results by Putnam (2000) and Piliavin and Charng 215

PAGE 229

( 1990) who have also noted the positive relationship between these two variables. Putnam (2000) places strong emphasis on organizational involvements as the main component of social capital. He has noted a decline in organizational involvements and organizational involvements have been tied to higher rates of other voluntary behaviors, including blood donation. Organizational involvements alone, however, may be an impoverished measure of social capital. Depending upon how skeptical one is about the usefulness of this measure, this dissertation may or may not have sufficient evidence to support a link between social capital and blood donation. In either case, we can support the relationship between organizational involvements and blood donationand it is that link which will be discussed in more detail. The practical implication of the link between organizational involvements and the donation rate is that the practice of partnering with organizations for blood drives is likely to be highly effective. Individuals in these surveys did not indicate that the sponsor of the blood drive was an important determinant of their decision to donate blood. However, organizational members are more likely donors. Hence, a benefit of targeting organizations, besides administrative ease, is that it provides blood banks with access to a group of individuals with a greater willingness to donate blood. 216

PAGE 230

Summary In summary, we have mixed support for our original hypotheses. The relationship between costs and benefits of donation is as hypothesized. When costs and benefits are considered independently, individuals donate more when the cost is low and when the benefit is high. When costs and benefits are considered together, we find that when the benefit ofblood donation is perceived as being low, individuals adjudicate their donation decision based on cost. Hence, cost functions as a gate in low benefit situations. However, when the benefits of donation are perceived to be high, the impact of cost declines, and no longer differentiates likely and unlikely donors (see Figure 6.1). The implication findings regarding the costs and benefits of donation is that decisions to donate blood are contextual. In terms of practical implications, different recruitment strategies are likely to be differently effective at different times. Incentives are designed to increase the perceived benefit of donation for potential donors. In this thesis, altruistic incentives appear as influential as self interested incentives. The donation to charity was more appealing than the t-shirt, but less appealing than the lottery ticket for a CD-player. The implication of these findings is that different individuals appear to be responding to different incentives. Hence, recruitment strategies should appeal to a multiplicity of incentives to capture the motivations of the various groups of potential donors. 217

PAGE 231

The results regarding the importance of perceived social capital for the donation decision are mixed. There is insufficient support for the social capital hypotheses regarding trust and community connectedness. However, we did find that individuals that spend more time with organizations were also more likely to donate blood. Hence, more research needs to be conducted to uncover the mechanism by which organizational involvements impact willingness to donate. These results could provide encouragement for continuing the old policy of having blood banks partner with organizations to conduct blood drives, as organizational members are more likely donors. Limitations ofthe Study This dissertation involved the use of policy-capturing techniques on two groups of potential blood donors. The nature of the study required assumptions regarding the comparability of two discreet populations, the comparability of stated intentions and actual behavior, and the selection, by the researcher, of factors to be included in the analyses. The findings should be evaluated in light ofthese limitations. The surveys were administered to two populations at two different time periods. While there are no obviously important differences between the populations, we cannot dismiss the possibility that there was some unobserved 218

PAGE 232

variable that accounted for the difference in results, other than exposure to the September 11th bombing (e.g., timing during the academic year). Some of the limitations of this study relate to issues of external and internal validity. See page 155 for a discussion of the role of internal and external validity in idiographic analyses. The internal validity of this dissertation could be suspect if, for instance, subjects really were not familiar with blood donation and hence, policy capturing might have been inappropriately used. Our analysis of task familiarity indicates that subjects reported high levels of remembering previous blood drives and remembering their rationale for either donating or not donating. However, we can not totally dismiss the possibility that subjects' reporting of these findings are not accurate, or that they insufficiently capture the concept of task familiarity. In addition there are several potential limitations regarding issues of external validity. First, there might be insufficient correspondence between individual's stated willingness to donate blood and actual behavior. Although previous research (Lee, Piliavin, and Call, 1999) indicated that stated willingness has been found to be a reasonable proxy, we can not dismiss the concern. The only alternative, then, would be to conduct a full-scale intervention where the characteristics of a blood drive were systematically altered, with donors and non donors being surveyed for each manipulation. The cost and complexity of such an 219

PAGE 233

intervention are beyond the scope of this dissertation, and hence, in the end, reliance on stated willingness is the best we can do. The second potential limitation from external validity relies upon the use of the linear model as being predictive of human judgments. Once again, previous research (Brehmer and Brehmer, 1988; Caroll and Johnson, 1990) has been supportive ofthe use oflinearmodels as being "paramorphic." That is, we make no claim that individuals make decisions using a linear model; however, the outcome of human decision making and the results oflinear models have been found to be agreeable (Cooksey, 1996). Nonetheless, we recognize that this is an assumption that is presumed but not verified in this research. The final concern regarding external validity regards the use of a single, non-randomly selected population. The population was drawn from students at a prestigious university. These findings may not be generalizable to other populations. It should be noted, however, that the next generation of donors is a group that keenly interests blood donation recruiters, and hence, studies should focus specifically on this population. Another potential limitation is that the dimensions in the policy capturing exercise were constructed based on a review of previous literature, unstructured interviews, and pre-testing of the survey instrument. However, it is possible that important elements have been omitted, or inappropriately operationalized. To the best of our ability, we have verified the results of the policy capturing exercise with 220

PAGE 234

questions from the second portion of the survey. Nonetheless, further investigation might reveal important dimensions that can supplement or augment the dimensions presented here. In summary, there are some limitations of this study. Nonetheless, it represents a solid and unique inquiry into the structure of decision making of potential donors; while the results need additional verification, they should not be unduly discounted. Theoretical Conclusions Several theoretical conclusions can be reached by considering this dissertation's findings. The conclusions coincide with the four literature areas addressed in Chapter Three: the rational actor model, altruism, social capital, and policy design. One of the major findings of this dissertation is the complex relationship that exists among some of these literature areas. Hence, interactions between the literature areas will also be discussed. Several theoretical insights regarding the rational actor model of human behavior can be derived from the findings of this dissertation. The most important of these findings is the relationship between altruism and self-interest. The research clearly supports the notions that individuals are responding to multiple motivations, and that responses are contextual. 221

PAGE 235

Rational choice theory presumes that all (relevant) behavior is ultimately self-interested. Hence, all motivations must fit within the rational choice framework; there is no recognition of non-self-interested motivations. Subjects from this study demonstrated that both altruistic and self-interested motives were driving their donation decision. When the perceived benefit of pro-social behavior was high, the relevance of costs was low. Alternatively, when the benefit was perceived low, the relevance of costs is high. This indicates that individuals were responding to both egoistic and altruistic concerns. Moreover, this indicates that there was a complex interaction between these motives that changed in light of contextual factors (in this case, the September 11th bombing). Other theorists (e.g., Elster, 1990; Sen, 1990; and Mansbridge, 1990) have noted the relationship between altruism and egoism. However, the specific nature of the relationship between these two motives has not been identified, nor has the contextual nature of the relationship. The contribution of the present research is the offering of a model relating the two motives (Figure 6.1 ), and the appreciation of the contextual nature of the interaction. Moreover, these findings support the observation of Holmes ( 1991) that rational choice theory strives for parsimony at the expense of richness-a trade-off that he concludes is ultimately unsatisfactory. Our review of the literature on rational choice theory also included consideration of the literature on social dilemmas and risk. Jane Mansbridge (1990) had noted that the classic resolutions to social dilemmas relied upon various appeals 222

PAGE 236

to self interest. She argued that other resolutions relying upon alternate motives exist. This dissertation has demonstrated one arena where a non-egoistic resolution to a social dilemma can be effective. Blood donation can be construed as a social dilemma, the resolution of which need not rely upon appeal to self-interested motives -providing practical backing to this theoretical supposition. Dawes, Van De Kragt and Orb ell ( 1997) had conducted game-theoretic research to identify the factors that would _increase cooperation in social dilemmas. They found group identity to be a highly effective. It is likely that the September 11th bombing increased group identity for many Americans. The idea of highly salient national emergencies increasing group identity was presented by Putnam (2000, 2002) in a discussion of World War II's impact. Putnam discussed the feeling of "we" that can emerge from feelings of external threat. Hence, this provides us with a potential link between some disparate literature areas, namely social capital and social dilemmas. Social dilemmas are just the sorts of situations that social capital would be hypothesized to help citizens overcome. One mechanism by which this might occur is through feelings of civic or group identity. An alternate mechanism that might increase cooperation in social dilemmas, suggested by this research, is situations in which the benefit is perceived to be high. It should be noted that this is not benefit to oneself, but benefit to others. Subjects in this study were more willing to engage in an altruistic act and to assume personal costs when the benefit to others was perceived to be high. Hence, increased 223

PAGE 237

understanding or appreciation for the benefits resulting from coordinated action might increase the chances for cooperation in social dilemmas. This represents a contribution to our understanding of the factors that facilitate resolution of social dilemmas. It also indicates that Mansbridge (1990) is correct in her assessment that resolutions to social dilemmas solely relying on self-interest may be overly narrow. More practical applications of the work from social dilemmas should be considered, as it might lead the way to resolution of intractable policy dilemmas. From the literature on risk, we noted there are often discrepancies between experts' and lay-persons' estimation of risk. While this dissertation did not explicitly compare perceptions of risk, the high rate of concerns over acquiring communicable diseases via donation hints that this may be another case where there is adivergence in perceptions of risk (Slovic et. al., 1980). One theoretical insight is the interaction between perceptions of risk and perceived benefit. There was no difference in perceptions of risk between the subjects on Study I versus Study II. However, subjects on Study II indicated that they were more likely to donate blood; hence, a willingness to accept risk appears to be related to the perceived benefit. This does not call into question the existing literature on perceptions of risk. However, it does suggest an extension of the literature that encompasses the joint consideration of the willingness to accept risk and perceived benefit. A more thorough modeling of the relationship between willingness to accept risk and 224

PAGE 238

perceptions of benefits would provide the necessary stepping stone for more thorough applied research in this area. One of the major theoretical issues plaguing the understanding of altruism has been the ongoing debate about whether altruism can somehow be reduced to self-interest. Although this dissertation cannot hope to resolve this long-standing debate, consideration ofhow egoistic, pseudo-altruistic and pure-altruistic accounts of altruism explain the findings is instructive. Pure egoism would argue that blood donations were high after the bombing because individuals were benefiting, perhaps because of the incentives, or for issues of reputation or self-image. Pseudo-altruistic theories, such as the arousal-reduction theory advocated by Piliavin (1991) would contend that individuals donated to reduce their own discomfort caused by having others suffer. Pure altruists would argue that individuals donated not for the benefit they received themselves but because of the benefits others would receive. Consideration of these findings in light of the very high levels of actual donation after the September 11th bombings is especially illuminating. Given the high rates of reported likelihood of donating blood in this dissertation, it seems that potential donors would have to have inordinately high levels of guilt to account for the arousal-reduction theory of altruism and would have to be rather (unnaturally) inclined towards t-shirts to account for the egoistic account of altruism. In the end, the voluntary actions of citizens in response to September 11th are difficult to account for in the egoistic or pseudo-altruistic accounts. John Elster ( 1990, 46) 225

PAGE 239

describes blood donation as unambiguously unselfish, since the donation goes from an unknown donor to an unknown recipient. Elster thus claims that this reduces the likelihood that an egoistic motive, such as prestige, is responsible for the action. Hence, in this instance, we are left more sympathetic to "pure altruism." Individuals assumed substantial costs at actual blood drives, and in this study, indicated that they would be willing to assume large costs, when the perceived benefit is high. It is difficult to explain these findings by anything other than a pure-altruistic account. Theoretical insights can also be gained from a consideration of the third literature area, social capital. The impact of exposure to a nationally unifying event has received some attention from Putnam (2000) in his discussion of World War II's effect. Putnam (2002) has also conducted some research aimed at assessing the impact of the September 11th bombings on social capital. He found higher levels of trust, 2 confidence in government, political participation, voluntary activity and a greater sense of''we" after September 11th (Putnam, 2002). Putnam notes that the feeling of"we" can be transitory, and that the challenge for civic renewal is to transform the change in attitudes into a change in behavior (Putnam, 2002). In this dissertation, we found that exposure to September 11th had a unique impact on individual's decision-making regarding blood donation. A comparison of finings from the two surveys revealed that subjects on the second survey were more 2 Putnam (2002) was using an extensive battery of questions regarding trust, as opposed to the single, simplified question used in this dissertation Future research should ascertain the correspondence between the various measures of trust. 226

PAGE 240

concerned about the need for blood, helping others, and were also more willing to donate blood. Hence, "something" about the experience ofbeing exposed to a national tragedy impacted the decision making process regarding a voluntary action. Keeping in mind our caveats regarding the relationship between social capital and organizational involvements, this dissertation provides tentative support for the theoretical contention that national tragedies boost social capital and, in turn, voluntary behavior. The final literature area considered in this dissertation is policy design. Two general theoretical recommendations for the policy design literature can be derived. The first finding relates to the insufficient attention within policy design to the psychology of targets of public policy. The second implication revolves around the near exclusive reliance on self interest as the model for human behavior. The findings from this study suggest that our theories of policy design should be augmented to include an appreciation and understanding for the complexities of the decision-making processes of the targets of public policy. The current literature on policy design, particularly literature with a post-positivistic bent, encourages the consideration of the targets of policies (e.g., Bobrow and Dryzek 1987; Schneider and Ingram, 1997). However, these prescriptions rely on a nearly unidimensional reading of the clientele. While there is recognition that policy targets may represent different constituencies, and hence different priorities, there is little consideration of the potential complexity of policy targets. For 227

PAGE 241

instance, there is no recognition that policy targets might be responding to multiple motivations. Concomitantly, there is little recognition that policy targets representing the same constituency might be reasoning about the same issue but in significantly different ways. In short, current policy design appreciates the importance of the targets of public policy but underestimates the complexities of these targets. More attention to the psychology of the targets of public policy should improve the outcomes of policy design. JDM research techniques provide one useful vehicle for exploring the decision making styles of policy targets. JDM research regularly considers policy issues. In fact, the use of JDM techniques in medical decision making is becoming rather commonplace. However, it is not a literature nor a way of thinking with which most graduates from most policy analytic schools are familiar. It should be noted that policy capturing is one of many analytic methods used in JDM. A broader consideration of other methods available to policy analysts would be helpful. One particularly useful recommendation for policy design, derived from JDM research is use of the idiographic (within-individual) level of analysis. Considering individual decision makers indepth is not a technique commonly used in policy analysis or design, except perhaps in qualitative analysis. The idiographic level of analysis encouraged by Cooksey ( 1996) permits a high level of understanding of an individual decision maker in a systematic, empirical manner. 228

PAGE 242

The final theoretical finding from this dissertation relates to the reliance upon self-interest as a model ofhuman behavior within policy design. Despite the numerous and rather persuasive critiques of rational choice theory as the basis for public policy, the theory maintains a large and committed following in some circles (e.g., political science) (see Schneider and Ingram, 1997; Petraea, 1991). There are several problems inherent in the near-exclusive reliance on rational choice theory as the epistemological bedrock of policy design. First, the rational choice theory marginalizes non-self-interested behavior and in tum legitimates self-interested behavior. Second, the theory discourages the consideration of other-oriented values within policy design. Third, institutions built on rational choice assumptions provide incentives for self-interested behavior and do not allow room for the expression of other-focused behavior. Finally, a society built on such institutions is ultimately rejecting the motives that are essential for provision of the public good and, as Schneider and Ingram argue, the smooth functioning of democracy. The first concern with the reliance on rational choice theory is the marginalization of pro-social behavior in favor of self-interested behavior. Schneider and Ingram summarize that "[t]he standard public choice theory labels public-spirited behavior, whether by citizens or public officials as 'irrational'" (Kelman 1987; Reich 1991)" (Schneider, and Ingram, 1997, 50) The second concern is that rational choice theory discourages the consideration of values other than efficiency, as noted by Amitai Etzioni in The 229

PAGE 243

Moral Dimension (1988). Peter deLeon points out that, in contrast to rational man arguments, "values cannot be assumed away under an utilitarian umbrella or Pareto optimality" ( deLeon, 1997, 93 ). Policies for encouraging blood donation are a prime example of the importance of considering values beyond efficiency. As Deborah Stone (1997) notes, safety, equity and other-mindedness are all important values in this context, and policies designed in the absence of their consideration will be found wanting. The third concern is that institutions built on assumptions of rational choice encourage the expression of self-interested behaviors and discourage the expression of other-interested behaviors. Schneider and Ingram summarize: The theory of public choice legitimates self-interested behavior as if such actions are "natural" and therefore inevitable in the political life of a society. When institutions and public policies are designed as if self-interest is the motivating factor, then such designs not only legitimate this behavior, but actually encourage and produce it. (Schneider and Ingram, 1997, 50). Shneider and Ingram go on to contend that such institutions built on self-interest do harm to the individuals with the proclivity to respond to alternative motivations. Institutions that presume self-interest are likely to produce such motivations, as the institutions legitimate self-seeking behavior and, furthermore, seriously disadvantage anyone who attempts to engage in more cooperative or collaborative efforts (Petracca, 1991 ). (Schneider and Ingram, 1997, 50). Richard Titmuss expressed similar concerns regarding the assumptions upon which public policies are built in the The Gift Relationship ( 1971 ). He lamented that if institutions by and large do not permit --much less encourage -the 230

PAGE 244

expression of motives other than self interested ones, they do harm not only to their cause, but to society at large. Schneider and Ingram continue, "[i]t is not likely that societies can survive if self-interest overwhelms collective considerations ('longterm' self-interest, as some public choice scholars call it)" (1997, 50). Commentators on civic life, beginning with de Tocqueville (1840/1984), have noted the high reliance on voluntary organizations in the United States is predicated on the expression of other interested behavior. A final concern is that if, as a society, we accept the rational actor model of human behavior, and hence build our institutions based on the premises of rational choice theory, then we do damage not only to our institutions, but to society at large. Schneider and Ingram, citing Petraea, raise concerns over how we, as a society, can pursue the public good in the absence of other motivated behaviors. "Petraea, ( 1991, p311) makes a compelling argument that public choice theory thwarts the 'transformation of self-regarding individuals into public-spirited citizens' and also fails to transform private-regarding political regimes into those that can achieve the public good" (Schneider, and Ingram, 1997, 50) Schneider and Ingram offer a leveling indictment of self-interest as the basis for public policy since these designs encourage self-interested motivations at the expense of good citizenship, social capital and ultimately a smooth running democracy. Many designs reinforce the self-interest motivations in U.S. politics and signal that people are expected to look after their own interest (with little regard for the elusive "public interest"), and everyone is expected to cut the 231

PAGE 245

best deal they can for themselves (sic.). Policy so designed creates a particular culture that permeates democratic institutions and has far-reaching negative consequences for justice .... The empathetic attitudes and willingness to compromise that are so essential for a working democracy and for justice may or may not have been learned through citizen interaction with policy and citizen interpretations of the meaning of policy. Excessive emphasis on self-interest can undermine notions of the collective good and result in political arenas without sufficient ethical grounding to create trust and social capital. When trust and social capital are destroyed, democracy becomes difficult or impossible (Putnam, 1993) (Schneider and Ingram, 1997, 50). The rational actor model of human behavior has been the foundation for much modem public policy, including blood donation policy. Despite the concerns over the reliance on the rational actor model, policy recommendations steeped in self-interest are being proposed to correct the disturbingly low levels ofblood donorship. American democracy will not crumble if we allow payment for donation, but the pattern of institutionalizing self-interest at the expense of more laudatory motives will, ultimately, be damaging. Implications for Blood Policy This dissertation clearly has a number of policy-pertinent implications for how blood should be solicited from potential donors. This dissertation does not address issues beyond the solicitation of donors. Research on blood distribution mechanism will be set aside for future research. The implications are briefly suggested below. 232

PAGE 246

As we saw after September 11th, individuals are more than willing to assume considerable costs to themselves when the perceived benefits are high. Hence, more thought should be given to what factors increase the sense of benefit to potential donors in non-emergency times. This presumes that Putnam (2000, 2002) is correct that the heightened sense of"we" is transitory. In order to recapture the feeling of "we" that seems to be enabling higher levels of civic participation the recommendation is not to create a feeling of artificial crisis. However, it is likely that the low donation rate might partially stem from a general lack of understanding of the need for blood. One solution might be to word solicitations for blood donation in proactive, as opposed to reactive, language. This would entail informing potential donors of the need to have sufficient blood in reserve in case of another large scale emergency. Along these lines, the American Blood Council (ABC), a non-profit membership organization for blood banks, is advocating and has implemented a policy (ABC, 2001) designed to deal with low levels of donation in times of national crisis. The ABC is requesting that individuals sign up for a blood-registry that could be activated in times of need. The American Association of Blood Banks has also adopted a similar policy. These policies are not aimed at increasing the blood donor level in non-crisis times but could be useful for certain unanticipated emergencies. This policy has the added benefit of raising the public consciousness about the potential for needing blood in the event of another crisis. 233

PAGE 247

From these analyses, we saw that incentives that appealed to altruism were as effective as incentives that appeal to self-interest. Blood banks can pursue alternative incentive structUres, designed to appeal to alternative motivations beyond self-interest. This has several important benefits. First, since there are groups of individuals responding to a range of motivations beyond self-interest, then strategies aimed at appealing to alternate motivations might be more effective than strategies predominantly aimed at one, (largely) self-interest. Second, as Titmuss ( 1971) noted, incentives that are more nearly pecuniary run the danger of encouraging donation from individuals with more risky blood.3 Incentives that appeal to altruism are less readily converted to cash, and are probably less likely to attract at-risk donors. Third, and as discussed in greater detail above, as a society we should leave room for the expression of civic and other-focused virtues, so that these virtues can be nurtured and encouraged. Blood donation provides an ideal venue for the expression of altruism, as blood from donors responding to incentives is likely to be of pdorer quality than the blood of an altruistic donor A third recommendation is to address the amount of distrust and misinformation surrounding blood donation. There are a substantial number of individuals, even among highly-educated college students, who are not confident that AIDS or other diseases cannot be acquired through the act of donating blood. As Slavic (2000) pointed out, perceptions of risk are a function of both the actual 3 To the best of our knowledge, the relationship between incentives and risks has not been empirically tested. 234

PAGE 248

risk and levels of trust. From the recognition of the relationship between risk and trust, two suggestions emerge. First, it is necessary to provide more information on the improbability of acquiring diseases through donation this addresses the risk perception problem. The second suggestion is that steps need to be taken to improve perceptions of the trustworthiness ofblood banks and their personnel.4 From a public health orientation, one of the disturbing results from this analysis is the finding that a test for communicable diseases was an incentive for donation.5 Blood donation is obviously not the appropriate venue for individuals concerned about their communicable disease status to seek testing. One of the goals of blood policy is to acquire safe blood. High risk individuals, motivated by the disease screens, place the blood supply in unnecessary jeopardy, as tests for communicable diseases are not perfect. This highlights the importance of providing additional information to potential donors on where they can be tested in a safe and confidential manner. Current policies of carefully screening blood for disease coupled with asking individuals if their blood should be discarded are appropriate, since high risk individuals, perhaps motivated by the test or donating because of social pressure, might be willing to indicate that their blood is unsafe. An additional step might be the regular offering of information on where interested donors can go to receive appropriate testing. 4 Later, in a discussion of additional research suggested by this inquiry a systematic investigation into the perceptions of blood banks and corresponding levels of trust are recommended 5 Inclusion of this item on the survey came at the recommendation of a pre-test subject. 235

PAGE 249

One final set of recommendations can be derived from consideration ofthe clusters of types of decision makers. The factor analysis and subsequent clustering of subjects revealed four distinct clusters of decision makers, each responding to different motivations. It should be noted that blood recruiters do recognize the need for clustering ofblood donors (AABB, 2000). However, the clusters are often socio demographic in nature. This dissertation has demonstrated that motivational differences provide more useful clustering criteria anci that specific recruitment strategies could be derived that would target the various clusters. The clusters derived from the factor analysis will be briefly re-introduced. Specific recruitment strategies for the various clusters would be best derived from marketing studies. However, some preliminary recruitment suggestions are offered. The first cluster, termed civic minded, was distinguished by the emphasis on distance to the blood drive as a determinant of the donation decision. Besides distance, this cluster is highly influenced by the perceived good deed of donating. This cluster appears uniquely "civic-minded" as compared to the other clusters, given their high rate of other charitable donations. Since the civic-minded group engages in a variety of voluntary activities, including donation, stressing the civic/voluntary virtues of donation might be an effective trigger for donation. The second cluster, termed self-interested, was mostly influenced by the lottery offering. It is also distinguished by the focus on self-interest, or, more precisely by the lack of motivation by anything other than self-interest. It should be 236

PAGE 250

noted that it is the group most likely to have donated blood. The self-interested group donates, in large part, for the incentives received from donation. Hence, the current recruitment strategy of stressing incentives, such as t-shirts and coffee mugs, would likely be effective. In addition, this group is very concerned about issues such as time and convenience. Addressing the "ease" of donation is likely to be appealing for this group. The third cluster, the worriers, is most influenced by the offering to charity and by non-self-interested motives, such as the societal need for blood. This group has high levels of perceived risk of donation, and a very high rate consider themselves to be ineligible. Hence, doing "good" is a significant motivator for this group, but the perceived cost of donation is very high, given their high levels of perceived risk. The major impediment to donation for the worriers is perceptions of risk. Educational campaigns aimed at reducing risk estimations would be helpful. It would also be useful to address fear of needles. The donation to charity was important for this group. However, given their high perceived cost of donation, the most promising first step towards encouraging donation is to address the perceptions of donation-associated risks. The final cluster, the principled donors, is distinguished by its emphasis on doing a good deed as a motivator for their donation decision. They are strongly influenced by the donation to charity and not by the lottery or the t-shirt options. These principled donors are the group that would likely be most repelled by the 237

PAGE 251

thought of payment for donation. An effective recruitment strategy for this group would focus on donation as a good deed and on the additional good, in the form of a charitable donation, that can result from their donation. In addition, testimonials from previous recipients of blood might be motivational for this cluster. These disparate donation groups appear to be responding to different motives, and hence unitary recruitment strategies would likely find inconsistent results amongst the groups. For instance, offering a lottery ticket and focusing on self-benefit as the basis for a recruitment effort would likely help motivate the self interesteq donors, but might easily repel the principled donors. Likewise, while information campaigns aimed at reducing perceptions of risk would likely be very useful for the worriers, it would likely not significantly improve donor rates in the other clusters. Two clusters-the worriers and the principled donors--are largely influenced by the offer of a donation to charity. This is important, as these two groups combined represent nearly fifty percent of the survey population. Offering donors altruistic incentives might be very helpful for recruiting potential donors from these two clusters. It also seems likely that the civic-minded group could be encouraged to respond to the altruistic incentives. It is possible, however, that the altruistic incentives would have little appeal for the self-interested groups. One strategy might be presenting donors with a choice of a self-interested or altruistic incentive as a reward for donation. 238

PAGE 252

In summary, the advantage of considering potential donors by cluster is that it demonstrates that potential donors are using different decision making policies. Hence, recruitment strategies solely aimed at one motive might miss large portions of the potential donor population. Further Research From this inquiry into the motivations for donation, several areas for further research are suggested. The research proposed is all of a practical nature, designed to provide more insights into remaining gaps in our understanding of the decision making processes that accompany blood donation. First, research (perhaps a marketing study) should be conducted into the appeal of various incentives, as well as the effect of incentives on both the donor rate and the quality ofblood. Such research should focus on the relative appeals of various incentives to various groups of donors. This could entail, for instance, having subjects rank order thevalue of various offered incentives as well as an in depth inquiry into perceptions of incentives. This research should include consideration of standardly offered incentives designed to appeal to self-interested motives (e.g. cups, mugs, lotteries for electronic gadgets) of potential donors as well as incentives designed to appeal to other motivations. Consideration of incentives, such as donations to charity, that appeal to altruistic motives, are likely to be of equivalent effectiveness in recruiting new donors and may tap into a group of 239

PAGE 253

donors that had been ineffectively targeted previously. Moreover, the use of these altruistic incentives would not raise concerns over the quality ofblood donated. If this research proves veridical, then further inquiry should be conducted that investigates various methods to appeal to altruistic motives. This research would have the additional benefit of moving incentives farther away from pecuniary rewards, which has obvious benefits for the safety ofthe blood supply. Large percentages of individuals in this study believe that blood banks are profiting from the sale of blood and that it is possible to acquire diseases through donating. Moreover, media reports of profiteering, competition and corruption may have introduced some justifiable skepticism into our common understanding of the reputability ofblood banks (e.g., Gaul, 1989). These findings, considered together, might indicate a lack of trust in banks.6 An understanding of the source of these concerns might prove useful for designing communications aimed at improving the perception of the trustworthiness and competence ofblood banks and their personnel -if indeed, it is these perceptions that are the root cause of the beliefs regarding profits and potential for disease. Third, there is a robust literature on risk communications. The applicability of this research to risk perceptions associated with blood donation has not, to our knowledge, been investigated. Given the high levels of perceived risk reported by respondents, studies diligently investigating the nature of the perceived risks are 6 A problem reported in previous research, e.g., Kohr and Sayers, 2000. 240

PAGE 254

necessary so that proper risk communications regarding blood donations can be developed. Extending their idea that low levels of trust can exacerbate perceptions of risk, perceptions of the high risks ofblood donation cannot be overcome without simultaneously addressing the issues oflack of trust in blood banks and their proxies. Further research into the relationship between risk and trust would, hence, be useful. Fourth, the current research on social capital laments the difficulty of operationalizing some of the key concepts related to social capital (Putnam, 2000). More research into identifying measures of social capital related concepts -such as trust and community connection --would be useful. Fifth, there are two shortcomings of previous research into motives for blood donation that should be redressed by future research. First, there has been little contemporary research that compares cognitive and behavioral differences between donors and non-donors; most of the research has considered differences between first-time and repeat donors. As the need for new donors becomes more acute, populationbased studies designed to identify motivational differences between donors and non-donors would improve blood recruiter's ability to propel non-donors to donate for the first time. Moreover, this research has demonstrated that the lack of correspondence between socio-demographic and behavioral motivations. Hence, methods for determining appropriate recruitment strategies for different populations are essential. 241

PAGE 255

An additional suggestion for further research involves the descriptive nature of the few studies that do compare donors to non-donors. The insignificance of the socio-demographic variables as predictors of either previous or likely donation lends credence to Piliavin's recommendation that research on blood donors must go beyond the merely descriptive (1998). Research prior to Piliavin's often focused on socio-demographic differences between donors and non-donors. While they might exist, they do not appear to be the driving force behind differences in blood donation predilections. This dissertation identified few, if any, relevant differences. between donors and non-donors. In the present inquiry, it is true that variation between groups was limited by the venue (college undergraduates). However, this is the very age cohort blood banks need to learn about to redress the chronically low blood donation rate. Hence, future research should consider motivational differences between donors and non-donors, and move beyond socio-demographic comparisons. This research should also include the consideration of methods for identifying individuals from the different judgmental clusters. A final area of further research involves the extremely high percentage of survey respondents that reported they were currently ineligible to give blood. This corresponds to Piliavin's (1998) observation that potential donors deferred from donation are very less likely to consider donating again in the future-even though they may no longer be ineligible. It appears that this belief of ineligibility is quite resilient to change. Hence, further research into identifying what types of 242

PAGE 256

information is necessary to encourage those that were previously ineligible to attempt to donate again would be valuable. Implications for Other Voluntary Medical Donations In this dissertation, we have investigated the motivational structure of decision making for potential blood donors. One final question is if these results can be applied to other sorts of voluntary behaviors, including other medical donations or donations of money and time. The relationship between donating time, money and blood has been investigated by Lee, Piliavin and Call (1999), who analyzed data on the relevant motivations. They then built a model of blood donor behavior and tested to see how well their model predicted the donation of time and money, and concluded that their model of"blood donor identity and donation can be generalized to charitable donation and volunteering" (Lee, Piliavin, and Call, 286, 1999). Hence, whether or not the results of this dissertation are useful for improving other sorts of voluntary behaviors requires separate investigation. In particular, these findings should have relevance for other voluntary medical donations. There is an amazing array of medical donations currently solicited. Some of these donations are for payment (e.g. sperm, eggs), others are not (e.g. milk, organs). For the types of donation that are currently non-reimbursed, the message is an optimistic one. Focusing on the salutary aspects of donation should 243

PAGE 257

go far to improve donor rates. However, shifting the focus onto increased benefit for the donors is likely to be less effective. For the donations that are reimbursed, the recommendation is not as clear. Finally, it is important to point out that there are advocates for returning to a system of payment for blood donation (Roberts and Wolkoff, 1988; Hernandez, 2001 ), as well as payment for organ donation. Concerns over adverse selection should dissuade proponents of these strategies. This is not to suggest that payment for donation would result in the full extent of the disastrous effects suggested by Titmuss. However, it is likely that Titmuss got part of the story right. Payment for donation has had negative effects on the safety of blood and will likely have similar effects for the safety of other medical donations because individuals motivated by payment are more likely to engage in risky behaviors. Requests for donations that resonate with feelings of doing a good deed, civic duty, and that make clear the need for the medical donation can go far towards improving donation rates. In addition, indicating that we should be proactive as opposed to reactive in our solicitations for medical donations is likely to be a rather effective strategy. Conclusion Over the last 40 years the blood donation rate has been steadily declining. Recent horrific events have resulted in an upswing in donation, and, in this research, an upswing in individuals' stated willingness to donate blood. This dissertation 244

PAGE 258

represented a multi-disciplinary multi-method approach to shed greater light on the motivations for blood donation. We can conclude that explanation for the persistently low donation rate in the United States lies, in part, with the design of our blood donation policies. The Red Cross and the AABB have the stated policy of relying on altruistic motivations for collecting blood. Their policies, however, send a mixed message Individuals are asked to volunteer for altruistic reasons, but the consistent provision of incentives gives mixed messages and may even be counterproductive (Gutherie and Smith, 1998). In addition, there has been increasing movement towards reliance on more substantial incentives. And, in a desperate move to deal with the chronically low donation rate, there have been calls for a return to payment for donation. The focus on self-interested motivations, as discussed previously, entails several unfavorable consequences. Consideration of alternate motivations may yield a variety of benefits, including an increase in the blood donation rate and an improvement of blood quality. In this dissertation we have shown that only a portion of the population responds to self-interested incentives. Providing potential donors with a range of incentives designed to appeal to a range of motives could increase rates of donation. Different groups of donors are sensitive to different motivations. Hence, a recruitment strategy aimed at only one motivation will miss segments of the 245

PAGE 259

potential donor population Moreover, donor's motivations are contextual. Recruitment strategies need to be sensitive to exogenous factors that might change the decision making calculus of potential donors. We also found that more organizationally involved individuals were more likely to donate blood, providing tentative support for a relationship between social capital and blood donation. Moreover, this research hints that the impact of national tragedy is as Putnam (2000) hypothesized_. A sense of "we" is created that facilitates civic engagement in a unique way. Individuals were more willing to donate after the bombing and were more concerned over the need for blood. Hence, determining how national tragedies alter our decision calculus can help improve civic involvement in less traumatic times. In sum, the various groups of donors were responding to different motivations. Recruitment strategies should recognize that sociodemographic breakdowns of the population are not as meaningful as breakdowns by motivation. Moreover, policy designers need to recognize the complex, contextual nature of the decision making policies of policy targets. When policy designers or blood recruiters are remiss in this recognition, policies are less effective, and blood donation rates are lower than they need be. If policy designers assume that citizens can be noble and good, and design policies around those assumptions then perhaps these motives can be nurtured. Alternatively, policy designers can continue to design policies around assumptions 246

PAGE 260

of self-interest, and they will be correct-individuals will in self-interested ways. As policy designers, and as a society, we need to be mindful of the assumptions embedded in public policies, for they will have long-term consequences. Richard Titmuss ( 1971) noted the dangers of building a policy on self interest. His conclusion was that the gift relationship could best be nurtured through a reliance on altruism. This dissertation supports the idea that motivations beyond self-interest should be encouraged. However, this does not imply an exclusive reliance on altruism. We have shown that policy targets are complex and that their motivations may include, but need not be limited to, self interest, or even altruism. 247

PAGE 261

APPENDIX A Summary of Policy Capturing Instrument The purpose of this study is to provide information regarding under what conditions individuals are more likely to donate blood. Imagine that there will be a blood drive on campus. On each attached piece ofpaper.in this stack you will find a description of the blood drive. Please read the description carefully and then decide how likely you are to donate blood in each situation. Assume that you will be eligible to donate blood. Consider the following blood drive and decide how likely you think you are to donate at the blood drive described. Mark your response on the scale at the bottom of the page. Charity 1. 2. Lottery 1. 2. Location l. 2. 3. For donating blood you will receive a commemorative t-shirt. For donating blood an anonymous $7 donation will go to a charity of your choice. For donating blood you will receive a commemorative t-shirt. For donating blood your name will be entered into a drawing for various items including a portable compact disc player. The blood drive is on your way to class. The blood drive is a ten-minute walk across campus. The blood drive is fifteen minutes away from campus by bus. 248

PAGE 262

Need Good-deed Sponsor 1. The blood drive is beir).g conducted because there is currently a shortage of blood in the local area. 2. The blood drive is being conducted because there has been a recent wellpublicized natural disaster and blood is needed. 3. The blood drive is being conducted because there is a shortage somewhere else in the country. 4 The blood drive is being conducted because there is always a need for blood. 1. A flier posted around campus stresses that donation is a good deed 2. A flier posted around campus reminds you that you will receive a gift for your donation 1. An organization you belong to is sponsoring the blood drive. 2. The blood drive is being sponsored by your university. 3. The blood is being conducted by a local blood bank. How likely do you think you are to donate blood at the blood drive described above? Not likely ------------------------------------------------Very likely 1 2 3 4 5 6 7 249

PAGE 263

APPENDIXB Section 2 of the Survey: Administered after the Policy Capturing Exercise Section 2. Motivations for and against donation Below is a list of common reasons people give for donating blood. Please indicate how important these reasons are for you when deciding if you will donate blood. 1. I want to help others. Not very important -------------------------------------------Very important 1 2 3 4 5 6 2. I like the t-shirts/mugs that they give out at blood drives. Not very important ------------------------------------------------Very important 1 2 3 4 5 6 3. I feel social pressure to donate blood. Not very important ------------------------------------------------Very important 1 2 3 4 5 6 4. I feel guilty ifl do not donate. Not very important ------------------------------------------------Very important 1 2 3 4 5 6 5. Someone asked me to donate blood. Not very important ------------------------------------------------Very important 1 2 3 4 5 6 250

PAGE 264

6. Free AIDS/STD test. Not very important ---------------------------------------------Very important 1 2 3 4 5 6 7. Understanding of the urgency of the need for blood (e g a natural disaster) Not very important------------------------------------------------Very important 1 2 3 4 5 6 If there are any reasons that you believe are important for your donation decision, please list them below, and indicate how important you believe they are for your decision to donate blood Feel free to use the back of the paper if you need more room. 8. Not very important---------------------------------------Very important 1 2 3 4 5 6 Below is a list of common reasons people give for NOT donating blood. Please indicate how important these reasons are for you when deciding if you will donate blood. 1. I don't have the time. Not very important--------------------------------------------Very important 1 2 3 4 5 6 2. Blood drives are not conveniently located. Not very important ----------------------------------------Very important 1 2 3 4 5 6 3 I haven't been asked to donate blood Not very important ---------------------------------------------Very important I 2 3 4 5 6 4 I don't like needles Not very important ----------------------------------------------Very important 1 2 3 4 5 6 5. I am afraid of getting diseases from donating blood Not very important -------------------------------------------Very important 1 2 3 4 5 6 251

PAGE 265

6. They ask too many personal questions Not very important ------------------------------------------------Very important 1 2 3 4 5 6 7. I think blood banks are making a profit off donated blood Not very important ------------------------------------------------Very important 1 2 3 4 5 6 If there are any reasons that you believe are important for your donation decision, please list them below, and indicate how important you believe they are for your decision to donate blood. Feel free to use the back of the paper if you need more room. 8. Not very important------------------------------------------------Very important 1 2 3 4 5 6 Section 3 0 Blood donation histmy 1. Have you ever donated blood before? No Yes If yes, how many times? __ How old were you the first time you donated blood? __ 2 Have you ever tried to donate blood, but were declared ineligible? No Yes 3. Do you believe you are currently eligible to donate blood? No_ Yes __ 4 Do any of your immediate family members regularly donate blood? No Yes What is the relationship of the family member(s) to you? (e.g. father? sister?) 5 Do any of your close friends regularly donate blood? No_ Yes __ 6. Have you or any members of your immediate family required blood in the past year? Yes No 7 How many blood drives would you estimate you have seen or heard of in the last year? __ 8. Please recall the last blood drive you remember seeing and answer the following questions If you cannot recall a recent drive, please go to question 8. 252

PAGE 266

8a. The blood drive was located: near work home school other 8b. Did you know about the blood drive before you saw it? Yes No 8c Would you consider the blood drive conveniently located? Yes_ No Ifno,whynot? ____________________________________ ___ 8d. Do you remember ever consciously to donate or not to donate blood at the blood drive? Yes No 8e. If you do remember your rationale please describe it below: 9 Did your high school have a blood drive? Yes_ No __ 9a. Did you donate at your high school's blood drive? Yes __ No __ 9b Do you remember deciding if you were going to or not going to donate blood? Yes No 9c. If you do remember your rationale, please describe it below : Please indicate how strongly you agree with the following statements. 10. When donating blood, it is likely that I could acquire health-related problems Agree -------------------------------------------------------------------Disagree 1 2 3 4 5 II. A person who donates blood runs the risk of contracting AIDS Agree ---------------------------------------------------------------------------Disagree I 2 3 4 5 12. It is risky to donate blood these days Agree ---------------------------------------------------------------------------Disagree I 2 3 4 5 253

PAGE 267

13. Needles make me nervous Agree 1 2 3 4 5 14. The sight of blood makes me uncomfortable. Agree ------------------------------------------------------------------Disagree 1 2 3 4 5 15. I quite enjoy taking risks. Agree ------------------------------------------------------------------------------Disagree 1 2 3 4 5 16. I am an adventurous person. Agree-----------------------------------------------------------------------------Disagree 1 2 3 4 5 17. I sometimes like doing things that are a bit frightening. Agree ----------------------------------------------------------------------:-------Disagree 1 2 3 4 5 Section 4. 1. Please list the organizations you currently belong to (e.g. clubs, voluntary organizations, religious organizations, etc.) and indicate approximately how many hours per week you spend in organization related activities. Organization Hours per week 2. Were you involved in student government in high school? Yes No 3. Are you involved in student government in college? Yes_ No __ 4. Have you donated any of the following to charitable organizations in the last year? Food Clothes Money Time Yes No 254

PAGE 268

5 Generally speaking, would you say that people can be trusted, or that in general you can never be too careful in dealing with people? People can be trusted Can't be too careful 1 2 3 4 5 6 7 6 Overall how "connected" do you feel to your community? Very Not at all 1 2 3 4 5 6 7 V. Socio-demographics 1 Year of birth 2 Male 3 Race : Caucasion (non-hispanic) African-American (non-hispanic) __ Hispanic Asian/ Pacific Islander Other 4 Number of years in college : __ 5 Major ____ 6 How many hours per week do you work for pay? __ 7 Would you describe your family as ? Low income Middle income Higher income 9 Highest level of education completed by a parent? Some high school Graduated high school Some college Graduated college Graduate degree 255 Female

PAGE 269

Soclo-Demographlcs Female gender % Age' Race While, not Hispanic Hispanic Asian/ Pacific Islander Other Years in college Donation History Donate previously % Previously ineligible, % Currently ineligible, % High school blood drive Motivators' Feelings of guilt T -shirt incentive Test for STDs Need for blood Want to he l p others Inhibitors' Convenience Fear of needles Amount of lime Fear disease Fear of getting AIDS Too many quest i ons Appendix C: Comparison by Previous and Likely Donation: By Survey ( selected variables) Survey I Survey II Non-Donors Donors P-value Unlikely Likely P-value NonDonors P-value Unl i kely Likely P-value Donors Donors Donors Donors Donors 56 75 54.55 0 8102 51 9 64 .29 0.1911 72 .97 54 9 0 .0366 54 55 73.91 0.0244 21 (20, 22)21.5 (20, 22) 0.4069 22 (20, 23) 21 (20, 22) 0.1973 19 (18, 20) 20 (19 21) 0 0425 9 (19, 21) 20 (18, 21) 0 .613 0 534 0.6322 0.7543 0.0833 36 .36 42.86 34 .62 46.51 36.49 43 14 38 18 39.13 13.64 7 14 12.82 6.98 6 76 9 8 1 .82 13 04 39.39 44 64 43.59 39. 53 50 41. 18 49.09 44.93 10 .61 5.36 8 .97 6.98 6 .76 5 .88 10 .91 2.9 3 (2, 4) 3(2, 4) 0 .4134 3 (2, 4) 3(1,4) 0.4548 2 (1, 3) 3(2,4) 0 0118 2 (1. 3) 2 (1, 3 5) 0 5124 37. 5 59.09 0 0208 32.73 46.38 0.1237 25 17.54 0.313 18.75 27.27 0 2712 39. 73 27.45 0 1576 23 64 44 12 0.0179 70 .59 84.21 0 0723 76.25 77 .27 0.8976 67.12 82 .35 0.059 87.27 61.76 0.0015 63 .08 82 14 0 .02 67.11 81.82 0 .082 86.49 86 0 9384 83.33 88.41 0.4186 2 .5(1,4) 2 (1, 3) 0 .5102 2(1, 3 .5) 2 .25(1,4) 0 5924 2 (1, 4) 3(2. 3 ) 0.5402 2 (1, 4) 2.5 (1, 3) 0.7144 2 (1, 3) 2 (1, 3 5) 0.4505 2 (1, 3) 2 (1, 4) 0 1001 2 (1, 3) 2(1,4) 0 7638 2(1, 3) 2 (1. 4) 0 .329 2 (1, 5) 2 (1' 5) 0 .8036 2 (1, 4) 4 (1' 6) 0.014 2(1,4) 1 (1' 3) 0.3802 2 (1, 4) 1 (1, 3) 0 1537 5 (4, 6) 5 (5, 6) 0.2949 5(4, 6) 6 (5 6) 0.0134 6 (5, 6) 6 (5, 6) 0.2661 6 (5, 6) 6 (5 6) 0.1852 5 (4, 6) 6 (5, 6) 0 .16 5 (4 6) 6 (5, 6) 0 0058 6 (5, 6) 6 (5 6) 0 6881 6 (5, 6) 6 (5, 6) 0 0388 5 (3, 5) 5 (3 6) 0 .3536 5 (3, 5) 5 (4 6) 0 0097 4 (3 5) 4 (3 5) 0 9361 4 (3 5) 4 (3 6) 0.9918 4 (2, 6) 2 (1, 4) 0 0061 4 (2 6) 2(1,4) 0.0166 4 (2, 5) 2 (1. 4) 0.0003 4 (2 6) 3 (1, 5) 0.0945 4 (3, 5) 5 (4, 6) 0 2269 4 (3, 5) 5(4, 6) 0 0129 4(2, 5) 4 (3 5) 0.9814 4 (2. 5) 4 (3, 5) 0 .6736 4 (3, 5) 5 (4, 5) 0 1598 4 (3, 5) 5 (4, 5) 0.1027 4 (3 5) 5 (4 5) 0 0196 4 (3, 5) 4 (3, 5) 0 9707 5 (4, 5) 5 (4, 5) 0 7529 4 (4 5) 5 (4, 5) 0.0832 5 (3, 5) 5 (4 5) 0 2767 5 (3, 5) 5 (3, 5) 0 .5497 2 (1, 3) 1 (1, 2) 0 1278 2 (1, 3) 1 (1, 2) 0 0012 2 (1. 3) 1 (1, 2) 0 0032 2 (1. 3) 1 5 (1, 3) 0.5112

PAGE 270

Appendix C: Comparison by Previous and Likely Donation: By Survey (Cont.) ( selected variables) Survey I Survey II Non Donors Donors P-value Unlikely Likely P value NonDonors P-value Un li kely Likely Donors Donors Donors Donors Donors Perceptions of Risk' Health risk of donat ion 3.7 (3.2, 4) 3.7 (3.3, 3.8) 0.9302 3 7 (3.3, 4)3.7 (3 .7, 3.7) 0 6216 3. 7 (3 3 4 )I 7 (3 .3, 3 7) 0 0745 .7 (3 .3, 4) : 7 (3.3 3.7) Overall risk seeking 3.7 (3.3, 4.3) 3 (2 .2, 3.5) 0.0001 3 7 (3, 4.3) 3 (2, 3 7) 0 0007 3 3 (2 .7, 4) 3 3 (2 .7, 4) 0 6459 3 7 (3 4) 3 2 (2, 4) Fear of donation 3.25 (2, 4) 4 (2 .8, 5) 0.0034 3.5 (2 25, 4) 4 3 (3; 5) 0 0161 3 (2, 4) 4 5 (3, 5) 0 0006 3 (2 5 4) 3 5 (2 5 5) Social Capital Government in HS 24.62 33.93 0 2599 24. 05 36 .59 0.1484 28.17 44.9 0 059 28.3 40.91 Government in college 4 .41 5 26 0 8245 2.5 9 .09 0 1017 6.85 15.69 0.114 1 82 17.65 Connected to 4 (3, 5) 4 (3, 5) 0.6606 4 (3, 5) 4 (3, 5) 0 53 4 (3, 5) 3 (2, 4) 0 032 4 (3 5) 3 (3, 4) Overall trust 4 (3 5) 3 (2, 4) 0.1419 4 (3, 4) 3 (2, 4) 0.0608 3 (3, 4) 4 (3, 5) 0.367 4 (3, 4) 3 (2, 4) Hours with organizations' 5 (3, 6)".5 (4 .3, 13.5) 0.0905 5 (3, 7.5) 10 (4 17) 0.0798 (4, 10) 8 (6, 12) 0.0492 5 (4, 10) 8 (4.5, 12) 1. Median (f' uartile, 3' ( q q uartile 2. Races with low Ns were bundled into the other" category. 3. Only of subjects who had spent time w i th organizations P-value 0 .101 0 .133 0 1695 0.1526 0.0045 0.095 0.7391 0.2112

PAGE 271

BIBLIOGRAPHY Adler, Ilan, & Kafry, Ditsa (1980). Capturing and clustering judges' policies. Organizational Behavior and Human Performance, 25, 384-394. Agence France-Presse (1998, Beijing (AFP), February 24). China-US-crime: China denies official policy to sell organs. ClariNet [On-line], 2/25/98. Available: http://start.earthlink.net/cgibin/psp ... eb.biz.industry .health.care&cobrand=eln Allison, James (1983). Behavioral economics. New York: Praeger Publishers, CBS Educational and Professional Publishing. Altman, L., (1970) Use of commercial blood increases with shortages in the U.S., New York Times. Altman, Lawrence K. (1989, September 5, late edition). Europe supplying blood for the U.S. The New York Times, p. Al. Altman Lawrence K. (1994, February 1, late edition). U.S. blood supply hits lowest level since World War II. The New York Times, p. Al. American Association of Blood Banks. AABB position manual (1980). Position Manual, 30.52. Arlington VA: American Association of Blood Banks. American Association of Blood Banks (1990). Technical manual: 101 h edition. Arlington VA: American Association of Blood Banks. Editor in chief Richard H. Walker, MD. 258

PAGE 272

American Association ofBlood Banks (1998, January). Donating blood: Where to donate blood; blood collection; donating for yourself; information on "national volunteer blood donor month" American Association of Blood Banks [On-line], 02/19/98. Available: www.aabb.org/docs/donate.html. American Association ofBlood Banks (1998, January). Facts about blood and blood banking. American Association ofBlood Banks [On-line], 02/19/98. Available: www .aabb.org/docs/facts.html. American Blood Commission, (1977), Recommendation for unified donor recruitment, July 26, 1977. American Red Cross (1997-98). Tissue services fact sheet (all figures based on 1995-96 data). American Red Cross--Tissue Services Fact Sheet [On-line], 9/15/98. Available: www.redcross.org/tissue/facttiss.html. American Survey. The solitary bowler. (1995, February 18). The Economist, 2122. Andaleeb, Syed Saad, & Basu, Amiya K. (1995, Spring). Explaining blood donation: The trust factor. Attitudes, personality, and demographics differentiate donors from nondonors. Journal of Health Care Marketing, .12(1), 42-48. Andrews, Lori, & Nelkin, Dorothy (2001). Body bazaar: The market for human tissue in the biotechnology age. New York: Crown Publishers. Aristotle ( -350 B.C./ 1999). Nichomachean ethics. Indianapolis: Hacket Publishing Company. Arkes, Hal R., & Hammond, Kenneth R. (Eds.). (1986). Judgment and decision making: An interdisciplinary reader. New York, NY: Cambridge University Press. Armas, Genarco C. (2000, Washington (AP), April21). Republican critics question census bureau calling card rewards. CNN.com allpolitics.com [On-line], 04/2112000. Available: www.cnn.com/2000/ALLPOLITICS/stories/04/21/callingcards.aplindex.htm I 259

PAGE 273

Arrow, Kenneth J. (1972, May). Gifts and exchanges." In Conference on Altruism and Economic Theory: Russell Sage Foundation (pp. 343-362). Arrow Kenneth J. (1982, January). Risk perception in psychology and economics. Economic Inguiry, 1-9. Associated Press staff ( 1999, Washington, January 11 ). Urgent need for blood donors: Weather woes cancel blood drives. ABCNEWS.com [On-line], 2/9/99. Available: http://more.abcnews.go com/sections/living/Daily News/blood990111.html Associated Press staff(l999, Sacramento, California (AP), November 22). California blood banks face finan,cial crisis. CNN.com [On-line], 11123/99 Available: www.cnn.com/HEALTH/9911/22/blood.banks.aplindex.html Associated Press staff (1999 Gaithersburg Maryland (AP), June 3). 'Mad cow' worry could bar travelers from donating blood. CNN interactive CNN .com [On-line], 6/3/99. Available : http:/ /cnn.com/HEALTH/9906/03/blood.madcow .ap/index.html Associated Press staff(1999, Washington (AP), June 27) Blood shortage could hit early next year, officials say. CNN interactive CNN.com [On-line], 6/29/99 Available: http:/ I cnn.com/US/9906/2 7 /blood.shortage.aplindex.html Associated Press staff(2000, New York (AP), January 18). Flu outbreak reduces nation's blood supply. CNN.com [On-line], 02/01/2000 Available: http://cnn com/2000/HEALTH/Ol/18/blood shortage.ap/index.html Associated Press staff(2000, Washington (CNN), January 31). Blood centers cite urgent need for donors. CNN.com [On-line], 02/01/2000. Available: wysiwyg://3/http:/ /cnn.com/2000/HEALTH/0 I /31 /blood.shortage/index.htm Auge Karen, (2002, Denver Post, January 22). Blood donors dwindle since 9/11. DenverPost.com [On"'line}, 1122/2002. Available: www.denverpost.com. Axelrod, Robert (1984). The evolution of cooperation. New York : Basic Books, Inc Axelrod, Robert ( 1997). The complexity of cooperation. Princeton, NJ: Princeton University Press. 260

PAGE 274

Bagozzi, Richard P. (1981, July). An examination ofthe validity oftwo models of attitude. Multivariate Behavioral Research, 16, 323-359. BarTal, Daniel ( 1985/86, Fall/Winter). Altruistic motivation to help: Definition, utility and operationalization. Humboldt Journal of Social Relations. U(1&2), 3-14. BarTal, Daniel (1976). Prosocial behavior: Theorv and research. Washington, DC: Hemisphere Publishing Corporation. Batson, C. Daniel (1987). Prosocial motivation: Is it ever truly altruistic? Advances in Experimental Social Psychology, 20, 65-122. Batson, C. Daniel (1990, March). How social an animal? The human capacity for caring. American Psychologist, 46(3), 336-346. Batson, C. Daniel (1991 ). The altruism question: Toward a social-psychological answer. Hillsdale, NJ: Lawrence Erlbaum Associates, Batson, C. Daniel and JayS. Coke. (1981). Empathy: A source of altruistic motivation for helping? 167-187. In, Rushton, J. Philippe and Richard M. Sorrentino, eds. ( 1981 ). A normative decision-making model of altruism. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Bazerman, Max H. (1986). Judgment in managerial decision making (3rd ed.). New York and Toronto: John Wiley & Sons, Inc Beal, R.W., & van Aken, W.G. (1992). Gift or good? A contemporary examination ofthe voluntary and commercial aspects ofblood donation. Vox Sang, 63, 1-5. Bellah, Robert N., Madsen, Richard, Sullivan, William M., Swidler, Ann, & Tipton, Steven M. (1985). Habits ofthe heart: Individualism and commitment in American life (updated ed. with new introduction). Berkeley, Los Angeles and London: University of California Press. Belle Bonfils Memorial Blood Center (1977). The Birth of a Blood Bank: A short history of the Belle Bonfiles Memorial Blood Center of Denver, Colorado. Denver, CO: Board ofTrustees Belle Bonfils Memorial Blood Center. 261

PAGE 275

Bendixen, Henrik, Manning, Frederick, & Sparacino, Linette (Eds.). (1996). Blood and blood products: Safety and risk, forum on blood safety and blood availability, Division of Health Sciences Policy, Institute of Medicine Washington DC: National Academy Press. Bendor, Jonathan (1995, December). A model of muddling through. American Political Science Review, 89(4), 819-840. Berry, Kathleen M. ( 1991, July 7, late edition). All about blood banks; A multibillion-dollar business in a nonprofit world. The New York Times, Section 3, p.8. Blood Supply Safety ( 1993, July 28). Hearing before the Subcommittee on Oversight and Investigations of the Committee on Energy and Commerce, House of Representatives, One hundred third Congress, first session. (Serial No. 1 03-42). Washington, DC: U.S. Government Printing Office. Bobrow, David B., & Dryzek, JohnS. (1987). Policy analysis by design. Pittsburgh, PA: University of Pittsburgh Press. Botten berg R.A. and Christal, R.E., ( 1961 ). An iterative technique for clustering criteria which retains optimum predictive efficiency. (WADDTN-61-30, ASTIA Document AD-261 615 Lackland Air Force Base, TX: Personnel Research Laboratory, Wright Air Development Division. As cited in: Cooksey, Ray W. (1996). Judgment analysis: Theory. methods, and applications. San Diego, CA: Academic Press, Inc., A Division of Harcourt Brace & Company. Bowersox, J. C. and J. R. Hess. (1995). Combat casualties, blood, and red blood cell substitutes: A military perspective in 1995. In: Winslow, Robert M., Kim. D. Vandegriffand Marcos Intaglietta Eds. (1995). Blood Substitutes: Physiological Basis of Efficacy. Boston: Birkhauser. Boyle, Robert ( 1666). Trials proposed by Mr. Boyle to Dr. Lower. Philosophical Transaction. 385-388 Brandt, Richard B. (1999). Utilitarianism and the rules of war, in: Moral Dimensions of the Military Profession, 51h edition. Forbes Custom Printing House. 262

PAGE 276

Breckler, S. J., & Wiggins, E. C. ( 1989). Scales for the measurement of attitudes towards blood donation. Transfusion. 29 (5), 401-404. Brehmer, Annica, & Brehmer, Berndt ( 1988). What have we learned about human judgment from thirty years of policy capturing? In: Brehmer, Berndt, & Joyce, C.R. B. (Eds .). (1988). Human Judgment: the SJT view. New York: North-Holland. Brehmer, Berndt, & Joyce, C.R. B. (Eds.). (1988). Human judgment: the SJT view. New York: North-Holland Bridges, Thomas (1994). The culture of citizenship: Inventing postmodern civic culture. Albany, NY: State University ofNew York Press. Briggs, N.C Piliavin, J.A., Lorentzen, D. & Becker, G.A (1986). On willingness to be a bone marrow donor. Transfusion, 26(4) 324-330. Brunswick, Egon (1952). The conceptual framework of psychology. Chicago: Chicago University Press. Buchanan, James M. (1968) The demand and supply of public goods. Chicago : Rand McNally & Company. Buchanan, James M. (1969). Cost and choice: An inquiry in economic theory. Chicago: Markham Publishing Company. Buchanan James M., & Tollison Robert D. (Eds.). (1972). Theory of public choice: Political applications of economics. Ann Arbor: The University of Michigan Press. Buchanan, James M., & Tullock Gordon (1965). The calculus of consent: Logical foundations of constitutional democracy. Ann Arbor: Ann Arbor Paperbacks, The University of Michigan Press. Burnet, J. J. (1982). Examining the profiles of the donor and nondonor through a multiple discriminant approach. Transfusion, 22(2), 138-142. Calabresi, Guido, & Bobbitt, Philip (1978) Tragic choices: The conflicts society confronts in the allocation of tragically scarce resources. New York and London: W.W. Norton & Company. 263

PAGE 277

Callero, Peter L & Piliavin, Jane Allyn. (1983). Developing a commitment to blood donation: The impact of one's fist experience. Journal of Applied Social Psychology, 13(1), 1-16 Carroll, JohnS. and Johnson, Eric J. (1990). Decision research: A field guide Applied Social Research Methods Series Vol. 22. Newbury Park: Sage Publications. Chang, HedyNai-Lin (1997). Democracy, diversity, and social capital. National Civic Review, 86(2) 141-147. Chergo, Jennifer (1999, June 10). Local blood supply, donors decline Denver Post. Chavis, David M. & Pretty, Grace M.H. (1999). Sense of community: Advances in measurement and application. Journal of Community Psychology. 27(6), 635-642. Chipuer, Heather M., & Pretty Grace M .H. (1999) A review of the sense of community index : Current uses, factor structure, reliability, and further development. Journal of Community Psychology, 27(6), 643-658. Churchill, Edward (1972), Surgeon to Soldiers, 47. Cimons, Marlene (200 1, June 18). Artificial blood is in the pipeline: Science: Universally compatible, the milky white substitute could hit market in two years. Los Angeles Times, Orange County Edition, pp. Al & A12. Civil societyand democratic citizenship (1998, Summer, Special Issue). Report from the Institute for Philosophy & Public Policy, 18(3), 1-39. Clark, Cheryl (1998a, January 18). Blood battle: Fierce competition drives changing marketplace for the precious 'gift oflife'. The San Diego Union Tribune, Special report, p. A38 Clark, Cheryl (1998b, January 18). Heartland supplies California with blood. The San Diego Union-Tribune Special report pp. A37-38 & A41. Clark, Cheryl ( 1998c January 18). FDA reviews claims of superiority by Red Cross. The San Diego Union-Tribune, Special report, pp. A40-41. 264

PAGE 278

Coleman, James S. ( 1990). Foundations of social theory. Cambridge, Massachusetts: The Belknap Press of Harvard University Press. Coleman, James S. (1994). A rational choice perspective on economic sociology. In Neil J. Smelser and Richard Swedberg (Eds.), The handbook of economic sociology (pp. 166-180). New Jersey: Russell Sage Foundation published by Princeton University Press. Condie, Spencer J., Warner, W. Keith, & Gillman, David C. (1976). Getting blood from collective turnips: Volunteer donation in mass blood drives. Journal of Applied Psychology, 61(3), 290-294 Connolly, Terry, Arkes, Hal R., & Hammond, Kenneth R. (2000). Judgment and decision making: An interdisciplinary reader (2nd ed.). Cambridge, UK: Cambridge University Press. Cooksey, Ray W. (1996). Judgment analysis: Theory. methods, and applications. San Diego, CA: Academic Press, Inc., A Division of Harcourt Brace & Company Cooksey, Ray W., Freebody, Peter, & Davidson, Graham R. (1986, Spring). Teachers' predictions of children's early reading achievement: An application of social judgment theory. American Educational Research Journal, 23(1), 41-64. Cooper, Michael H., & Culyer, Anthony J. (1968). The price ofblood: An economic study of the charitable and commercial principle. Hobart Paper 41, Institute of Economic Affairs. Crandall, James E., & Harris, Michael D. (1991, March). Social interest, cooperation and altruism. Individual Psychology, 4 7 ( 1 ), 115-119. Creswell, John W. (1994). Research design: qualitative & quantitative approaches. Thousand Oaks, CA: Sage Publications, Inc. Crick, Bernard (2001). Citizens: Towards a citizenship culture. Oxford, UK: Blackwell Publishers. Cuyer, A.J. (1968). Ethics and economics in blood supply. Lancet, 602-603. 265

PAGE 279

D'Augelli, Anthony R., D'Augelli, Judith Frankel, & Danish, Steven J. (1981). Helping Others. Monterey, California: Brooks/Cole Publishing Company. Dahl, Robert A. (1982). Dilemmas of pluralist democracy: Autonomy vs. control. New Haven: Yale University Press. Dahl, Robert A. (1993, March/April). Finding competent citizens. Improving democracy. Current, 351, 23-30. Dahl, Robert A. (1997, Summer). On deliberative democracy. Dissent, 44(3), 5458. Dahl, Robert A. (1989). Democracy and its critics. New Haven: Yale University Press. Danziger, Marie (1995). Policy analysis postmodernized: Some political and pedagogical ramifications. Policy Studies Journal, 23(3), 435-450. Darlington, Richard B. (2002). Factor analysis. Available: www. psych. cornell. edu/ darlington/factor.html Dasgupta, Partha (1993, September). Altruism and the allocation of resources. Social Service Review, 374-387. Dasgupta, Partha, & Serageldin, Ismail (1999). Social capital: A multifaceted perspective. Washington DC: The World Bank. Dawes, R.M. and Corrigon, B. (1974). Linear models in decision making. Psychological bulletin. 81, 95-106. Day, Michael (1999, January 23). Testing times ahead: If you were incubating "human BSE", would you want to know? New Scientist, 5. deLeon, Peter (1992a, March/April). Democracy, analysis, and entrepreneurship. The democratization of the policy sciences. Public Administration Review, 52(2), 125-129. deLeon, Peter (1992b, Autumn/Winter). Policy formulation: Where ignorant armies clash by night. Policy Studies Review, 11(3/4), 389-405. 266

PAGE 280

deLeon, Peter (1994a). Democracy and the policy sciences: Aspirations and operations. Policy Studies Journal, 22(2), 200-212 de Leon, Peter ( 1994b ). Reinventing the policy sciences: Three steps back to the future. Policy Studies Journal, 27, 77-95. deLeon, Peter (1995). Democractic values and the policy sciences. American Journal of Political Science, 39( 4), 886-905. deLeon, Peter (1997). Democracy and the policy sciences. Albany, NY: State University ofNew York Press. deLeon, Peter (1998a). Advice and consent: The development ofthe policy sciences. New York: Russell Sage Foundation. deLeon, Peter (1998b). Models of policy discourse: Insights versus prediction. Policy Studies Journal, 26(1), 147-161. de Leon, Peter ( 1998c ). Introduction: The evidentiary base for policy analysis: Empiricist versus postpositivist positions. Policy Studies Journal, 26( 1 ), 109-113. del Pozo, Pablo Rodriguez (1994). Paying donors and the ethics ofblood supply. Journal of Medical Ethics, 20, 31-35. De Tocqueville, Alexis (1,84011984). Democracy in America. New York: Penguin Books. Dickson, David. (l988). The new politics of science. Chicago: The University of Chicago Press. Dionne, E.J., Jr. (1997, Fall). Why civil society? Why now? The Brookings Review, 5-8. Dovidio, John F. (1984). Helping behavior and altruism: An empirical and conceptual overview. Advances in Experimental Social Psychology, 17, 361-427. Downs, Anthony ( 1957). An economic theory of democracy. New York: Harper & Row Publishers. 267

PAGE 281

Drake, Alvin W., Finkelstein, StanN., & Sapolsky, Harvey M. (1982). The American blood supply. Cambridge, Massachusetts: The MIT Press. Dryzek, John S. (1990). Discursive democracy: Politics, policy, and political science. New York, NY: Cambridge University Press. Dryzek, JohnS. (1990, September). Political inclusion and the dynamics of democratization. American Political Science Review, 475-487. Dulles, Foster Rhea (1950). The American Red Cross: A history. New York: Harper & Brothers Publishers. Dunteman, George H. (1989). Principal components analysis. Newbury Park: Sage Publications. Eastis, Carla M. (1998, September). Organizational diversity and the production of social capital: One of these groups is not like the other. American Behavioral Scientist, 42( 1 ), 66-77. Eckert, Ross D. (1988) AIDS and the blood bankers, in: Swartz, Thomas R. and Frank J. Bonello. Taking sides: Clashing views on controversial economic issues, 41h edition. The Dushkin Publishing Group, Inc. Eckert, Ross D., & Wallace, Edward L. (1985). Securing a safer blood supply: Two views. Washington: American Enterprise Institute for Public Policy Research. Edwards, Bob, & Foley, Michael W. (Eds.). (1997, March/April). Social capital, civil society, and contemporary democracy. American Behavioral Scientist, 40(5). Edwards, Bob, & Foley, Michael W. (1998, September). Civil society and social capital beyond Putnam. American Behavioral Scientist, 42( 1 ), 124-139. Eisenberg, Nancy (Ed.). (1982). The development ofprosocial behavior (Developmental psychology series). San Diego: Academic Press. Elshtain, Jean Bethke (1997, Fall). Civil society creates citizens; it does not solve problems. Not a cure-all. The Brookings Review, 9-12. Elster, Jon (Ed.). (1986). Rational choice. Oxford: Basil Blackwell. 268

PAGE 282

Elster, Jon (1992). Local justice: How institutions allocate scarce goods and necessary burdens. New York: Russell Sage Foundation. Elster, Jon (1993). Justice and the allocation of scarce resources. Logical Perspectives on Justice. Mellerc & Barron. Epstein, Richard A. (1993, September). Altruism: Universal and selective. Social Service Review, 388-405. Etzioni, Amitai (1996). The new golden rule: Community and morality in a democratic society. New York, NY: BasicBooks. Fischer, Frank ( 1998). Beyond empiricism: Policy inquiry in postpositivist perspective. Policy Studies Journal, 26(1), 129-146. Fischer, Frank, & Forester, John (Eds.). (1993). The argumentative turn in policy analysis and planning. Durham: Duke University Press. Fischhoff, Baruch (1988). Judgment and decision making. 153-187. In Sternberg, Robert J., & Smith, Edward E. (Eds.), The psychology ofhuman thought (153-187). Cambridge: Cambridge University Press. Fischhoff, Baruch, Paul Slavic and Sarah Lichtenstein. "The public" vs. "the experts": perceived vs. actual disagreements about risks of nuclear power. Source lost. Flap, Henk (1999). Creation and returns of social capital. A new research program. La Revue Tocgueville/The Tocgueville Review, (1), 5-44. Foddy, Margaret, Smithson, Michael, Schneider, Sherry, & Hogg, Michael, (Eds.). ( 1999). Resolving social dilemmas: Dynamic, structural, and intergroup aspects. Ann Arbor, MI: Edwards Brothers. Foley, Michael W., & Edwards, Bob (1997, March/April) Editors' introduction: Escape from politics? Social theory and the social capital debate. In, Edwards, Bob, & Foley, Michael W. (Eds .). Social capital, civil society, and contemporary democracy. American Behavioral Scientist, 40(5), 550-561. 269

PAGE 283

Foley, Michael W., & Edwards, Bob (1998, September). Beyond Tocqueville: Civil society and social capital in comparative perspective. American Behavioral Scientist, 42( 1 ), 5-20. Foley, Michael W., & Edwards, Bob (1999). Is it time to disinvest in social capital? Journal Public Policy, 19(2), 141-173. Forbes, J.M., & Laurie, M.L. (1994). Blood collections, 1988 through 1992. Transfusion, 34(5), 392-395. Frantz, Douglas (1996, May 30). Blood bank politics-A special report.; Elizabeth Dole and Red Cross: 2 Powers at Work. The New York Times [On-line], Name: Elizabeth Dole, Section A, pi, c2, National Desk, 3/2/99. Available: http://web.lexisnexis.com/univers ... 5=6a9a606fc540d403757c975b88340c09 Fratantoni, Joseph C., MD. (1998). Chapter 2. Demonstration ofthe efficacy of a therapeutic agent. In, Rudolph, Alan S., Rabinovici, Reuven, & Feuerstein, Giora Z. (Eds.) Red blood cell substitutes: Basic principles and clinical applications. New York: Marcel Dekker, Inc. Fukuyama, Francis (1995). Trust: the social virtues and the creation of prosperity. New York: The Free Press. Fukuyarna, Francis (1999a). The great disruption: Human nature and the reconstitution of social order. New York: Simon & Schuster. Fukuyama, Francis (1999b, May). The great disruption: Human nature and the reconstitution of social order. The Atlantic Monthly, 55-78. Galston, William A., & Levine, Peter (1997, Fall). America's civic condition: A glance at the evidence. The'Brookings Review, 23-26. Ganley, Elaine, Associated Press staff(1999, Paris, February 9). Tainted blood trial begins. ABCNEWS.com [On-line], 2/9/99. Available: http://www.abcnews.go.com/sections/world/Daily N ews/taintedblood990209 .html Gassier, Robert Scott. ( 1986). The economics of nonprofit enterprise: a study in applied economic theory. New York: University Press of America. 270

PAGE 284

Gigerenzer, Gerd, Todd, Peter M., & the ABC Research Group (1999). Simple heuristics that make us smart. New York: Oxford University Press. Gaul, Gilbert M. (1989, September 24). How blood, the 'gift oflife,' became a billion-dollar business. Philadelphia Inquirer [On-line], Section: National, AO 1, 3/3/99. Available: http://newslibrary.krmediastream.com/cgi-... ocument/nl2 auth?DBLIST=pi89&DOCNUM=57856 Gaul, Gilbert M. (1989, September 25). The loose way the FDA regulates blood industry. Philadelphia Inquirer [On-line], Section: National, A01, 3/3/99. Available: http:/ /newslibrary.krmediastream.com/cgi-... ocument/nl2 auth?DBLIST=pi89&DOCNUM=58218 Gaul, Gilbert M. (1989, September 26). Fear of aids spurs change. Philadelphia Inquirer [On-line] Section: National, A01, 3/3/99. Available: http:/ /newslibrary .krmediastream.com/ cgi-... ocument/nl2 auth?DBLIST=pi89&DOCNUM=58402 Gaul, Gilbert M. (1989, September 27). Tax exemption on blood business challenged. Philadelphia Inquirer [On-line], Section: National, A12, 3/3/99. Available: http://newslibrary.krmediastream.com/cgi-... ocument/nl2 auth?DBLIST=pi89&DOCNUM=58705 Gaul, Gilbert M. (1989, September 27). Red Cross: From disaster reliefto blood. Philadelphia Inquirer [On-line], Section: National, AOl, 3/3/99. Available: http:/ /newslibrary .krmediastream.com/ cgi-... ocument/nl2 _auth?DBLIST=pi89&DOCNUM=58684 Gaul, Gilbert M. (1989, September 28). America: The OPEC of the global plasma industry. Philadelphia Inquirer [On-line], Section: National, AO 1, 3/3/99. Available: http:/ /newslibrary.krmediastream.com/cgi-... ocument/nl2 auth?DBLIST=pi89&DOCNUM=58775 General Accounting Office, Health, Education, and Human Services Division. (1999, 187R). Blood supply: Availability ofblood to meet the nation's requirements (GAO Office No. B-283557, September 20). Heinrich, Janet: Author. Glaeser, Edward L., Laibson, David, & Soutter, Christine (1999, May). What is social capital? Unpublished transcript, MacArthur Foundation, part of the MacArthur Foundation Network on Norms and Preferences. 271

PAGE 285

Glazer, Nathan (1971). Blood, The Public Interest. Goldstein, Avram (1999, June 6). U.S. blood donations decline: Supply shortfall looms. Washington Post [On-line], Page A02, 6/7/99. Available: http:/ /search.washingtonpost.com/wp-srv/WPlate/1999-06/06/2321-060699idx.html Goldstein, Avram (1999, June 7). Plea for blood donations is no case of crying wolf. Chicago Sun-Times [On-line], 5XS; Issue: PSA-2364; NEWS Section, 7/2/99. Available: http://library.northernlight.com/PN19990616080008386.html?cb=O&sc=O Goldstein, William M., & Hogarth, Robin M. (Eds.). (1997). Research on judgment and decision making: Currents, connections, and controversies. New York, NY: Cambridge University Press. Gonzales, Angela, The Business Journal staff(1998, March 2). Blood services affected by managed care trend. nptimes.com [On-line], The Business Journal, Phoenix, 2/22/99. Available: http://www .am city .com/journals/health_ care/phoenix/ 1998-0302/story6.html Gorsuch, Richard L. (1983). Factor analysis (2nd ed.). Hillsdale, New Jersey: Lawrence Erlbaum Associates, Inc., Publishers. Govier, Trudy (1998). Dilemmas oftrust. Quebec, Canada: MeGill-Queen's University Press. Greeley, Andrew (1997, March/April). Coleman revisited: Religious structures as a source of social capital. In, Edwards, Bob, & Foley, Michael W. (Eds.). Social capital, civil society, and contemporary democracy. American Behavioral Scientist, 40(5), 587-594. Green, Donald P ., & Shapiro, Ian ( ( 1994 ). Pathologies of rational choice theory: A critique of applications in political science. New Haven and London: Yale University Press. Green, Leonard, & Kagel, John H. (1996). Advances in behavioral economics. Volume 3: Substance use and abuse. Norwood, NJ: Ablex Publishing Corporation. 272

PAGE 286

Hair, J.F., Anderson, R.E., Tatham, R.K. and Black, W.C. (1992). Multivariate data analysis with readings. New York: Macmillan Publishing Company. Hammond, Kenneth R. ( 1996). Human judgment and social policy: Irreducible uncertainty, inevitable error, unavoidable injustice. New York and Oxford: Oxford University Press. Hammond, Kenneth, R., McClelland, Gary H., & Mumpower, Jeryl (1980). Human judgment and decision making: theories. methods and procedures. New York: Preager Special Studies, Hemisphere Publishing Corporation. Hardin, Garrett (1968). The tragedy of the commons. The population problem has no technical solution; it requires a fundamental extension in morality. Science, 162, 1243-1248. Hardin, Russell (1993, September). Altruism and mutual advantage. Social Service Review, 358-373. Hardin, Russell (1993, December). The street-level epistemology of trust. Politics & Society, 21(4), 505-529. Hardin, Russell (2000). The public trust. In: Pharr, Susan, J., & Putnam, Robert D. (Eds.). (2000). Disaffected democracies: What's troubling the trilateral countries? Princeton: Princeton University Press. Hatcher, Larry, PhD. (1994a). A step-by-step approach to using the SAS system for factor analysis and structural equation modeling. Cary, NC: SAS Institute Inc. Hatcher, Larry, PhD., & Stepanski, Edward J. ( 1994b ). A step-by-step approach to using the SAS system for univariate and statistics. Cary, NC: SAS Institute Inc. Healy, Kieran (2000, May). Embedded altruism: Blood collection regimes and the European Union's donor population. American Journal of Sociology, 1 05( 6), 1633-57. Heffner, Richard D. (Ed.). (1956, renewed by Heffner 1984). Democracy in America: Alexis de Tocqueville (A Mentor Book abridged edition). New York, NY: Penguin Group, Penguin Books, 273

PAGE 287

Hill, Jean L. (1996). Psychological sense of community: Suggestions for future research. Journal of Community Psychology, 24(4),431-438. Hoffman, P .J. ( 1960). The paramorphic representation of clinical judgments. Psychological bulletin, 57 16-131. Hogarth, Robin M., & Reder, Melvin W. (Eds.). (1989). Rational choice: The contrast between economics and psychology. Chicago: The University of Chicago Press. Hogarth, Robin M. (Ed.). (1990). Insights in decision making. Chicago: The University of Chicago Press. Holmes, Thomas P. (1990, May). Self-interest, altruism, and health-risk reduction: An economic analysis of voting behavior. Land Economics, 66(2), 140-149. Horton, Raymond L., & Horton, Patricia J. (1993, Spring). Improving the current system for supplying organs for transplantation. Journal of Health Politics, Policy and Law, 18(1), 175-187. Hosmer, Larue Tone ( 1995). Trust: The connecting link between organizational theory and philosophical ethics. Academy of Management Review. 20(2), 379-403. Hosmer, D.W., & Lemeshow, S. (1989). Applied logistic regression. New York: John Wiley and So.ns. Hough, Douglas E. (1978). The market for human blood. Lexington, Massachusetts: Lexington Books, DC Heath and Company. Hurd, Charles (1959). The compact history ofthe American Red Cross. New York: Hawthorn Books, Inc. James, Estelle (1998). Commercialism among nonprofits: Objectives, opportunities, and constraints. In Weisbrod, Burton (Ed.), To profit or not to profit: the commercial transformation of the nonprofit sector. Cambridge University Press. Jennings, C.E.J (1888) On transfusions ofblood and saline fluids, pp. 106-115: London: Bailliere, Tindall and Cox 274

PAGE 288

Johnson, David B. (Ed.). (1976). Blood policy: issues and alternatives. American Enterprise InstitUte for Public Policy Research, Washington DC Joseph, Jenifer, ABCNEWS staff(1999, March 3). Does a brain disease threaten our blood supply? ABCNEWS.com [On-line], 2/9/99. Available: http://more.abcnews.go.com/sections/living/Daily N ews/bloodrecall0302.html Kahn, Jeffrey P., PhD., M.P.H. (1998a). Is there a difference between selling eggs and selling kidneys? CNN.com [On-line], 10/05/1998. Available: http://cnn.com/HEALTH/indepth.healthlbioethics/9805/selling.eggs/index.ht ml Kahn, Jeffrey P., PhD., M.P.H. (1998b, July 9). Would you give a stranger your kidney? The ethics of''unknown" kidney donors. CNN.com [On-line], 1 0/05/1998. Available: www .cnn.com/HEALTH/bioethics/9807 /stranger.kidney/index.html Kahneman, Daniel & Tversky, Amos (1973). On the psychology of prediction. Psychological Review, 80, 237-251. Kahneman, Daniel and Amos Tversky (1974). Judgment under uncertainty: heuristics and biases. Science, 185,1124-1131. Kahneman, Daniel, Slovic, Paul, & Tversky, Amos (Eds.). (1982). Judgment under uncertainty: Heuristics and biases. New York: Cambridge University Press. Kant, Immanuel (178511993). Grounding for the metaphysics of morals. Indianapolis: Hacket Publishing Company. Keown, John (1997). The gift ofblood in Europe: an ethical defense ofEC directive 89/381 [European Community directive]. Journal of Medical Ethics, 23, 96-100. Keown, John (1998). A reply to McLachlan. Journal ofMedical Ethics, 24,255256. 275

PAGE 289

Kessler, David (1993). In: Blood supply safety (1993, July 28). Hearing before the Subcommittee on Oversight and Investigations of the Committee on Energy and Commerce, House of Representatives, One hundred third Congress, first session. (Serial No. 103-42). Washington, DC: U.S. Government Printing Office. Keynes, Geoffrey (1941). Blood transfusion. Bristol: John Wright and Sons. Kim, Jae-On, & Mueller, Charles W. (1978). Introduction to factor analysis: What it is and how to do it (series, number 07-013: Quantitative applications in the social sciences). Beverly Hill, California: Sage Publications, Inc. Kim, Jae-On, & Mueller, Charles W. (1978). Factor analysis: Statistical methods and practical issues (series, number 07-014: Quantitative applications in the social sciences). Newbury Park, CA: Sage Publications. King, Gary, Keohane, Robert 0., & Verba, Sidney (1994) Designing social inquiry: Scientific inference in qualitative research. New Jersey: Princeton University Press. Kitcher, Philip (1993, October). The evolution of human altruism. The Journal of Philosophy, XC (I 0), 497-516. Kohr, Lorraine, MS, & Sayers, Merlyn, MD, PhD. (Eds.). (1998). Motivating donors in today's world: Recruitment and retention. Bethesda, Maryland: AABB Press. Kottow, Michael H., MD. ( 1990). Against the magnanimous in medical ethics. Journal ofMedical Ethics, 16, 124-128. Lappe, Frances Moore, & DuBois, Paul Martin (1997, Summer). Building social capital without looking backward. National Civic Review, 86(2), 119-127. Landy, Marc (1993). Public policy and citizenship. In: Ingram, Helen M. and Smith, Steven Rathgeb. Public policy for democracy. Washington D.C.: Brookings Institute. Lee, Lichang, Piliavin, Jane Allyn, & Call, Vaughn R.A. (1999). Giving time, money, and blood: Similarities and differences. Social Psychology Quarterly, 62(3), 276-290. 276

PAGE 290

Lee, Stephanie J., Liljas, Bengt, Neumann, Peter J., Weinstein, Milton C., & Johannesson, Magnus, (1998). The impact of risk information on patients' willingness to pay for autologous blood donation. Medical Care, 36(8), 1162-1173. Lee, S.J., Liljas, B., Churchill, W.H., Popovsky, M.A., Stowell, C.P., Cannon, M.E., & Johannesson, M. ( 1998, August). Perceptions and preferences of autologous blood donors. Transfusion, 38, 757-763. Leeds, R. ( 1963). Altruism and the norm of giving. Merrill-Palmer Quarterly, 9, 229-240. Leikola, Juhani ( 1998). Achieving self sufficiency in blood across Europe: European countries should encourage unpaid donors and dismantle barriers (editorial). British Medical Journal [On-line] Citation information: (v316n7130) Start Page: p489(2) ISSN: 0959-8146,7/2/99. Available: //library.northernlight.com/LW1998 .. .iolPAnwVAmkiXQhSUUMGEhdHd gA%253D Lerner, Melvin J., & Meindl, James R. (1981). Justice and altruism. 213-231. In, Rushton, J. Philippe, & Sorrentino, Richard M. (Eds.). (1981). A normative decision-making model of altruism. Hillsoale, NJ: Lawrence Erlbaum Associates. Levi, Margaret (1996, March). Social and unsocial capital: A review essay of Robert Putnam's Making democracy work. Political & Society, 24(1), 4555. Light, Alan (2002, February). National Public Radio. Lightman, Ernie S. (1981 ). Continuity in social policy behaviors: The case of voluntary blood donorship. Journal Social Policy, 10(1), 53-79. Lingeman, Richard R. (1970). Don't you know there is a war on? The American home front. New York: G. P. Putnam's Sons. London, Perry and Bernice M. Hemphill ( 1965). The motivations of blood donors Transfusion, 559-568. MacRae, Duncan, Jr., & Whittington, Dale (1997). Expert advice for policy choice: Analysis & discourse. Washington, DC: Georgetown University Press. 277

PAGE 291

McClelland, Gary, H., Schulze, William D., & Hurd, Brain ( 1990). The effect of risk beliefs on property values: A case of a hazardous waste site. Risk Analysis, 10(4), 485-497. McCloskey, Donald N. (1994). Knowledge and persuasion in economics. New York, NY, USA and Cambridge, UK: Cambridge University Press. McCullough, Jeffrey, M.D. (1993, May 5). The nation's changing blood supply system. JAMA, 269(17), 2239-2245. McLachlan, Hugh V. (1998). The unpaid donation of blood and altruism: A comment on Keown. Glasgow Caledonian University: Journal of Medical Ethics, 24, 252-254. McMillan, David W., & Chavis, David M. (1986, January). Sense of community: A definition and theory. Journal of Community Psychology, 14. Malloy, D., B. McDonough and M. Fuller (Eds.) (1991). A look to the future of blood banking and transfusion medicine: Report from the American association ofblood banks foundation managerial think tank Transfusion, .ll(S), 450-465. Manning, Frederick J., & Sparacino, Linette (Eds.). (1996). Blood donors and the supply of blood and blood products. forum on blood safety and blood availability, Division of Health Sciences Policy, Institute ofMedicine. Washington DC: National Academy Press. Mansbridge, Jane J. (Ed.). (1990). Beyond self-interest. Chicago: The University of Chicago Press. Mayer, F. Stephan, Shelley Duval, RolfHoltz, and Craig Bowman, (1985, June). Self-focus, helping request salience, felt responsibility, and helping behavior [Published by the Society for Personality and Social Psychology, Inc.]. Personality and Social Psychology Bulletin, 11 (2), 133-144. Mayer, Klaus (1987, April). The community: Still the best source ofblood. Hastings Center Report, 5-7. Mazo, Ellen (1998, September 1). The blood business; A community endeavor becomes a big industry. Pittsburgh Post-Gazette, Section Health, p. G 1. 278

PAGE 292

Mill, John Stuart (1863/1993). On liberty and utilitarianism. New York: Bantam books. Ministry of Health Press Release (1998, February 26). Precautionary measures on blood products-theoretical NVCJD link. Britain in the USA. [On-line], 9/15/98. Available: http://www.britain info.org/BIS/FORDOM/HEALTH/27F eb98.stm Murray, Thomas H. (1987, April). Gifts ofthe body and the needs of strangers. Hastings Center Report, 30-38. Nagel, Thomas (1970). The possibility of altruism. Ely House, London, England: Oxford University Press and Clarendon Press. National Blood Data Resource Center, (1998) unpublished data. National Heart, Lung, and Blood Institute expert panel, chair: Toy, Pearl T.C.Y., MD ( 1998). Transfusion alert: Use of autologous blood (Introduction; Preoperative autologous blood donation; Perioperative blood salvage; Acute normovolemic hemodilution; Transfusion of autologous blood components; Summary) [On-line], 9115/98. Available: http://www .nhlbi.nih.gov/nhlbi/bloodltransf/prof/logo.htm Naylor, James C., & Wherry, Robert J., Sr. (1965). The use of simulated stimuli and the "Jan" technique to capture and cluster the policies of raters [research was supported in part by the Air Force Systems Command, Contract AF41(609)-1596]. Educational and Psychological Measurement, XXV(4), 969-986. Neiman, Max and Stephen J. Stambaugh. Rational choice theory and the evaluation of public policy [Symposium on Politics and Policy]. Policy Studies Journal, 26, 3. Neuberg, Ldand Gerson (1999, Winter). A book review. A solution to the ecological inference problem: Reconstructing individual behavior from aggregate data. Journal of Policy Analysis and Management, 18( 1 ), 191198. 279

PAGE 293

Neustadt, Richard E., & May, Ernest R. (1986). Thinking in time: The uses of history for decision makers. New York, NY: The Free Press, A Division of Macmillan, Inc. Newton, Kenneth (1997). Social capital and democracy. American Behavioral Scientist, 40(5), 575-586 Nickerson, Carol A.E., McClelland, Gary H., & Petersen, Doreen M. (1991, September 11 ). Measuring contraceptive values: An alternative approach. Journal ofBehavioral Medicine, 14(3), 241-265. Norman, Geoffrey R., PhD., & Streiner, David L., PhD. (1997). PDQ statistics (2nd ed.). St. Louis, Missouri: Mosby-Year Book, Inc., A Times Mirror Company. Norman, Geoffrey R. PhD., & Streiner, David L., PhD. (2000). Biostatistics: The bare essentials (2nd ed.). Hamilton, Ontario: B.C. Decker Inc. Company. Nucci, Mary L., & Abuchowski, Abraham (1998, February). The search for blood substitutes. Scientific American, 73-77. Oakley, Ann, & Ashton, John (ed.) (1997). The gift relationship: From human blood to social policy. London: LSE Books. Office ofTechnology Assessment, U.S. Congress (1985). Blood policy and technology. January 1985, OTA-H-260. Olson, Mancur, & Kahkonen, Satu (Eds.). (2000). A not-so-dismal science: A broader view of economies and societies. New York: Oxford University Press. Onyx, Jenny, & Bullen, Paul (2000, March). Measuring social capital in five communities. The Journal of Applied Behavioral Science, 36(1), 23-42. Orfinger, Becky (2000, July 12). Red Cross calls national blood shortage 'a disaster'. Http://www .redcross.org/news/inthnews/00/7 -12a-OO.html Osborne, D J., and S. Bradley (1975). Blood donor and non-donor motivation: a transnational replication. Journal of Applied Psychology, 60 409. 280

PAGE 294

Ostander, Susan, A. and Stuart Langton (ed) (1987). Shifting the debate: Public/private sector relations in the modem welfare state. Transaction Books, New Brunswick. Ostrom, Elinor (1990). Governing the commons: The evolution of institutions for collective action. New York: Cambridge University Press. Oswalt, Robert M. and Marie Napoliello (1974). Motivations ofblood donors and nondonors. Journal of Applied Psychology, 59(1), 122-124. Oswalt, R.M. and T.E. Hoff (1975, January/February). The motivations of blood donors and nondonors: A community survey. Transfusion, 15(1), 68-72. Oswalt, R.M. (1977, March/April). A review ofblood donor motivation and recruitment. Transfusion,.17(2), 123-135. Parfit, Derek (1986). Prudence, morality, and the prisoner's dilemma. In, Elster, Jon (Ed.), Rational choice. Oxford: Basil Blackwell. Pearle, Jone L. (1993). Volunteers: The organizational behavior of unpaid workers. New York: Routledge. Petracca, Mark P. (1991). The rational choice approach to politics: A challenge to democratic theory. Review of Politics 53(2), 288-319. Pharr, Susan, J. and Robert D. Putnam, Eds. (2000). Disaffected democracies: What's troubling the trilateral countries? Princeton: Princeton University Press. Piliavin, J.A. (1987). Temporary deferral and donor return. Transfusion, 27(2), 199-200. Piliavin, Jane Allyn and I. M. Piliavin (1973). The good Samaritan: Why does he help? Unpublished manuscript, University of Wisconsin. Piliavin, Jane Allyn, Peter L. Callero, and Doorcas, E. Evans (1982). Addiction to altruism? Opponent-process theory and habitual blood donation. Journal of Personality and Social Psychology 43(6), 1200-1213. 281

PAGE 295

Piliavin, Jane Allyn, D. E. Evans and Peter L. Callero (1984 ). Learning to "give to unnamed strangers": The process of commitment to regular blood donation. In E. Staub D. Bar-Tal J. Karylowski, and J Reykowski, (Eds ) The Development and Maintenance of Pro social Behavior: International Perspective. New York: Plenum. Piliavin Jane Ally, and Donald Libby (1985) Personal norms, perceived social norms, and blood donation. Humboldt Journal of Social Relations 12, 1&2: 159-194. Piliavin, Jane Allyn (1987). Temporary deferral and donor return. Transfusion, 27, 199-200. Piliavin, J. A. ( 1990). Why do they give the gift of life? A review of research on blood donors since 1977. Transfusion, 30 (5), 444-459. Piliavin, Jane Allyn and Hong-Wen Charng (1990). Altruism: A review of recent theory and research. Annual Review of Sociology, 16,27-65. Piliavin, Jane Allyn and Peter L. Callero (1991). Giving blood: The development of an altruistic identity. Baltimore: The John Hopkins University Press Piliavin, Jane Allyn, Dorcas E. Evans and Peter Callero. (1984). Learning to "give to unnamed strangers" : The process to commitment to regular blood donation. In, Staub, Ervin, Bar-Tal, Daniel, Karylowski, Jerzy, and Reykowski, Janusz (Eds.). (1984). The development and maintenance of prosocial behavior: International perspectives on positive morality. New York: Plenum Piliavin, Jane Allyn and Donald Libby (1985 / 86). Personal norms, perceived social norms, and blood donation. Humboldt Journal of Social Relations, 13(1&2).. 159-194. Plant, Raymond (1977). Gifts exchanges and the political economy ofhealth care. Journal ofMedical Ethics, 3, 166-173. Poole Ross (1999). Nation and Identity. New York : Routledge, Taylor & Francis Group. 282

PAGE 296

Potapchuk, William R., Crocker, Jarle P., & Schechter, William H., Jr. (1997, Summer). Building community with social capital: Chits and chums or chats with change. National Civic Review, 86(2), 129-139. Pressman, Jeffrey L., & Wildavsky, Aaron (1984). Implementation: How great expectations in Washington are dashed Oakland; or, Why it's amazing that federal programs work at all this being a saga of the economic development administration as told by two sympathetic observers who seek to build morals on a foundation of ruined hopes (3rd ed., expanded). Berkeley: University of California Press. Prottas, Jeffrey M. (1992). Competition for altruism: Bone and organ procurement in the United States. The Milbank Quarterly, 70(2), 299-317. Prottas, Jeffrey M. (1993). Altruism, motivation, and allocation: Giving and using human organs. Journal of Social Issues, 49(2), 137-150. Putnam, Robert, D (1993a, Spring). The prosperous community: Social capital and public life. The American Prospect, 35-42. Putnam, Robert D., with Leonardi, Robert, & Nanetti, Raffaella Y. (1993b). Making democracy work: Civic traditions in modem Italy. Princeton, NJ: Princeton University Press. Putnam, Robert D. ( l995a, December). Tuning in, tuning out: The strange disappearance of social capital in the United States. PS: Political Science and Politics 28 (4), 664-683. Putnam, Robert D (1995b). Bowling alone: America's declining social capital. Journal ofDemocracy, 6(1), 65-78. Putnam, Robert D. ( 1995c, Spring). Foundations of democracy: Bowling alone, revisited The Responsive Community, 19-33. Putnam, Robert D. ( 1996, March/ April). Robert Putnam responds. The American Prospect, 25, 26-28 Putnam, Robert D. (2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon & Schuster. 283

PAGE 297

Quigley, Barbara, Gaes, Gerald G., & Tedeschi, James T. (1988) Does asking make a difference? Effects of initiator, possible gain, and risk on attributed altruism. The Journal of Social Psychology, 129(2), 259-267. Radin, Beryl A. ( 1995). Presidential address: The evolution of the policy analysis field: From conversation to conversations. Journal of Policy Analysis and Management, 16(2), 204-218. Reaney, Patricia (1998, London, February 27). Britain to import blood plasma due to 'mad cow' risk. http://www.nando.net and Reuters News Service [On line], 9/15/98. Available: www .nando net/newsroom/ntn/health/022798/health22 1994 7 .html. Reitman, Judith (1996). Bad blood: Crisis in the American Red Cross. New York, NY: Kensington Publishing Corp. Reuters staff(2000, Geneva (Reuters), April 7). UN says developing world at risk from unsafe blood. CNN com health [On-line], 04/07/2000. Available: http:/ I cnn.com/2000/HEAL TH/04/07 /health. blood.reut/index.html. Reynolds, Charles H. & Norman, Ralph V. (Eds.). (1988). Community in America The challenge of habits ofthe heart. University of California Press, 1-34, 269-288. Rich, Paul (1999, September). American voluntarism, social capital, and political culture. Annals of the American Academy, 565, 15-34. Roberts, Russell D & Wolkoff, Michael J. (1988, Spring) Improving the quality and quantity of whole blood supply: Limits to voluntary arrangements [Duke University]. Journal ofHealth Politics, Policy and Law,I3(1), 167179 .** Rohrbaugh John, McClelland, Gary & Quinn, Robert ( 1980). Measuring the relative importance of utilitarian and egalitarian values: A study of individual differences about fair distribution. Journal of Applied Psychology, 65(1), 34-49. Rohrbaugh, John and Paul Wehr (1978). Judgment analysis in policy formation: A new method for improving public participation [Published by Elsevier North-Holland, Inc.]. Public Opinion Quarterly by The Trustees of Columbia University, 521-532 284

PAGE 298

Rosenberg, Alexander (1988). Philosophy of social science (2"d ed., Dimensions of philosophy series). Boulder: Westview Press, Inc. Rosenblum, Nancy L. (1998). Membership & morals: The personal uses of pluralism in America. New Jersey: Princeton University Press. Rosenhan, David L., Peter J. Salovey, Jerzy Karylowski, and Kenneth Hargis. (1981). Emotion and altruism. 233-248. In, Rushton, J. Philippe and Richard M. Sorrentino, eds. ( 1981 ). Altruism and helping behavior: Social, personality and developmental perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates. Rundall, Christine S. & Weiss, David J. (1994, January/March). Nurses' fear of contagion: A functional-measurement analysis. Medical Decision Making, 14(1), 40-45. Rushton, J. Philippe. (1981). The altruistic personality. 251-266. In, Rushton, J. Philippe and Richard M. Sorrentino, eds. (1981 ). Altruism and helping behavior: Social, personality and developmental perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates. Rushton, J. Philippe. (1984). The altruistic personality: Evidence from laboratory, naturalistic and self-report perspectives. 271-290. In, Staub, Ervin, Bar-Tal, Daniel, Karylowski, Jerzy, and Reykowski, Janusz (Eds.). (1984). The development and maintenance of pro social behavior: International perspectives on positive morality. New York: Plenum. Rushton, J. Philippe, Roland D. Chrisjohn and G. Cynthia Fekken (1981). The altruistic personality and the self-report altruism scale. Personality Individual Differences, 2, 293-302. Rushton, J. Philippe, & Sorrentino, Richard M. (Eds.). (198la). Altruism and helping behavior: Social, personality and developmental perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Rushton, J. Philippe, & Sorrentino, Richard M. (1981b). Altruism and helping behavior: An historical perspective. 3-16. In, Rushton, J. Philippe, & Sorrentino, Richard M. (Eds.). (1981). Altruism and helping behavior: Social, personality and developmental perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. 285

PAGE 299

SAS/STAT User's Guide, Volume 1, ACECLUS-FREQ. Version 6, Fourth Edition. ( 1990) Cary NC: SAS Institute, Inc. Sabatier, Paul A., & Jenkins-:Smith, Hank. C. (Eds). (1993). Policy change and learning: An advocacy coalition approach. Boulder: Westview Press. Salamon, Lester M., & Anheier, Helmut K. (1997, January/February, Special Edition). The civil society sector. Society, 60-65. Salvatore, Steve, medical correspondent (1999, Washington (CNN), October 6). House holds hearing on blood safety and supply. CNN.com [On-line], 10/07/99. Available: cnn.com/HEALTH/991 0/06/blood.supply/index.html San Diego Blood Bank. (1999) personal communications, February. Sapolsky, Harvey M., & Finkelstein, StanN. (1979). Blood policy revisiteda new look at "the gift relationship". The Public Interest, 46, 15-27. Schall, Ellen (1995). Learning to love the swamp: Reshaping education for public service. Journal ofPolicy Analysis and Management, 14(2), 202-220. Schambra, William A. (1997, Fall). Building community top-down or bottom-up? Local groups are the key to America's civic renewal. The Brookings Review, 20-22. Schindler-Rainman, Eva and Ronald Lippitt (1975). The volunteer community: creative use of human resources, second edition. NTL Learning Resources Corporation, VA. Schlager, Edella ( 1995). Policy making and collective action: Defining coalitions within the advocacy coalition framework. Pollicy Sciences, 28, 243-270. Schmidtz, David ( 1993 ). Reasons for altruism. Social Philosophy & Policy, 1 0( 1 ), 52-68. Schmitt, Neal, Gogate Jagadish, Rothert, Marilyn, Rovner, David, Holmes, Margaret, Talarcyzk, Geraldine, Given, Barbara, & Kroll, Jill ( 1991 ). Capturing and clustering women's judgment policies: The case of hormonal therapy for menopause. Journal of Gerontology: Psychological Sciences, 46(3), 92-101. 286

PAGE 300

Schneider, Anne Larason, & Ingram, Helen (1997). Policy design for democracy. Lawrence, Kansas: University Press of Kansas. Schoemaker, Paul J .H. ( 1993). Determinants of risk-taking: Behavioral and economic views. Journal ofRisk and Uncertainty, 6, 49-73. Schon, Donald A., & Rein, Martin (1994). Frame reflection: Toward the resolution of intractable policy controversies. New York: BasicBooks, A division of HarperCollinsPublishers. Schudson, Michael (1996, March/April). Controversy. What if civic life didn't die? The American Prospect, 25, 17-19. Schwartz, Barry (1993, September). Why altruism is impossible ... and ubiquitous. Social Service Review, 314-343. Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (Ed.), Advances in experimental social psychology 10,221-279. New York: Academic Press. Schwartz, Shalom H. and Judith A. Howard. ( 1981 ). A normative decision-making model of altruism. 189-21 I. In, Rushton, J. Philippe and Richard M. Sorrentino, eds. (1981 ). A normative decision-making model of altruism. Hillsdale, NJ: Lawrence Erlbaum Associates. Seligman, Adam B. (1997). The problem of trust. Princeton, NJ: Princeton University Press. Sen, Amertya K. (1990). Rational Fools: A critique ofthe behavioral foundations of economic theory. In Mansbridge, Jane J. (Ed.), Beyond self-interest. Chicago and London: The University of Chicago Press. Shapiro, Ian, & Hacker-Cordon, Casiano (Eds.). (1999). Democracy's edges. New York, NY: Cambridge University Press. Sharpe, Virginia A., & Faden, Alan I. (1998). Medical harm: Historical, conceptual, and ethical dimensions of iatrogenic illness. New York: Cambridge University Press. 287

PAGE 301

Shilts, Randy ( 1987). And the band played on: Politics, people. and the aids epidemic. New York: St. Martin's Press. Shroyer, AL, Grover FL and Edwards FE ( 1998). 1995 Coronary artery bypass risk model: The Society ofThoracic Surgeons Adult Cardiac National Database. Annals ofThoracic Surgery, 65, 879-884. Simon, Herbert A. ( 1979). Information processing models of cognition. Annual Review ofPsychology, 30, 363-393. Sinclair, Matthew ( 1998, August). Red Cross, independents fight for blood donors. nptimes.com [On-line], 2/22/99. Available: http://www.nptimes.com/August?8/augfrol.html Singer, Peter. Altruism and commerce: A defense ofTitmuss against Arrow. Skocpol, Theda (1996, October 12). The Tocgueville problem: Civic engagement in American democracy. Presidential address for the Annual Meeting of the Social Science History Association, New Orleans. Skocpol, Theda (1996, March/April). Unraveling from above. The American Prospect, 25, 20-25: Skocpol, Theda (1997, Fall). Building community top-down or bottom-up? America's voluntary groups thrive in a national network. The Brookings Review, 16-19. Skolnick, Andrew A. (1994). Nation's blood supply dangerously low, banks call need for donors urgent. JAMA. 271(4), 255. Slavic, Paul (1972a). Psychological study of human judgment: Implications for investment decision making. The Journal of Finance, 27, 779-799. Slavic, Paul (1972b). Information processing, situation specificity, and the generality of risk-taking behavior. Journal of personality and social psychology, 22: 544-551. Slavic, Paul (2000). Risk, trust and democracy. In: Connolly, Terry, Arkes, Hal R., & Hammond, Kenneth R. (2000). Judgment and decision making: An interdisciplinary reader (2"d ed.). Cambridge, UK: Cambridge University Press. 288

PAGE 302

Slovic, P., Fischoff, B. and Lichtenstein, S. (1980). Facts and fears: Understanding perceived risk. In R. Schwing & W.A. Albers. Jr. (Eds.) Society risk assessment: How safe is safe enough? Plenum: New York Slovic, Paul, Baruch Fischhoff, and Sarah Lichtenstein (1982). Facts versus fears: Understanding perceived risk. In Daniel Kahneman, Paul Slovic, and Amos Tversky, eds., Judgment under uncertainty: Heuristics and biases. Cambridge, U.K.: Cambridge University Press. Smith, V. Kerry, Desvousges, William H., & Payne, John W. (1995). Do risk information programs promote mitigating behavior? Journal of Risk and Uncertainty, 203-221. Smith, Tom W. (1997), Factors relating to misanthropy in contemporary American society, Social Science Research. 26. 170-196. Sober, Elliott, & Wilson, David Sloan ( 1998). Unto others: The .evolution and psychology of unselfish behavior. Cambridge Massachusetts: Harvard University Press. Solomon, Robert C., & Flores, Fernando (2001). Building trust in business, politics, relationships, and life. Oxford and New York: Oxford University Press. Solow, Robert, M. (1971). Blood and thunder. The Yale Law Journal, 80, 1696-171 L Spears, Tom, The Ottawa Citizen staff(1997, March 11). Blood shortage threatens Kanata teen's life: Low supplies delay surgery. Ottawacitizen.com [On line], 9/15/98. Available: http://www.ottawacitizen.com/city/970311/853809.html Spurr, Stephen J. (1993 Spring). The proposed market for human organs. Journal ofHealth Politics, Policy and Law, 18(1), 189-201. Starr, Douglas (1998). Blood: An epic history of medicine and commerce. New York: Alfred A. Knopf. Starr, Paul (1982). The social transformation of American medicine: The rise of a sovereign profession and the making of a vast industry. New York: BasicBooks, A Division of HarperCollinsPublishers. 289

PAGE 303

Stewart, Robert M. (1984). Morality and the market in blood. Journal of Applied Philosophy, 1(2), 227-237. Stewart, Thomas R. (1988). Judgment analysis procedures. In: Brehmer, Berndt, and C.R. B. Joyce, (Eds.). (1988). Human Judgment: the SJT view. New York: North-Holland. Stokey, Edith, & Zeckhauser, Richard (1978). A primer for policy analysis. New York and London: W.W. Norton & Company. Stolle, Dietlind, & Rochon, Thomas R. (1998, September). Are all associations alike? American Behavioral Scientist, 42(1), 47-65. Stone, Deborah A. ( 1997, 1988). Policy paradox: The art of political decision making. New York: W.W. Norton & Company. Summer, William Graham (1911). Folkways: A study of the sociological importance of usages, manners, customs, mores and morals. Boston: Ginn. Surgenor, Douglas, Wallace, Edward L., Hao, Steven H.S., &, and Chapman, Richard H. (1990). Collection and transfusion ofblood in the United States, 1982-1988. The New England Journal ofMedicine, 322 (23), 1646-1651. Swartz, Thomas R. and Frank J. Bonello (1988). Taking sides. Clashing views on controversial economic issues (4th ed.). The Dushkin Publishing Group, Inc. Tabachnick, B. G., and Fidell, L.S. ( 1989). Using multivariate statistics. New York: HarperCollins Publishers. Thaler, Richard H. (1992). The winner's curse: Paradoxes and anomalies of economic life. Princeton, New Jersey: Princeton University Press Tomasulo, Peter, M.D. (1995) Transfusion Alternatives: Impact on Blood Banking Worldwide. In: Winslow, Robert M., Kim D. Vandegriffand Marcos Intaglietta Eds. ( 1995). Blood Substitutes: Physiological Basis of Efficacy. Boston: Birkhauser. Tommasi, Mariano, & Ierulli, Katheryn (Eds.). (1995). The new economics of human behavior. New York: Cambridge University Press. 290

PAGE 304

Titmuss, Richard M. (1971). The gift relationship: From human blood to social policy. New York: Pantheon Books, A Division of Random House. Tukey, John W. (1977). Exploratory data analysis. Reading, MA: Addison-Wesley Publishing Co. Tullock, Gordon (1972). Economic Imperialism. In: Buchanan, James M., & Tollison, Robert D. (Eds.). (1972). Theory of public choice: Political applications of economics. Ann Arbor: The University of Michigan Press. Tversky, Amos, & Dahneman, Daniel ( 1981 ). The framing of decisions and the psychology of choice. Science, 211, 453-8. Valelly, Richard M. (1996, March/April). Couch-potato democracy. The American Prospect, 25, 25-26. Viscusi, W. Kip, Magat, Wesley A., & Forrest, Anne (1988). Altruistic and private valuations of risk reduction. Journal ofPolicy Analysis and Management, 1(2), 227-245. von Schubert, Hartwig ( 1994). Donated blood gift or commodity? Some economic and ethical considerations on voluntary vs. commercial donation ofblood. Social Science Medicine, 39(2), 201-206. Wakefield, Jerome C. (1993, September). Is altruism part of human nature? Toward a theoretical foundation for the helping professions. Social Service Review, 406-458. Weimer, David L. (1992). Claiming races, broiler contracts, heresthetics, and habits: ten concepts for policy design. Policy Sciences, 25, 135-159. Weimer, David L. (1998). Policy analysis and evidence: A craft perspective. Policy Studies Journal, 26(1), 114-128. Weiss, Rick (2001, March). Body Bazaar explores today's burgeoning market for human Tissue. Scientific American. Weisbrod, Burton (Ed.) (1998). To profit or not to profit: the commercial transformation of the nonprofit sector. Cambridge: Cambridge University Press. 291

PAGE 305

Whittington, Keith E. (1998, September). Revisiting Tocqueville's America. American Behavioral Scientist, 42(1), 21-32. Wiesenthal, David L., Austrom, Douglas, & Silverman, Irwin (1983). Diffusion of responsibility in charitable donations. Basic and Applied Social Psychology, 1{1), 17-27. Wigton, R.S., Hoellerick, V.L., and Patil, K.D. (1986). How physicians use clinical information in diagnosing pulmonary embolism: An application of conjoint analysis. Medical Decision Making, 6, 2-11. Wildavsky, Aaron (1988). Searching for safety. New Brunswick, USA and Oxford, UK: Transaction Publishers. Wilson, James Q. (1993). The moral sense. New York, London and Toronto: The Free Press. Wilson, John, & Musick, Marc ( 1998). The contribution of social resources to volunteering. Social Sciences Quarterly, 79(4), 799-814. Winslow, Robert M. (1998). Chapter 3. A physiologic basis for the transfusion trigger. In, Rudolph, Alan S., Rabinovici, Reuven, & Feuerstein, Giora Z. (Eds.) Red blood cell substitutes: Basic principles and clinical applications. New York: Marcel Dekker, Inc. Winslow, Robert M., Kim D. Vandegriff and Marcos Intaglietta Eds. (1995). Blood Substitutes: Physioloigcal Basis of Efficacy. Boston: Birkhauser. Wolfe, Alan (1997, Fall). Is civil society obsolete? Revisiting predictions ofthe decline of civil society in whose keeper? The Brookings Review, 9-12. Wuthnow, Robert (1993, September). Altruism and sociological theory. Social Service Review, 343-357. Wuthnow, Robert (1999). The role of trust in civic renewal. In Fullinwider, Robert K. (Ed.), Civil Society, Democracy, and Civic Renewal. (pp. 209-230). Boulder: Rowman & Littlefield Publishers, Inc. Youniss, James, McLellan, Jeffrey A., & Yates, Miranda (1997, March/April). What we know about engendering civic identity. In, Edwards, Bob, & Foley, Michael W. (Eds.). Social capital, civil society, and contemporary democracy. American Behavioral Scientist, 40(5), 620-631. 292

PAGE 306

Zajac, Edward E. (1995). Political economy of fairness. Cambridge, Massachusetts: The MIT Press. Zedeck, Sheldon, & Kafry, Ditsa (1977). Capturing rater policies for processing evaluation data. Organizational Behavior and Human Performance, 18, 269294. Zey, Mary (1992). Decision making: Alternatives to rational choice models. Newbury Park, CA: Sage. In Zey, Mary (1998). Rational choice theory and organizational theory: A critique. Thousand Oaks: Sage Publications. Zey, Mary (1994). Review of"Rational Choice Theory: Advocacy and Critiques," edited by James Coleman and Thomas J. Fararo. Social Forces 77:273-75. In Zey, Mary (1998). Rational choice theory and organizational theory: A critique. Thousand Oaks: Sage Publications. Zey, Mary ( 1998). Rational choice theory and organizational theory: A critique. Thousand Oaks: Sage Publications. 293