Key diffusion variables in effective referenda campaigns

Material Information

Key diffusion variables in effective referenda campaigns a comparative study
Faatz, Jeanne
Publication Date:
Physical Description:
vii, 177 leaves : illustrations ; 29 cm


Subjects / Keywords:
Referendum ( lcsh )
Referendum -- Case studies -- Colorado -- Denver Metropolitan Area ( lcsh )
Referendum ( fast )
Colorado -- Denver Metropolitan Area ( fast )
Case studies. ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )
Case studies ( fast )


Includes bibliographical references (leaves 170-177).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Department of Communication.
Statement of Responsibility:
by Jeanne Faatz.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
16859932 ( OCLC )
LD1190.L48 1985m .F32 ( lcc )

Full Text
B.S., University of Illinois, 1962
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Arts
Department of Communication
Jeanne Faatz
1985 ?

This thesis for the Master of Arts degree by
Jeanne Faatz
has been approved for the
Department of
Commun icat ion

Faatz, Jeanne (M.A., Communication)
Key Diffusion Variables in Effective Referenda Campaigns: A
Comparative Study
Thesis directed by Associate Professor Samuel A. Betty
In thirty-nine states and two territories important public
policy decisions can be made directly by the people through
referenda, wherein the electorate votes on proposed statutory
or Constitutional changes. Persons wishing to promote or defeat
these referenda will want to develop communication strategies
which persuade voters to support the advocated position on Election
A review of diffusion of innovation I iterature points
to key variables that may affect the outcome of an election.
This thesis summarizes the literature and tests the hypothesis
that effective referenda campaigns contain key diffusion variables:
use of credible change agents, use of opinion leaders, use of
a persuasive message, use of appropriate channels of communication,
and use of measurement tools to assess client attitudes and to
develop campaign strategy.
This hypothesis is tested through a comparative case study
of tv/o Denver transit elections, one successful and one unsuc-
cessful. In 1973 and in 1980 Denver voters were asked to approve
tax increases for funding rapid transit. The campaigns promoting
the tax increases have been recreated through documentary research
and interviews with individuals who had first-hand knowledge about
either campaign. The appearance and effect of key variables in

one case have been compared with the presence and effect of the
same variables in the second case. The study shows that the
winning campaign used the key variables according to the principles
of innovation diffusion; the losing campaign did not. The results
of the comparative study support the hypothesis.

1. INTRODUCTION ..................................
2. DIFFUSION THEORY ..............................
An Overview .................................
Use of Credible Change Agents ...............
Need for Credibility ......................
Likemindedness is Important................
Recipient Attitudes .......................
Effect of Time ............................
Implications for Campaign Consultants ....
Use of Opinion Leaders ......................
Characteristics of Opinion Leaders ........
Identification of Opinion Leaders .........
Women as Opinion Leaders ................
Influence of Opinion Leaders ............
Effect of Homophily .......................
Implications for Campaign Consultants ....
Use of Appropriate Channels of Communication
Role of Interpersonal Channels ............
Timing is Important .......................
Interpersonal Channels and Audience
Relationship .............................

Mass Media Channels ..................................... 35
Mass Media as an Information Source ............... 35
Credibility of Mass Media ......................... 36
Media and Audience Relationship ................... 38
Relationship of "News" and Television
Commerc i a I s ................................ 40
Interrelationship of Media Channels ............... 41
Importance of Print Media, Radio and
Telephone ............................................ 42
Media as a Persuasive Tool .......................... 44
Mass Media Drawbacks .............................. 45
Implications for Campaign Consultants ................. 46
Use of a Persuasive Message ......................... 47
Message Purpose is to Influence ................... 47
Perceived Attributes are Important ................ 50
Advertising: The Promotion of Perceived
Attributes ........................................ 52
Selective Attention of the Message Receiver ........... 54
Repetition of Message ................................ 56
Effect of Advertising on the Minimally-
Informed Receiver ................................. 58
Differences Between Candidate and Referenda
Elections .......................................... 58
Use of Fear as a Motivational Appeal .............. 60
Use of Ambiguous Messages ............................. 62
Can Information Campaigns be Persuasive? .......... 65
Implications for Campaign Consultants ................. 67
Use of Measurement Tools to Assess Client
Attitudes and to Develop Campaign Strategy ............ 68

Types of Surveys .................................. 69
Weaknesses of Polling ............................. 72
Use of Information from Polls ..................... 74
Use of Information in Political Advertising ....... 78
Formats for Collecting Information .................... 80
Implications for Campaign Consultants ................. 81
Summary ................................................. 82
3. METHODOLOGY ............................................... 98
4. HISTORY OF THE CAMPAIGNS .............................. I 10
5. A COMPARATIVE STUDY ................................. 115
Use of Opinion Leaders ................................. 115
Use of Measurement Tools ............................... 118
Use of Persuasive Message .............................. 121
Use of Appropriate Channels of Communication ...... 128
Use of Credible Change Agents ......................... 139
Conclusions ............................................ 147
APPENDIX .......................................... 152

Transportation has long been key to the viability of
cities. Access to points of transportation between cities, such
as waterways, railroads, and highways, often suggested the location
of new communities. As these communities grew, internal systems
of mobility became important matters of public concern.
Denver metropolitan area residents were asked to make
a crucial decision about their mobility in 1973should rapid
transit be built In their community? The state legislature had
authorized the question to be put to a vote. Did residents value
improved mobility enough to choose to tax themselves for it?
Each qualified voter had an opportunity to express his
opinion on election day. Did he wish to continue depending on
the automobile, complete with its air pollution and noise? Did
he wish to lose travel time in heavy traffic? Did he wish to
continue with the status quo or would he rather choose a different
Iifestyleone in which places would be easier to go and where
better public transportation planning could lead to better land
use planning? Did he want a better place to live?
On election day the public policy decision was made.
More than fifty-seven percent of the voters said "Yes" to better

Seven years later the Denver metropolitan area qualified
voters were asked once again to vote for the construction of rapid
transit. Improved mobility was still the issue. A better life-
style was still at stake. The public had another voice at the
ballot box. This time more than fifty-four percent said "No.
Why did one election elicit a favorable vote for rapid
transit while the other was rejected?
With important public policy decisions such as these
depending on the outcome of such referenda elections, it becomes
important to understand the strategies which guide referenda
campaigns. By understanding the strategies used to successfully
advocate a position, It may be possible to identify and develop
communication strategies which enhance the odds of success.
With such important matters of public policy being decided
directly by the people, referenda must be acknowledged as an
Important part of the polItical process. It is, in fact, a right
reserved to the people In the Constitutions of more than seventy-
five percent of the states.* Any strategies which guide the
outcome of such decisions In such a large portion of the United
States of America could ultimately have considerable effect on
that state's and nation's future.
Before examining communication theories and strategies
which may help to shape the citizens' quality of life, it is
important to review the provisions of the political process which

allow policy decision-making directly by the people.
In all states, legislatures and other policy-making bodies,
comprised of members elected by the citizenry, make many polIcy
decisions through authority delegated by the people In their
Constitutions and city charters. According to Thomas M, Durbin,
Legislative Attorney, thirty-nine states and two territories have
Constitutional provisions which also allow the electorate to vote
directly on proposed statutory of Constitutional changes. Such
proposed changes can be placed on the ballot by referendum, the
process by which the state legislature or other authorized policy-
making body refers proposed legislation or Constitutional changes
to the electorate.*5
In eighteen states and one territory such legislative
changes also can be placed on the ballot through the initiative
process, also known as cite petitioning.^ Betty Chronic, Director
of Elections and Licensing of the Department of State in Colorado
defines cite petitioning as a formal request developed into a
written ballot issue draft by citizens who wish to have proposed
changes in legislation of Constitutional provisions submitted
to the electorate.^
Both of these methods of submitting proposed legislation
or Constitutional amendments are used in Colorado.^ Citizens
who find themselves advocating or opposing the content of the
issues to be voted upon often find themselves as newcomers to
the political campaign process and look for skilled consultants
to guide promotion of their side of the story.

Skilled consultants are available to guide such campaigns,
according to political scientist Larry Sabato, though not in great
numbers. In fact, Sabato contends, It was during an Initiative
campaign that modern political consultants first had a major
California was in the forefront of the popular initiative
and referendum movement and, because of that, had an exceptionally
long ballot. In 1933, the California legislature passed a bill
authorizing a flood control and irrigation development in northern
California (called the Central Valley Project), which the Pacific
Gas and Electric Company (PG&E) believed to be a threat to private
power. The utility promptly launched a ballot initiative to
reverse the decision.
Then Central Valley Projects proponents hastily enlisted
Clem Whitaker, a Sacramento newsman and press agent, and Leone
Smith Baxter, a public relations specialist, to mastermind a
campaign to defeat PG&E's initiative. On a limited budget of
$39,000, and using radio and newspaper appeals, Whitaker and Baxter
managed to save the Central Valley Project.
Not only did PG&E hold no grudge, it actually put Whitaker
and Baxter, who had incorporated themselves, on annual retainer
as consultants. There were two decades of smooth sailing for
the Whitaker and Baxter firm, operating out of San Francisco,
and the lack of extensive competition enabled it to post a ninety
percent success rate in seventy-five campaigns. Eventually the
free-market system gave rise to competitive consultants and

therefore reduced Whitaker and Baxters edge and win-loss record.
Two decades later, political consultants (especially for candi-
dates) had become a campaign standard across the United States.
Professional literature reveals relatively few consultants
who specialize in initiative or referenda campaigns. More than
a few consultants refuse such campaigns, evaluating them much
as consultant Douglas Bailey did after a couple of disappointing
initiative campaigns. "Frankly, said Bailey, "a...candidate
race is a far more exciting thing normally than an issue (Initia-
tive or referenda) race.... I can tell stories about how difficult
it is to get associations and organizations of people to take
decisive action. I would much rather work with a candidate who
can say yes or no than with a board of fifty people who must say
yes or no."*
A few consultants who have seized upon ballot measures
for much of their livelihood naturally disagree with Bailey's
assessment. They believe the political consultant is almost
certainly given greater leverage In many referendum or initiative
efforts. First of all, there Is not party involvement or partisan
identification to contend with either in the workings of the
campaign or the electorate. Second, most groups sponsoring (and
opposing) ballot measures are we 11-funded and have a mass member-
ship. Moreover, these groups often have little electoral experi-
ence and desperately need expert campaign guidance.* Winner
Wagner and Associates is one of several California firms that
handle ballot questions. Forty percent of its work is in this

area, with the balance controlled by nonpolitical "issues communi-
cation" for major corporations and agencies. Chuck Winner hotly
disputes the contention that ballot campaigns are less interesting
that the candidate variety, terming this previous initiative
efforts "Intense and intriguing." He also finds them more profit-
able and less contentious than a candidate and his family and
his staff.* *
Winner Is not alone. The San Francisco based Woodward
and McDowell agency had great success in defeating an anti smoking
ballot proposition in California and became the first choice of
the tobacco industry to fight similar initiatives all over the
country. Butcher-Forde agency is handling tax and spending limita-
tion referenda across the country after its guidance of Howard
Jarvis* Proposition 13 in California, and Solem and Associates
of San Francisco specializes In rent-control measures.*^
Whereas some consultants prefer the initiative and refer-
enda campaigns, citing specific advantages of noncandidate work
(In addition to finding the field quite lucrative),*^ there are
not great numbers of professionals who specialize in this growing
area of campaign activity. It seems that the demand for indi-
viduals trained for referenda or initiative promotion or opposition
will continue to increase.
New people wishing to enter the consultant arena for
initiatives and referenda will benefit from more information about
campaigning as their skills shape important policy decisions.
It is this probIem of a need for more information which this paper

Such individuals would benefit from knowledge about the
diffusion of innovations. A thorough examination of diffusion
theory and of measurement techniques to assess voter attitudes
will enhance their ability to promote complex Issues in a manner
which will enhance their ability to promote complex issues in
a manner which will persuade and mobilize people to support their
position of Election Day and is one ob iective of this paper.
Another ob iective is to learn more about potential key
campaign variables through examination of such literature and
to compare and contrast the use of these key variables in a
specific setting.
In order to examine the use of diffusion theory and
attitude measurement In a specific setting, this paper will
investigate the use of such elements in two referenda campaigns
conducted in the Denver metropolitan area on the same topic.
In 1973, a referendum calling for a tax Increase for Improved
public transit was successfully submitted to the electorate.
In 1980, a similar referendum calling for increased taxes for
improved public transit was defeated by the Denver area voters.
The purpose of this thesis is a comparative analysis of
the two public transit referenda in the Denver metropolitan area.
In completing this analysis, this researcher expects that
the following hypothesis will be substantiated: Effective
referenda campaigns contain key diffusion variables: use of
credible change agents, use of opinion leaders, use of persuasive

message, use of appropriate channels of communication, and use
of measurement tools to assess client attitudes and to develop
campaign strategy.
To properly test this generalization, definition of terms
is essential. As used In this study, these terms have the
following meanings:
EFFECTIVE: Effective means winning. As stated by William
Butcher of Butcher-Forde consulting agency, "When it comes to
political consulting, the name of the game is winning on Election
Day."*4 Effective campaigns produce at least fifty percent-plus-
one votes on Election Day.
REFERENDA: Proposed measures submitted to the electorate
by the legislature or other policy-making body legally authorized
to submit such measures for the purpose of electorate approval.
CAMPAIGNS: "A continuous exercise in the creation,
re-creation, and transmission of significant symbols through
communication. Campaigns incorporate the active participation
of both campaigners and voters."**5 The campaigners (those
promoting an issue) try to manage the impressions voters have
of the "issues by expressing symbols they hope will appeal to
the electorate."*
KEY: Significant enough (singly or in combination with
other events) to jeopardize achieving a fifty percent-plus-one
DIFFUSION: "The process by which new ideas are communi-
cated to the members of a social system."*^

VARIABLES: "Influential factors."18
USE: Selection and employment of.
CREDIBLE: "The degree to which a communication source
Is perceived as trustworthy and competent by the receiver."*^
CHANGE AGENT: "The helper, the person or group who Is
attempting to effect change."^8
OPINION LEADER: An individual who approves or sanctions
an Innovation and who, by the nature of the social power he
possesses and his representation of the social system's norms
and values, can add credibility to the innovation: i.e., "Provide
justification, the license to act."^*
PERSUASIVE: Presentation of facts so forcefully or master-
fully that people are aroused to adopt the position advocated.
MESSAGE: The words and symbols that express the perceived
attributes of the innovation.
APPROPRIATE: Choice of a communication channel which
is designed to meet the receiver's needs based on his stage in
the innovation-decision process.
CHANNELS OF COMMUNICATION: The means by which a message
gets from a source to a receiver.^
MEASUREMENT TOOLS: In this context the instrument used
to quantify public opinionone of various polling methods and,
ultimately, the vote itself.
CLIENT: Member of the social system which comprises the
ATTITUDE: "A relatively enduring organization of an

Individuals beliefs...that predisposes his actions.
STRATEGY: A step-by-step plan, often written, for
achieving fifty percent-plus-one of the votes cast.
INNOVATION: "Any idea, process or thing that is perceived
of as new."^
In this thesis, the principles of diffusion will be
examined at length according to scholarly writings in professional
journals and other publications, and according to communication
specialists, undoubtedly apply to both initiatives and referenda.
This paper, however, will be limited to non-petition referenda
in order to minimize the pre-campaign variable of petition
Chapter 2 will review diffusion literature, including
a generalized discussion of theories and a detailed discussion
of literature relevant to key variables. Research will draw from
literature In the disciplines of communication, social change,
and marketing.
Chapter 3 will examine the methodology used in this study,
including a description of the research design, methods of data
collection and processing, and methodological assumptions.
Chapter 4 will outline the history of the two transit
Chapter 5 will analyze the data, comparing likenesses
and differences in each of the key variables. It will also Include
conclusions about the role.of the key variables.
Hopefully, upon conclusion of this comparative analysis,

students of communication theory will recognize the practical
application of diffusion theory within the concept of referenda,
and consultants who promote referenda will approach the challenge
with a deeper understanding of the process of planned change and
of ways to develop more effective strategies using key variables
in order to increase their success ratios in the win-loss columns.
By doing so, they will have enormous impact on policy making and
will help to determine the direction of the future.

* Thomas M. Durbin, "Initiative, Referendum and Recall:
A Resume of State Provisions, Congressional Research Service,
Report No. 81-63A, 751/110, March 9, 1981.
Durbin, p. unk.
^ David H. Everson, "Ballot Initiatives: How Popular
Are Popular Referendums?," Legislative,Pol icy. March/April 1984,
p. 42.
^ Durbin,
J Betty Chronic, Director of Elections and Licensing,
Department of State, Colorado, Telephone interview, November 29,
Chronic, interview.
7 Larry J. Sabato, The Rise of Political Consultants^
New Wavs of,.Winning Elections (New York: Basic Books, Inc., 1981),
p. II.
Sabato, pp. 11-12.
Sabato, p. 55.
Sabato, p. 56.
Sabato, pp. 56-57.
Sabato, pp. 56-57.
Sabato, p. 56.
Sabato, p. 17.
5 Dan NImmo, Po l i t lea l Commun I cat i on and Pub I. ic. Qd i n Ion
In America (Santa Monica, California: Goodyear Publishing Co.,
1978), p. 371.
NImmo, p. 371.

^ Everett M. Rogers with F. Floyd Shoemaker, Communication
of- Innovations. 2nd ed. (New Yorks The Free Press, 1971), p. 7.
*8 Ann Bonar Blalock and Hubert M. Blalock, Jr., Introduc-
tion to. Socf a I- .Research. 2nd ed. (Englewood Cliffs, NJ: Prentice-
Hall, 1982), p. II.
1 * Rogers and Shoemaker, p. 244*
Rogers and Shoemaker, p. 227.
1 Rogers and Shoemaker, p. 280.
Rogers and Shoemaker, p. 251.
Rogers and Shoemaker, p. 109.
Jon WInterton, Professor of Communication, University
of Colorado at Denver, November 24, 1982.

An Overvl.ew
A referendum campaign involves collective decision-making.
At election time fifty percent-plus-one of the people to be
affected by a decision and who are qualified to vote must approve
a measure for change to occur. Rogers and Shoemaker explain the
complexity of such decisions by stating, "The collective decision
process is really composed of a multitude of individual decision:
to initiate a new idea in a social system to adapt the new
proposal to local conditions, to sanction the Idea, to ^support
the innovation, and so on. Each of these different behaviors
may be carried out by different individuals in the collectivity."*
To provide a general framework for analyzing collective-
innovation-decision-making, Rogers and Shoemaker suggest that
such diffusion occurs in stages:
1. Simulation Someone becomes aware that a need exists
for a certain innovation within a social system.
2. Initiation The new idea receives Increased attention
by members of the social system and is further adapted to the
needs of the system.
3. Legitimation Collective Innovation is approved

or sanctioned by the power-holders or opinion leaders (those who
Informally represent the norms and values of the social system).
4. Decision To act, by members of the social system
through conducting a survey, holding a referendum, circulating
petitions, etc.
5. Action Implementation of the new idea.^
As Rogers and Shoemaker point out, one consequence of
collective decision-making Is Increased stability of change, once
effected, because a decision made by a group can generally be
changed only by that group.
The diffusion model for collective decision-making closely
parallels the individual decision-making model, also suggested
by Rogers and Shoemaker: Individuals pass from AWARENESS (knowl-
edge of an innovation) to PERSUASION (forming a favorable or
unfavorable attitude about the Innovation) to a DECISION (to adopt
or reject) and CONFIRMATION of the decision (as the individual
evaluates his choice).^
These models are not only ones suggested by researchers
of diffusion. Donohew and Springer suggest an information-seeking
model (a "bottom-up model as opposed to the Rogers-Shoemaker
source-oriented, trickle-down model), suggesting that human beings
are capable of initiating actions rather than "merely serving
as reactive targets of persuasion."^
The Donohew-Springer model focuses on the individual at
the heart of the change process. It also has five stages:
I) Give attention to a stimulus cluster, such as defining a

problem; 2) Compare wl+h cognitive Image of reality, set prior-
ities; 3) Search for information, assess situation; 4) Take action;
and 5) Revise image.
In referenda campaigns, the Rogers-Shoemaker model would
more closely resemble the flow of communication commonly experi-
enced. Perhaps a little more insight into the components of the
Rogers-Shoemaker model v/ould clarify the process of diffusion.
The stimulator is very often an outsider to the social
system or a member of the system who has much contact with outside
systems (i.e., cosmopolite, message-oriented). He acts as a gate-
keeper, controlling the flow of which new ideas of a collective
nature enter the social system.
The initiator incorporates the innovation into a specific
plan of action that is adapted to the social system, including
the ability to predict certain consequences of the new idea, once
it is adopted. He is an "insider," a localite, whose advantage
is that he knows the system.
Change agents are usually stimulators or Initiators.
According to Rogers and Shoemaker, they perform a sequence of
roles. They develop need for change, develop a rapport and empathy
with clients (including building credibility), diagnose the problem
to determine the proper alternative to pursue, motivate the client
to want to change, promote compliance (through action or behavior)
with the change program advocated, give reinforcing messages to
those who have adopted, and eventually step out of the picture
after the client learns to rely on himself.^

According to Jon Winterton, Professor of Communication
at the University of Colorado at Denver, change involves risk.
Even a bad situation is sometimes preferable to a change for
betterment because RISK is minimized and people are spared
critically evaluating their old ways of "doing business/'
In a farming study of awareness and trial of Innovation
by Gartrell and Gartrell, awareness of the innovation was trans-
lated Into trial at a fairly high rate when risk was almost
nonexistent, due to total government assistance and underwriting
of the experiment. The study also showed that the higher the
education level of the clients (members of society) and higher
the level of social status, the faster the adoption of the innova-
Legitimation is the approval or sanctioning of an innova-
tion by "those who informally represent the social system In its
norm and values and in the social power they possess."*9 Opinion
leaders perform this role. Whereas the role is primarily one
of screening ideas for approval, the legitimizer may alter or
modify the proposals brought to him.
Initiators may circumvent the legitimization stage, but
when legitimizers are Ignored, they are more likely than not to
scuttle the collective Innovation.**
The decision stage is the determination of a favorable
or unfavorable attitude about the Innovation by members of the
social system and the expression of the decision at the polls
on Election Day, in the case of a referendum.

The Individuals PERCEIVED participation in the process
of reaching the collective decision is more Important than his
objective participation (how much he participates according to
others) in explaining his satisfaction with the collective
I ?
If he participates fully in the decision-making process,
an individual learns that most others in the system are willing
to go along with the decision, and he becomes more satisfied with
it himself. Moreover, his input will make the content of the
decision more appropriate to his needs. In addition, widespread
participation allows the opinion leaders In the system to assume
a major part In making the decision. Hence, the opinion leaders
position is reinforced, and the members are encouraged to abide
by the decision and be more satisfied with it.^
The action stage Is the implementation of the new idea
and begins If the referendum is successful.
Promoters of referenda, such as those promoting the Denver
transit referenda, should be aware of each of the above stages
In collective decision-making. Although the promoters usually
enter the decision-making stage after a stimulator has already
identified a need, they must be prepared to assist with the execu-
tion of the phases and to assure the adequacy of each phase's
development. More careful examination of aspects Important to
this process will aid the promoters in their development of
strategy and their critical evaluation of each stage.
They must also be crucially aware of both the populace's

reluctance to challenge their past decisions and of the necessity
for the promoters to constantly challenge their own decision
strategy, despite the tension and conflict they will encounter.
At any given time, readjusting the strategy for appealing to the
voters may make the difference between a successful and unsuc-
cessful referenda campaign.
Use .Qfr,Xr^di.fc|ig.iChanag,./\ One of the most Important decisions a campaign consultant
can make in promoting a referendum is the selection of a change
agent. The consultant may select an individual or an agency to
serve the change agent role, but he must be certain that the change
agent possesses certain traits. These traits and variables
surrounding them will be examined in this section.
A change agent may have a mammoth task before him. His
technical knowledge and his widespread social relationships provide
a firm basis for calling new ideas to the attention of the system's
leaders. He can Identify legitimizers and urge their usage.
He can help the members of the society evaluate the financial
and social costs of the collective decision they are weighing,
in order to achieve his goal of moving the client's attitudes
in the direction he advocates, he must" be credible.*^
Need-.for .Credibil ity
As Rogers and Shoemaker state, "A change agent's success
is positively related to his credibility in the eyes of his
clients."*-* In studies by Burgoon, et al., criteria for

credibility include competence, composure, extroversion, and socia-
biIi ty.16
According to Herbert Kelman, "An agent possesses credi-
bility if his statements are considered truthful and valid and
hence worthy of serious consideration. Hovland, Janis, and Kelley
distinguish two bases for credibility: expertness and trustwor-
thiness. In other words, an agent may be perceived as possessing
credibility because he Is likely to KNOW the truth, or because
he Is likely to TELL the truth. Trustworthiness, In turn, may
be related to over-all respect, IIkemindedness, and lack of vested
Conversely, several researchers have found that biased,
untrustworthy sources are not effective in producing attitude
change. Politicians are viewed as biased sources of information,
said these researchers.*
A study by James Jacard showed that confidence in the
source, as a measure of credibility, is crucial to belief change.
He states, "Whenever people are more confident in themselves than
the source, relatively little belief change results. However,
when the confidence in the source exceeds an Individual's own
confidence, belief change will be directly related to the discrep-
ancy of the assertion."1:7
Sternthal, Phillips and Dholakia discuss a study performed
by Bochner and Insko (1966) concerning the Joint effects of source
credibility and communication discrepancy. Using the topic of
the number of hours of sleep needed, the findings showed that

when the communicator was highly credible, increasing message
discrepancy enhanced persuasion. When the source was less cred-
ible, a moderate level of discrepancy was most persuasive.
Furthermore, the highly credible source induced greater Influence
than the moderately credible communicator only when discrepancy
was relatively extreme. At lower levels of discrepancy, source
credibility had no systematic effect.^
Sternthal, Phillips, and Dholakia also found that identifi-
cation of a highly credible source at the beginning or In the
middle of a message induces greater support for the communicators
position. It Is better to identify a low credibility source at
the end, the study showed. Credibility has a systematic persuasive
effect when the source is identified prior to the message but
not when identification is deferred.^*
These studies give guidance to the promoters of referenda,
both in the importance of using credible change agents and in
the placement of identification of the source should he not be
as credible as would be desirable.
LIkem faded ness is,, Important
Likemindedness, as a component of credibility, creates
more effective communication with clients, according to Rogers
and Shoemaker. Sharing certain attributes, common meanings and
interest U.e., being homophilous) creates greater change agent
success. More time for a diffusion campaign must be allowed,
however, because according to Rogers and Shoemaker, when there
Is a high degree of homophlly, information spreads horizontally,

slowing the time for diffusion.
Promoters of referenda need to select change agents who
are likeminded with their clients, also keeping In mind that the
Information may take longer to spread through the social system
because of that Iikemindedness.
Recipient Attitudes
Just as Iikemindedness with recipients enhances the effect
of a change agent, recipient attitudes affect the way a change
agent presents ideas.
Where message recipients have strong initial opinions
stored in long-term memory, the contents of a message registered
In short-term memory triggers the retrieval of the thoughts from
long-term memory. Sternthal, Phillips, and Dholakia found that
this thought retrieval Is mediated by source credibility. A highly
credible source inhibits the retrieval of counterarguments when
message recipients are negatively predisposed to an appeal, whereas
a less credible source stimulates this retrieval of support argu-
ments among those with a favorable predisposition to the message
The researchers also found that, when subjects have neutral
opinions, opinionated messages cause greater persuasion than
nonoplnIonated messages when source credibility is high but not
when it is low. If subjects have intensely negative initial
opinions, Increasing opinionatedness decreases persuasion for
both high and low credibility sources.^
In addition to preconceived ideas, personality

characteristics of the responder affect the persuasiveness of
the source, the researchers found. "Highly authoritarian indi-
viduals are strongly influenced by source credibility cures,
whereas less authoritarian people make greater use of message
cues in determining their attitude. On the other hand, when the
message cue is moderate in complexity, it will serve as the
dominant cue and source credibility is predicted to have no system-
atic effect."^
To achieve maximum effectiveness, the change agent,
according to the literature, must know preexisting attitudes of
his clients and must understand their personality character I sties.
Effect, of. .Time
There is conflicting data as to whether time reduces the
effect of source credibility. Called, In communication theory,
the "sleeper effect," it is suggested that the character of the
source is forgotten at a more rapid rate than the content of the
communication so that the influence of a highly credible source
and a low credibiIity source is similar over long periods of time.
Schulman and Worrall suggest that, according to a study they
conducted concerning the adoption of drug policy, the more crucial
test is rather that "the individual must be less likely to asso-
ciate spontaneously the content with Its source."^7
LmDlii.catlons .for Campaign ConsuItants
The literature outlined above suggests that consultants
promoting referenda would be wise to select credible agents who

are llkeminded wi+h their clients. Source credibility is espe-
cially important with highly authoritarian individuals or with
individuals who would be negatively predisposed toward the refer-
endum. Source credibility was less important with clients already
favoring the referendum or where much time will elapse between
the presentation of the change agent and the election.
Use,.of.. Op In I,on, Leaders
As the overview in diffusion of innovation theory
suggested, one of the roles of a change agent is to identify
legitimizers. Opinion leaders can serve as legitimizers.
In "Whose Opinion Do You Trust," Richmond and McCroskey
define an opinion leader as "the individual a person turns to
for advice or information on a particular topic." Lazarsfeld,
Berelson, and Gardet and Rogers and Shoemaker expand on this defi-
nition by saying that these people are seen as being more competent
and more influential than their followers. By "influential these
authors mean that the followers are more likely to accept the
opinion leaders advice than they are to accept other peoples
advice on the same subject."^9
Another way of stating this concept is that opinion leader-
ship is the degree to which an individual Is able to informally
influence other individuals attitudes or overt behavior in a
desired way with relative frequency. According to Rogers and
Shoemaker, opinion leadership is earned and maintained by the
individuals technical competence, social accessibility, and

conformity to the system's norms.
Character I.stl.cs of Leaders
What are the characteristics of opinion leaders?
Rogers and Shoemaker state, "Legitlmizers of collective
innovation decisions possess higher social status than other
members of the social system. An opinion leader has status and
Several researchers Indicate that when the social system
is modern, the opinion leaders are quite innovative; but when
the norms are traditional, the leaders also reflect this norm
In their behavior. By their close conformity to the systems
norms, the opinion leaders serve as apt models for the innovation
behavior of their followers.^
In general, when opinion leaders are compared with their
followers, they I) are more exposed to all forms of external
communication, 2) are more cosmopolite, 3) have higher social
status, and 4) are more innovative. Opinion leaders are usually
members of the social system In which they exert their influence.
These influential persons can lead In the promotion of new Ideas
or they can head an active opposition, according to Rogers and
Paul Lazarsfeld and Herbert Menzel list as characteristics
of opinion leaders: I) they have Interest in the subject matter
concerned, 2) they occupy positions In communities regarded as
giving them special competence in the matter at hand, 3) they
are accessible and gregarious and know many people, and 4) they

have contact with relevant information coming from outside their
immediate circle.34
These researchers state, "Leaders do not deviate very
far from the norms of the group they lead; if anything, they live
up to them with special consistency." As one of Lazarsfelds
colleagues stated, "Those men can better lead who are traveling
the same road as their followers but are a little ahead.
I dent i f.IcatLon of Opinion. Leaders
Whereas researchers tend to agree on the basic character-
istics of opinion leaders, there Is a distinct difference of
opinion between political science researchers and sociologists
about the methodology for identifying these leaders.
The political scientist, according to Robert Dahl, Is
concerned with decision-making and reconstructs case studies to
trace the types of power structures (coalitlonal, factional or
pluralistic). He operates on the assumption that community power
Is situational to the decision at issue.3
Studies by Lazarsfeld tend to confirm the political scien-
tist's contention that "opinion leadership is not a general charac-
teristic of a person but is always limited to particular issues."3^
One study was conducted by Katz and Lazarsfeld In 1955. The
researchers deliberately sought out people who had recently changed
their opinions or who had otherwise made new decisions in four
very different fields: political opinions, the purchase of
groceries for the household, fashions, and motion picture prefer-
ences. "Indeed there was very little overlap of leadership:

a leader in one. sphere was not especially likely to be influential
in another, unrelated sphere. Once again it turned out that
personal influence played a more frequent and effective role than
any of the mass media. And once again, the influential people
proved to be fairly evenly distributed through all educational
and income classes.... Only with regard to political opinions
was there even a moderate degree of concentration of leadership
in the highest socioeconomic status."^
On the other hand, the sociologist views community power
as a generalized ability, similar to a pyramidal structure,
according to Floyd Hunter. Sociologists use the "reputational
method" with community leaders determined through surveys.
In a study designed to yield a measure of opinion leader-
ship based on self-identification (whether respondents went to
someone during a political campaign to persuade him to vote in
a particular wayand whether someone during a political campaign
came to the respondent for suggestions), John KIngdon found that
opinion leaders had attained higher levels of formal education,
were more white than nonwhite, more male than female. He also
found, contrary to findings of some other studies of opinion
leadership (Katz), leaders are "by no means evenly distributed
among occupational strata. Using Duncans socioeconomic scale,
forty-two percent of respondents in households where the head
of the household is of high occupational status Identify themselves
as leaders of some sort, compared to twenty-three percent for
the low status and twenty-seven percent medium status."^ In

addition, Kingdon found that leaders tended to rely on print media
or combinations of sources, while nonleaders placed heavy reliance
on television as a news source.4*
In choosing their opinion leaders, change agents should
be aware of the political science and the sociology schools of
thought. As long as they consider both methods of identification,
they will not be as likely to overlook a potentially important
opinion leader who could provide invaluable support in legiti-
mizing a referendum.
Women as Op i n i on .Leaders
In a previously cited study, John Kingdon found that
opinion leaders tended to be more male than female. This study
raises the question of the role of women as opinion leaders.
In a 1975 study to determine the role of women as opinion
leaders, Richmond and McCroskey found that married adults chose
general opinion leaders of the same sex. Single female graduate
students also chose female opinion leaders.4^ "The prior research,
bias that females judge males to be more competent than other
.females held true in the present study only on the topic of
politics," said Richmond and McCroskey.4^
The researchers concluded that considerable research of
diffusion and social change suggests that adoption of Innovations
in this case, acceptance of women's opinionsis frequently earlier
among the well-educated. Thus any progress suggested by the data
should be Interpreted as an early trend toward change rather than
a clear societal pattern.44

Change agents may wish to Investigate whether more recent
research has addressed a more current reading on the societal
trend before they would select a female as the primary opinion,
leader for a political issue. The female would still be able
to have considerable influence with specific individuals, however,
so long as she possessed traits attributable to opinion leaders
in general.
I rcfIuence.of Op i n ion,Leaders
If change agents choose opinion leaders wisely, the leaders
can be highly influential. A study attempting to characterize
the persons most susceptible to the Influence of formal and
informal leaders showed that most people imputed to their leaders
opinions like their own; those who saw a discrepancy between their
own opinion and that of some of their named leaders tended in
the course of the campaign to change their opinions In the direc-
tion of those Imputed to their leaders. This effect was about
equally strong for formal and informal leaders and was strongest
among the politically most apathetic. Adjustments to the perceived
opinions of informal leaders occurred most strongly among the
least-educated and oldest segments; to those perceived opinions
of formal leaders, among the moderately educated and youngest
Rogers and Shoemaker state that, given the Influence of
opinion leaders, "rate of adoption of a collective innovation
is positively related to the degree to which the social system's
legitimizers are involved in the decision-making process."^

As was stated earlier, Initiators may circumvent the
legitimization stage, but if legitimizes are Ignored, depending
on the situation, they are more likely than not to scuttle the
collective innovation.^
Another warning by Rogers and Shoemaker: there is research
evidence that opinion leaders can be "worn out" by change agents
who overuse them. Opinion leaders may be perceived by their peers
as too much like the change agents; thus, the opinion leaders
lose their credibility with their former followers.^
Hamophl l>y
Homophlly Is defined by Rogers and Shoemaker as "the degree
to which pairs of individuals who interact are similar in certain
attributes, such as beliefs, values, education, social status,
and the Iike."^ Homophlly can affect the speed with which
information and attitudes about innovations travel through
society. Change agents must be alert to the Influence of homophily
so that they will be able to establish realistic time goals In
the promotion of referenda.
In fact, to speed the flow of Information, should time
be short, Rogers and Shoemaker encourage change agents to work
with different opinion leaders throughout the social structure.
This suggestion is made primarily to counteract the Invisible
barrier homophily can create to the rapid flow of Innovation within
a system. The researchers state, "New ideas usually enter by
means of a higher status and more innovative members of the
system. When a high degree of homophily Is present, these elite

Individuals interact mainly with each other, and there is little
"trickle-down of the innovations to the nonelite. Homophilic
diffusion patterns cause new ideas to spread horizontally, rather
than vertically, within a system. Homophily, therefore, acts
to slow down the rate of diffusion. One implicatlon of homophily
as a barrier to diffusion is that change agents should work with
different opinion leaders throughout the social structure."-^
Implications, for Samoaion, Consul.tants
Those individuals promoting referenda must recognize
opinion leadership traits and be able to identify opinion leaders
who can legitimize the issue with clients of the social system.
In addition, change agents, who have the selection of opinion
leaders within their "job description," need to be alert to the
possible overuse of opinion leaders, the most effective use of
women opinion leaders if the issue is political, and the importance
of not ignoring opinion leaders. The success of the referendum
may depend on the constructive Involvement of the appropriate
opinion leaders.
Use,.of. Appropr late,Channels^of...Communication
Whereas opinion leaders can legitimize a message, for
that message to reach the receiversi.e., the voters, in the
case of referendait must travel from the sender to the receiver
via channels of communication.
Any tradesman or handyman knows that having the right
tool for the right task is essential. Employment of channels

of communication involves two categories of decisions: I) what
channel of communication is most effective for a given purpose,
and 2) how can we maximize the effect of the channel selected.
This section will deal largely with the first category. The second
category will appear prominently in the later discussion of
designing persuasive messages.
The promoter of referenda must be familiar with two basic
channels of communicationInterpersonal (involving a face-to-face
exchange between two or more Individuals or, as in a speech, one-
to-many) or mass media ("all those means of transmitting messages
that Involve a mass medium, such as radio, television, film, news-
papers, magazines, and so on, which enable a source of one or
a few individuals to reach an audience of many").'5*
Rogers and Shoemaker state that, In the diffusion process,
"Mass media channels are primarily knowledge creators, whereas
interpersonal channels are more Important at persuading, that
is forming and changing attitudes."-^ |n other words, mass media
channels are more important at the knowledge (awareness and infor-
mation) function and interpersonal channels are more Important
at the persuasion function (creating favorable or unfavorable
Role of. I nterpersona I Channels
Discussing first the role of Interpersonal channels,
Lazarsfeld and Menzel believe that face-to-face communication
has some advantage over mass media (print or electronic) in matters
of political persuasion: persons of opposing views are more easily

reached by face-to-face communication as spontaneity reduces
filtering or selective attention. "They are less defensive when
reached; they experience the immediate gratification of the inter-
locutors approval if they come to agree with him."-^
Also addressing the role of interpersonal channels, a
study by Kenneth Starck showed that interpersonal Information
sources were regarded most important in helping the respondents
achieve value-prescribed goals. More than half of the respondents
(54.9$) indicated they looked to relatives and family (35.3$)
or friends and acquaintances (19.6$) for guidance where decisions
concerning values were Involved. They also preferred print media
over the electronic media (66.6$ to 17.6$).^
Confirming the importance of interpersonal channels at
the attitude formation stage, Lazarsfeld and Menzel found, through
the panel method ("whereby the same people are interviewed
repeatedly In the course of an election campaign so as to make
possible the detection of changes in their attitudes"),-* that
the media had "a rather small" effect in changing attitudes.
"People appeared to be much more influenced in their political
decisions by face-to-face contact with other peopIemembers of
their family, friends and neighbors, and people with whom they
workedthan by the mass media directly."-7
Thomas Robertson reports similar findings in marketing
products, stating that communication with other consumers can
reduce risk in the purchase decision. "In many ways personal
Influence may be superior to advertising, but it is difficult

+o utilize personal Influence channels, Robertson said.,s.. [fflp.pnlatri;
Interpersonal communication Is often only one phase of
a communication process. Being exposed to that channel at the
right time Is Important to effective communication.
Lazarsfeld and Menzel state, "Effective communication
demands the performance of several different functions, which
are often best served by different channels."' They relate that
rural sociologists note that different channels of Information
are associated with the various phases of a decision process.
Studies such as one by Copp, Sill, and Brown show that those
farmers who have been exposed to the "wrong channel at a given
phasefor example, who have been made aware of the new practice
by their neighbors rather than by farm magazineshave been more
likely to reject the innovation than those who have been reached
by the "right channel at the "right" time.**0
Interpersonal Channels and. Audience. Rel.ationshId
Despite the widespread research pointing to the value
of interpersonal channels at the persuasion stage, some research
points to specific circumstances under which people prefer inter-
personal channels as information sources. William Frederick,
Herbert S. Dordlck, and Frederick Horstmann state, "It is clear
that there are marked differences among communities or subgroups
in a large urban area not only in terms of general media habits
but also in the sources of Information on different topics."

Set In Los Angeles, this study asked respondents for the source
where thy might go to get information on specific topics. It
showed that there was a bias in Watts (almost totally black, low-
income) for interpersonal networks as information sources, in
Boyle Heights (predominantly Mexican-Amerlean) for institutions
or agencies, and Reseda (middle-income, predominantly white) for
print media and telephone.*
Because of this variation in preferences, it is important
for the promoter of referenda to know the source preferences of
his social system clients In order to have maximum impact,
In addition to using interpersonal channels of communica-
tion in a referendum campaign, mass media channels serve valuable
There is perhaps a fine line in differentiating between
the function of channels. "The closer one observes the workings
of mass media, the more it turns out that their effects depend
on a complex network of specialized personal and social influ-
ences," say Lazarsfeld and Menzel.^
Maas,, ,^,.lnf,Qrj[ia.t, I,on^S,ounce
Within this complex network, mass media is still identified
by researchers as a prime information channel. On this subject,
Dan Nimmo reported findings (I960) of the Survey Research Center
of the University of Michigan, which gathered data from a nation-
wide panel on voting habits. They asked, "Would you say you found

out about the campaign this year more by talking to other people
or from things like newspaper and television? Of those respond-
ing, more than eighty percent named the mass media over inter-
personal contacts.^
Political scientist Larry Sabato feels that of the mass
media sources, television has profound influence as an information
source on the political scene. The average American now watches
over three hours of television per day.... It comes as no
surprise, then, that television Is the most Important source of
campaign Information for voters.... Almost nine out of ten Amer-
icans say they follow television reports about campaigns, and
media Information (not just advertising but new programs as well)
is always at the top of a voters list of causal factors explaining
his ballot decision. In fact, about two-thirds of all Americans
consider television their primary source of news and about half
claim to rely on television whenever there are conflicting reports
in the various media.^
Credibility of Mass Media
The credibility of television, as mentioned by Sabato,
is supported by other researchers. Another testimonial to the
perceived credibility of television comes from Dan Nimmo. Not
only do Americans rely heavily on the televised media for news,
they are more likely to believe what they see than what they read
in newspapers. In recent surveys almost half of the persons inter-
viewed list television as the 'most believable' medium as opposed
to less than a third who list newspapers; generally only one in

ten Americans credits either radio or magazines with being most
believable.1 Moreover, these same studies reveal that Americans
think television is less biased, more colorful, and more complete
than newspapers or radio."65
' Nimmo stresses the importance of media specialists knowing
patterns of audience exposure in order to identify the campaign
audience. He reports that larger proportions of women than men
believe a television news report more than they do their news-
paper... as do less educated persons and younger persons. Older
respondents prefer newspaper accounts, as do men with at least
some college, residing in an urban area, professional or In a
reasonably high-status occupation. Farmers are inclined to place
high credibility on radio as an Information source. Audiences
of electronic media would include those members who do not partake
of print mediaoften the disinterested or alienated voteithe
ones many campaigns wish to reach.66
The media and/or campaign consultant must know the strength
of each media channel in order to advise his clients as to the
most effective channel for the specific purpose under the specific
conditions. At times, if expense is not a factor, the channel
may not make a large difference. AndreolI and Worchel found that
when the communicator agreed with his audience, neither the medium
nor the source differentially affected the audience's acceptance
of the message.6^ However, credibility of the source may make
a difference in selection of the media channel when the communi-
cator's message is discrepant with the audience's initial

position. Television is the most effective medium for a trust-
worthy source, but the least effective medium for an untrustworthy
communicator, say Andreoli and Worchel. They found that one
of the most trustworthy sources is a newscaster. One of their
studies concluded that a news reporter's Influence may be even
greater than that of elected officials serving in political
Media ,and Audience Relationship
The use of mass media may be highly affected by the nature
of the audience selected (targeted) to receive the message and
by prior attitudes of that audience.
Edward Jay Whetmore, author of believes 'tele-
vision commercials seem to work best In close elections or in
those where there is a large, undeclared vote.... Television
Is images, not issues.6^
Mass media may, Indeed, be more effective with the
marginal Iy-interested, undecided voter. Dan Mimmo found that
newspapers were most influential with that group. He says that
the principal proposition that emerges from studies of the campaign
effects of newspaper readership is that newspapers influence
(through stories, advertising, and editorials) the marginally-
interested voter. Under conditions of low Involvement, the voter
has few guidelines for what to believe or how to vote. In these
circumstances, readers frequently acquire knowledge of issues
and vote on the basis of newspaper content.^8
The general populaceor the margina11y-interested voter

Is not always the Intended recipient of campaign messages, how-
ever. Nimmo states, "Some aspects of campaign communication,
however, are directed not at the mass but at differentiated groups
with special concerns in hopes of getting endorsements, financial
support and votes. Direct mailings, door-to-door canvassing,
newspaper advertising and regional television are adaptable to
these more selective appeals."7*
Included In selective-appeal media Nimmo discusses display
media (bumper stickers, billboards, yard signs, placards, buttons
and other visual aids); "Few opinions of voters are changed by
such displays.... Their main purposes are building morale and
promoting name recognition or campaign identification."7^
Acknowledging televisions limitations as a persuasive
channel, Larry Sabato says, "Television may well be used more
effectively when it subtly reinforces our existing attitudes and
prejudices rather than when it attempts to change current ones
or create new ones."7^
Coming to the same conclusion, Dan Nimmo points to studies
on the effects of televised debates which suggest that few voters
change their minds as a result of watching debates but that a
significant number crystallize their decisions (that Is, the
undecided reach conclusions).7^
Jones and Saunders comment, also, on the Information and
attitude reinforcement qualities of mass media. In a before and
after assessment of mass media campaign effects, the researchers
state, "An evaluation of a 1972 multimedia pesticide information

campaign in Quincy, Illinois, concluded that the effort increased
audience levels of knowledge and strengthened attitudes toward
the pesticide labels to the safe use of pesticides."^
In summary, mass media holds tremendous potential,
according to several researchers, as a valuable medium for the
marginal Iy-interested voter and as a reinforces of attitudes the
viewer may already hold. Mass media can also be used to target
specific audiences for specific messages,
ReLationshlp of "News" and Tel ev iaion Commercials
To elaborate on the results of an Andreoli and Worchel
study reported earlier in this paper, which identified the news-
caster as one of the most trustworthy sources,^ this section
investigates the attempt of the campaign consultant to manipulate
the news and make the campaign seem newsworthy.
In the use of television in campaign commercials, Dan
Nfmmo says the overall ploy is contrived spontaneity. Next to
the spot commercial, the filmed documentary Is the most enter-
taining format for political appeals, he believes. Moreover,
manipulating the news is part of the goal of campaign specialists,
according to Nlmmo, "Campaign specialists thrive on the image
publicizing properties of television by generating material for
news reports (live, filmed, and taped), documentaries, interviews,
panel, and guest shots on celebrity shows. Brief exposure on
an evening news program is particularly prized.... The Baus and
Ross agency regards thirty seconds on television news as more
valuable than a half-hour paid ad on television.

Ultimately, however, it must be recognized that, on a
non-pa id basis, the media determines what will be publicized and
what won't. As Marshall McLuhan observed, the press has observed
that "news was not only to be reported but also gathered, and
indeed, to be made. What went into the press was news. The rest
was not news."^
The campaign consultant who can publicize the campaign
by making aspects of it newsworthy enough to be mentioned on other
television or radio formatsand especially the television newsor
who can make the paid commercials appear to be legitimate, spon-
taneous news advances the cause of the campaign.
Interrelationship of Media ChanneIs
It would be a mistake to think of each media channel
isolated from the others. In the real world, people are exposed
to more than one channel. A study by Lazarsfeld, et al. (1948)
based on voters in the 1940 presidential election confirmed that
more educated people attend to more of the mass media. "Thus,
most of those who learn of an event through one channel, for
example, a given magazine, are also likely to learn of it through
a second and third channel, for example, a second magazine and
a radio broadcast.
Whetmore, in discussing the role of media, relates a story
which emphasizes the complementary nature of all media to real
life events. "I was struck by this at a recent baseball game.
It was San Francisco at Cincinnati and we had choice seats just
over home plate. I noticed that the guy next to me had a portable

television set and a radio. Both were tuned to the came we were
watching. After a few beers I got up the courage to ask him Why
all the paraphernalia? He looked up from the television, turned
up the radio, and gazed at the field through his binoculars.
'Whadya mean? he said impatiently. 'Why, without this stuff
I wouldn't have any idea what was goin on.* Whetmore concluded
that many people must have "Realife" put through the mass media
channels before they really feel they have experienced "the total
Importance of Print Media. Radio and Telephone
Print media, as part of a campaign "realife" experience,
includes newspapers (publicity and advertising) and campaign liter-
ature. Nimmo points out that campaign literature can be adapted
for general appeal or targeted direct mail. "There is little
evidence available on the effectiveness of general audience and
clientele literature delivered by direct mail, special groups
or the campaign organization.... There is evidence that literature
promotes voter turnout. One study of a local charter revision
election revealed that of persons not contacted, only one-third
voted, of those contacted by mail, sixty percent voted; and three-
fourths of those contacted personally voted."*
Not only is print media valuable in promoting voter turn-
out, it also provides message cues to voters in some types of
elections. Dan Nimmo states, "Implications are that (newspaper)
editorial endorsements provide cues for voters, especially In
state and local referenda where party loyalties are less

relevant. Moreover, says Nlmmo, studies reveal that level of
education Is positively correlated with the use of printed media.0
Atkin, Galloway, and Nayman further elaborate, based on
a secondary analysis of national sample data, that "education
and social class are moderately associated with print media but
are not correlated to electronic media exposure."OJ
In addition to print media and television, radio Is a
valuable source of Information. With the average suburban commuter
spending ninety minutes of every working day isolated in his auto-
mobile, radio is his link with the world, says Nimmo. It provides
the capability of free-image building through interview and talk
shows and is less expensive than television. Transistors have
created a sizable audience of young adults still "hooked" on radio
as they grow out of rock-and-roll teens, Dan nimmo explains.
Yet another media tool, the telephone, has become a
standard method of achieving high voter turnouts in soft areas."
According to Nimmo it is also a tool for getting "personal"
recorded messages to voters and is valuable as a campaign research
survey tool
Also discussing the use of telephones in political
campaigns, Adams and Smith report that, in a field study environ-
ment, advocacy phone calls (calls which urge support for a specific
candidate or point of view on an issue) were made to turn out
the vote for a particular candidate in a special election. Results
of the study showed that the calls were effective in increasing
voter turnout but there was no evidence that the campaign phone

calls affected candidate preference of voters.
Ithiel de Sola Pool would indicate that this increased
turnout (hopefully, of a targeted group) could be a valuable tool.
"The main effect of a political campaign is to mobilize, not
convert, he states.
The campaign consultant needs to be aware of the potential
for using radio as a less expensive knowledge-creating medium
than television and to especially be aware of the mobilizing effect
in voter turnout of print media and of the telephone. The
consultant may wish, based on research results, to not take the
time to promote a specific candidate or issue during the telephone
call, however. Reminding a targeted voter audience to go to the
polls may be a more profitable use of time.
Media-as a Persuasive Too]
Although mass media are lauded especially for their role
In creating knowledge, not all research agrees that media channels
are Ineffective in the persuasion function. Jones and Saunders
report that In a 1976 multimedia campaign conducted by the American
Civil Liberties Union, with heaviest emphasis on television (71$)
and radio (25$), attitudes were changed about privacy and wire-
tapping. As shown by a before and after random samp 11ng of
different groups, heavy media watchers experienced the greatest
change in attitude toward the direction advocated.
Media as an opinion changer was effective In this instance,
according to Jones and Saunders, for two reasons: I) the privacy
Issue was easy to sell because it could be perceived as close

to the self-interest of many Individuals but has not been on the
public agenda so long that a great many people could have developed
firm positions. The former feature makes people more likely
to pay attention to the message, the latter aspect Increases the
chances that a new message could cause a change in formation,
saliency or attitude. 2) The assessment documented again tele-
vision's ability to reach a much wider audience than does any
other medium, according to the subjects' recall as to v/here thy
had seen or heard the message.*
Also exploring mass media and attitude change, Wall and
Boyd, in an experiment concerning channel variation and attitude
change on the subject of acceptance of women's liberation goals,
found that, after an analysis of variance, the written message
produced more change in attitude than either a live presentation
or a video-taped version of the live presentation.^ The results
also suggested that the use of video-tape was as effective as
a live presentation and could provide the experimenter more control
of message and delivery variables.^
The previous studies provide encouragement to campaign
consultants who may wish to use mass media to produce attitude
change. Under certain circumstances, the research has shown that
mass media can be a tool for persuasion.
Mass,Med ia Drawbacks
Whereas many campaigns have used media channels effect-
ively, Larry Sabato does not see the use of media as without draw-
backs. In a highly critical assessment, he states, "Mixing style

with substance and imagery with reality, media consultants have
developed a wide range of formats, strategies, techniques, and
gimmicks to both inform and deceive a television-addicted elec-
In the same attitudinal vein, Arthur Schlesinger, Jr. said,
Television and polling have bred a new profession of electronic
manipulators. Assembled in election management firms, the media
specialists...reduce campaigns to displays not of content but
of technique.9^
Continuing the ethical criticism of his own profession,
media consultant Robert Goodman contends, "Media consultants are
merely the stage managers and the creative writers of living
theater politics."9^
Imp Ucat i ons for Campa ion Consu I tants
The literature on channels of communication is diverse
and not without contradiction. The studies reported In this
section are largely without focus and difficult to categorize
because of their diversity. Taken together, however, they suggest
Important tips for campaign consultants charged with channel selec-
tion for referenda.
The results of much research would suggest that interper-
sonal channels should be used for persuasion and mass media for
information. However, the literature presents apparent contradic-
tions by showing that under some circumstances media can persuade
and that print information, for some voters, is the preferred
information source.

Campaign consultants should also time voter exposure to
channels of communication to the appropriate stage of decision-
making, should choose media and Interpersonal channels appropriate
to each population subgroup, and reinforce the effect of one
communication channel by using other channels to capture the "total
Drive-time radio use, telephone calls to increase voter
turnout, and reserving television exposure for the most trustworthy
communicators are among the numerous helpful hints presented in
this discussion of channel usage.
Useof a,. Persuas i ve Message
Although some of the channels of communication available
to promoters of referenda have been examined, the nature of the
message which will be sent over these channels also needs to be
considered. Dan Nimmo believes that the message of a referendum
campaign must provide "sympathetic perceptions" for the voter
in support or opposition of a specific voting choice.^ A campaign
consultant often devises the campaign message, and in doing so,
can define the purpose of the campaign.
MafififlqiPr PiiEacsaJA
The purpose of a campaign is reflected in its message...and
campaign experts disagree on the purpose of campaigns. Ithiel
de Sola Pool believes the purpose of a campaign is "to mobilize,
not convert.^ Kenneth Bailey, former Republican National
Committee field representative, often says of candidate races,

"The campaign that seeks to educate loses." Reflecting the same
philosophy concerning candidate races, Edward Whetmore says, "My
father, a long-time politician In southern California, has a
favorite saying, The worst thing a candidate can do is get bogged
down In the issues.'"00
Whetmore, approaching the issue of messages from a public
relations viewpoint, expresses the crux of referenda messages
as seen by many researchers. "(The message) must influence atti-
tudes, beliefs and behaviors. No matter how clever a message
campaign may be, if it fails to influence, it fails.... This
may mean reinforcing existing attitudes or creating new ones."*00
Though the philosophies of campaign purposes may differ,
each expert sees the message as persuasive communication. Dan
Nimmo discusses two philosophies of the purpose of persuasive
communication. "Since political campaigns consist largely of
persuasive communication, we can begin to explain their Impact,
large or small, by reviewing the findings of communication research
into the types of effects persuasive messages have on personal
attitudes.... The leading assumption is that a person's attitudes
direct his behavior by predisposing him to respond in certain
ways to a stlmuluspositively, negatively or with indifference.
The assumption that attitudes guide behavior carries with it the
Implication that to change behavior a leader must first change
the attitudes related to it."*0*
This philosophy, according to Nimmo, falls prey to limiting
conditions on persuasive effect on voting behavior. The personal

predfsposi+lons and personalities of voters, their social ties,
the credibility of sources, the form of the message, and the media
employed all limit the independent effectiveness of communication
in changing attitudes, and through attitudes, behavior, "In fact,"
Nimmo says, "research has revealed them to be so influential that
they often result in the majority of voters reaching decisions
before the campaign begins," Nimmo suggests that campaign communi-
cation might have a purpose different from that of changing
attitudesone that might be more effective in producing desired
votes.* ^
In describing this alternate theory, which he calls "a
theory of perceptual effects," Nimmo states, "The purpose of
persuasion is not to change the attitude of the committed, but
to shift the perceptions of voters with low involvement. If
persuasive communications are successful, the audience learns
the messages and modes of behavior acceptable to the persuader
without being converted,"*^
In relating the process of perceptual change to politics,
we offer two points," says Nimmo, "First, perceptual shifts need
not be limited to trivial or unimportant matters. Second, a theory
of perceptual change Is especially useful In discussing the conse-
quences of political campaigns because many Americans have only
low Involvement in their political community,.,. The repetition
of symbolic, almost nonsensical, appeals...changes the way they
look at politics and wins votes without really engaging fundamental
predisposition. For the lesser involved of our citizens a campaign

is effective to the degree that it gratifies inner needs rather
than converts basic beliefs. It rests in the theory of the
. , , ,,104,105
campaign as para-social pay." '
Television raises the possibility of para-social parti-
cipation by viewers in election campaigns, providing a sense of
participation with little demand for involvement and making the
para-social interaction pleasurable for its own sake. Nimmo said,
"The remarkable capacity of television to convey the images of
confI lets...and political moods, yet allow the viewer to perceive
in those images what he expects to see, to see for himself,
results in citizens who are electronic participants in political
events rather than mere targets for campaign messages."^
"Persuasion is a creative transaction in which the
persuadee takes active part in constructing meaningful responses
to the symbols in the persuaders appeal," says Nimmo in Political
CommunLcation and.Publie Oninion in America.107
Regardless of the guiding philosophy behind the creation
of the persuasive message, researchers point out basic elements
of persuasion that are crucial to diffusion campaigns.
Perce i ved Attr? butes are Important
A campaign consultant asked to devise a message for a
referendum campaign should be able to analyze and articulate why
the Innovation is to the benefit of the voters. Those most impoi
tant characteristics of the innovation are known as perceived
attributes and often provide the crux of the referendums message
to the voters

Rogers and Shoemaker discuss the need +o express the
perceived attributes of the innovation in any diffusion campaign.
the perceived attributes are crucial to the innovation's adoption.
Elements of perceived attributes include: relative advantage,
compatibility, complexity, observability, and trlalablIity.
The relative advantage is the degree to which an innovation
is perceived as being better than the idea ft supersedes.
The degree of relative advantage is often expressed in economic
profitability but the relative advantage dimension may be measured
in other ways. "The relative advantage of a new idea, as
perceived by members of a social system, is positively related
to its rate of adoption," say Rogers and Shoemaker.**^
"Compatibility is the degree to which an Innovation is
perceived as consistent with the existing values, past experiences
and needs of the receivers," according to Rogers and Shoemaker.***
Compatibility ensures greater security and less risk to the
receiver and makes the new idea more meaningful to him. An innova-
tion may be compatible I) with sociocultural values and beliefs,
2) with previously introduced ideas, or 3) with client needs for
"One indication," say Rogers and Shoemaker, "of the compat-
ibility of an Innovation Is the degree to which it meets a felt
need by the clients.... Change agents must have a high degree
of empathy and rapport with their clients in order to assess their
needs accurately. Such techniques as informal probing in inter-
personal contacts with individual clients, client advisory

committees to change agencies, and surveys are sometimes used
to determine needs for innovations.... The compatibility of a
new Idea, as perceived by members of a social system, is positively
related to its rate of adoption."**^
"Complexity, says Rogers and Shoemaker, "is the degree
to which an innovation is perceived as relatively difficult to
understand and use. The complexity of an innovation, as perceived
by members of a social system, Is negatively related to its rate
of adoption."**^
"TrialabiIity," according to Rogers and Shoemaker, "is
the degree to which an innovation may be experimented with on
a limited basis. New ideas that can be tried on the installment
plan will generally be adopted more rapidly than innovations that
are not devisable."
"Observability is the degree to which the results of an
innovation are visible to others. The observability of an innova-
tion, as perceived by members of a social system, is positively
related to its rate of adoption," Rogers and Shoemaker state.**
The campaign consultant promoting referenda must be alert
to these attributes and be ready to design a message which articu-
lates the benefit(s) of the innovation to the voter.
Advertising:, The Prompt ion of. Perce i ved i butes
With expression of the perceived attributes of the innova-
tion a prime goal of the referendum campaign, the expression of
these attributes is based in a form of communication known as
advertising. For purposes of a discussion about persuasion, it

is not necessary to differentiate between paid advertising and
unpaid advertising. Rather, techniques for promotion of the attri-
butes are at issue.
Atkin, et al. states, "The basic political advertising
has a basis In three key propositions from behavioral research;
exposure to more types of information is highly selective; level
of message availability is an important determinant of audience
reception patterns; and concept familiarity leads to positive
Expressed in a different manner, Joseph Plummer cites
three unique aspects of advertising as a form of communication
that needs to be recognized in a model of advertising communi-
cation. *
The first aspect, says Plummer, is the repetitive nature
of advertising over time. Second is the highly competitive nature
of the environment where advertising messages exist (commonly
called "clutter"). Third, because advertising has become an
accepted part of our popular culture, things can be shown or said
that outside of advertising would be rejected as absurd. As a
result of this cultural role and the "addictive" nature of tele-
vision as a communication tool, "people probably 'see* most tele-
vision advertising, but the degree of their Interest, involvement,
comprehension, retention and responses vary significantly." Given
this three-part perspective, Plummer proposes that advertising
communication be viewed as a process that incorporates both the
message and the receivers, with major emphasis on the viewers'

Plummer's perspective would be a theoretical concept that
would encourage campaign consultants to gear any message to the
receiver's perspective, keeping in mind the repetitive aspect
of advertising, which tends to create clutter and sometimes
"absurd" (but acceptable to the viewer) advertising.
SeIect i ve Attent ion of the Message. ReceIver
A common thread runs through the preceding philosophical
views of political advertising as a form of communication: before
the receiver will perceive attributes of an innovation gleaned
from political advertising, the advertiser must first get the
receiver's attention,
A campaign consultant hired to, among other things, cover
key messages about a proposed referendum, must know that his
message will have to compete with other advertising messages
(including, conceivably, other political advertisements) and
possibly the viewer's own preconceived attitudes to have an oppor-
tunity to influence that viewer.
Nimmo states, "Persuasion specialists realize that a
person's attention is selective;" selective attention means
that we only pay attention to messages that interest us, reinforce
what we be Iieve, and are the most agreeable of those competing
for our awareness. The selective attention barrier must be manipu-
lated to produce votes. One method, he says, is to package the
stimulus attractively by associating it with universally assuring
slogans"Progress is our most important product" (General Electric

slogan) or "Give your fair share" (the United Fund appeal).
Another method.Is to attempt to remove or control the elements
of the communication setting that compete with the stimulus.
Still another method, the most practical method according to Nimmo,
is repetition. Repeating a message increases the chances that
people will eventually pay attention and assures that the message
will reach the individual in a variety of contexts so that a
congenial setting will eventually prompt a desired response.-
Nimmo gives the example of a weary housewife who may ignore
a perfume commercial when badgered by her children during the
day, yet pay close attention later In the evening when they have
been tucked into bed. He suggests that, if heard frequently
enough, the message may modify attitudes as well as eliciting
positive response. He concludes, "Modern campaign techniques
approach the electorate as a differentiated mass which is condi-
tioned to behave in certain ways and has little Interest in
political affairs."*^
Two conceptsseIective listening and effect of repeti-
tionneed further elaboration. McCroskey and Combs discuss the
operation of selectivity in a persuasive setting. A number
of options are available to the receiver, they say. He may I)
derogate the source, 2) attend only to portions of the message,
3) redefine the meaning of the message consistent with his own
attitudes, or 4) ignore the message.
Nimmo also points to the personal defenses people have
in filtering out undesirable messages. "Because of these personal

interests people tend to expose themselves to communications
congenial to their existing attitudes and to avoid messages that
challenge those attitudes.
The campaign consultant, therefore, must view the voter's
selective filtering of messages as a major obstacle. The consul-
tant must overcome clutter and preconceived biases against the
referendum he is promoting for his messages to "sneak information
about the innovation's attributes to the voter.
Repeti t. ion- of- Message
The recommendation for repetition of message as a strategy
to offset selective filtering has support from numerous
researchers. Nimmo states, "Message repetition plays an Important
part in shifting perceptions. Since perceptual persuasion Is
aimed at people with low involvement in an issue, It is an audience
that pays little attention to the communication. But repetition
of a commercial or political spot helps overcome their lack of
awareness as well as their weakened perceptual defenses."
Atkin, et ai. state that enthusiasts for frequent and
repetitive spots find scientific basis for their beliefs in the
psychological experimentation of Zajonc, who showed that mere
I og
exposure leads to positive evaluation change. Also, they
contend, after assessing the literature, Berelson and Steiner
observed that "people tend to see and hear communications to the
degree to which they are available." Atkin concludes that campaign
messages must be prominently accessible if they are to reach
neutral voters who have little motivation to seek them out.

Therefore, the effective campaign strategy relies on spot adver-
tising techniques that can overcome the initial exposure barriers
created by low interest.
Tucker and Ware state that "mere exposure" with novel
or unfamiliar topics was persuasive. "Goodness of meaning
increased In...a linear fashion with an Increase in the number
of exposures. Our experiment, we feel, is evidence in support
of the contention that mere exposure to a 'meaningful1 stimulus
will in fact result in the enhancement of the subject's attitude
toward that stimulus object.
In the way of background, the idea of spot advertising,
as suggested by Atkin, was created by Prosser Reeves through what
he called television "electorate penetration" during the 1952
Eisenhower campaign. "Reeves argued that, unlike the customary
long-winded speeches that the audience could tune out, the spot
ad was brief, to the point, and difficult to avoid," says Atkin.
"His tentative research data showed that less than ten percent
of a sample of voters recalled content from Stevenson or MacArthur
speeches, but more than ninety percent could recall content from
Eisenhower spot announcements."*^*
Research provides support for the campaign consultant
who wishes to repetitively saturate the mass media with the refer-
endum's message(s). Repetition has shown to be an effective
offense against the voter's selective filtering, one of the biggest
obstacles the message may face in influencing the voter.

Effect, of Advertising on the Min Receiver
Keeping in mind the benefits of repetition, a campaign
consultant might wish to design his message (which presumably
expresses the priority benefits of the proposed referendum) for
the minimally-informed voter.
Atkin and Heald note that minimal Iy-informed voters appear
to be most strongly influenced by the agenda set (prioritization
of issues) in political advertising and that the frequency of
viewing is positively correlated with knowledge, and advertising
exposure is positively correlated with interest generated in the
campaign. They conclude that a we11-designed and we11-financed
political advertising campaign In the broadcast media can serve
to I) Increase the electorates level of knowledge, 2) elevate
emphasized issues and attributes higher on the voters agenda
of decisional criteria, 3) stimulate the electorate's interest
in the campaign, 4) produce a more positive effect toward a candi-
date as a person, and 5) intensify the polarization of evaluations
of the candidate.*^
Given the assumption that this strategy will work with
referenda campaigns as effectively as with candidate campaigns,
these benefits of targeting minimal Iy-informed voters and repeating
the messages sufficiently should enable the campaign consultant
to influence the voter with the issues the consultant feels the
voter should know.
Differences.Between Candidate.and Referenda Elections
Although there are many similarities between candidate

and referenda campaigns, the messages selected for a referendum
campaign will differ because there are several basic differences
between the nature of the two campaigns.
With the last two effects of political advertising relating
to candidates, it might be well to clarify some of those differ-
ences. Dan Nimmo says, "A popular referendum differs considerably
from the election of a candidate. In bond elections, for example,
no major personality commands public attention; instead the elec-
tion involves complex debates about the needs for roads, sewers
or schools and the technicalities of funding. The principal
problem is to sell the notion that passage of the issue would
be a public benefit when in fact some elements of the community
have more to gain or lose than others. The Baus and Ross agency
suggests that the general strategy for such elections should be
to; I) convince the voters that the bonds are self-liquidating
that whatever is built will eventually pay for itself, 2) win
the support of homeowners by demonstrating that the Improvements
proposed In the issue will increase property values, 3) create
the Impression that unless the Issue passes additional taxes may
have to be levied to pay for the projects, and 4) tastefully
frighten voters into believing that failure of the issue would
be disastrousthat failure to replace bad sewers, for example,
would spread disease. One final stratagem is to create the
illusion that the officials supporting the bond issue have noting
to hid, usually by associating them with a Citizens Committee
of prestigious backers of community causes.

Whereas Nimmo points out differences between candidate
and referenda campaigns which campaign consultants should consider
before designing a message, not every aspect Nimmo has mentioned
exclusively relates to referenda campaigns. Candidate campaigns
also often have a citizens committee leading a feeling of openness
and prestige to the candidate.
Use of Fear as a Mot ivat i ona I,. Appeal
When in the prior referenda strategy discussion Baus and
Ross suggested "tastefully frlghten(ing)" voters, the subject
of fear as a motivation appeal surfaced.
The Baus and Ross recommendation introduces questions
about persuasive appeals which will motivate the electorate in
the desired direction. Literature on persuasion suggests five
popular motivational appeals: hope, loyalty, love, hatred of
unworthy things, and fear. The use of fear appeals is perhaps
the most controversial and deserves elaboration.
Most research, beginning with a study by Jan is and
Feshback, supports the hypothesis that minimal fear appeals are
more successful in promoting the desired behavior in an audience
than are those containing strong fear appeals. They conclude
that the strong fear appeal In a persuasive communication tends
to produce various kinds of Interference that reduces effective-
Subsequent studies show specific situations in which strong
fear appeals are effective. Sternthal, Phillips, and Dholakla
report a study by Hewgill and Miller (1965) which showed that

in dealing with threats concerning physical consequences on noncom-
pliance, the greatest shift in attitude tov/ard the position
advocated occurred when a high-credibility source presented a
strong, threatening message.*^
Kraus, El-Assai, and DeFleur report findings in a field
study concerning fear threats of eye damage for direct viewing
of an eclipse which suggest that in some cases appeals containing
elements of strong fear may be used quite successfully. They
suggest that a possible reason for this result Is that the unfamil-
iarity of eye damage as a topic made it possible to accept recom-
mendations for avoiding damage without making changes in habitual
attitudes and behavior.*^
This line of reasoning is born out in studies concerning
strong and minimal fear threats advocating the adoption of auto-
mobile seat belt use, finding that those who had little experience
driving or riding (thus no strong habits) reacted greater in the
direction advocated with strong fear appeals. The researchers
conclude, "When a 'genuine* danger confronts a community, the
use of strong fear appeals may be quite successful in gaining
attention, in avoiding resistance through hostility and in
eliciting the desired response."
The research on the use of fear would indicate that
campaign consultants could recommend strong fear appeals when
a genuine danger to the public exists or when the topic is
unfamiliar to the voter. However, barring these specific situa-
tions, minimal fear appeals work better than strong fear appeals

as the recipient does not erect interference barriers, and the
campaign message has a better chance of not being filtered because
of the appeal.
Use of Ambiguous Messages
The use of fear appeals is only one contentious area in
the study of persuasion. Another area of controversy concerns
the desirability for ambiguous messages in persuasive communica-
tions, Dan NImmo contends, "It is important that the content
of a message.,,be sufficiently ambiguous that members of an
audience can project into it the precepts relevant to their own
cognitive needs."
NImmo says that people with strong cognitive needs ("for
clarity, meaning, organization, integration, and reasonableness
of experiences") who are quick to draw clear distinctions
("sharpeners") pay more attention to communications and may undergo
more attitude change than persons with equal needs who simplify
and Ignore distinctions ("levelers")."Interesting enough,"
says NImmo, "sharpeners seem particularly responsive to ambiguous
communication; this suggests that campaign appealsoften the
essence of ambiguitymay be particularly effective in meeting
the needs of individuals who strive to project their own meanings
into the message."
Chester Insko and others, in "Communication Discrepancy,
Stimulus Ambiguity and Influence," found that subjects conformed
more to a confederates suggestions when the stimulus was vague
rather than clear.

M. Lee Williams Investigated the use of deliberate vague-
ness" and determined, within the political arena of communication,
the deliberate use of vague messages is a well-established prac-
tice. George Gerbner and William Alston both present evidence
Indicating that there are contexts in which it is better to use
vague rather than specific terms. Being too specific could
alienate certain interest groups, limit the speaker's flexibility
in future negotiations, or provide the opposing forces with the
opportunity to gain a distinct advantage in devising a strategy
for encounter.
In the Williams' study, "deliberate vagueness" was used
to refer specifically to a situation in which a speaker purpose-
fully encodes a message by use of words and phrases whose meaning
boundaries are fuzzy (not clearly drawn). For example, words
such as "middle-aged," "most," "some" are vague because they lack
precision and allow room for a broad or narrow interpretation
among respondents. Professor Jon Winterton points out that such
vagueness also provides the respondents considerable potential
for rationalizing their decisions.
Williams says, "It should also be emphasized that 'delib-
erate vagueness' as used in this study applied only to issues
with which the audience disagrees and hence is conceived as a
rhetorical strategy to avoid premature negative responses from
the audiences. As such, it represents not an attitude change
strategy, but a stalling strategy."
According to the proponents of social judgement theory,

when a message falls within or near the receiver's latitude of
acceptance, it is assimilated and distorted so as to conform with
the receiver's beliefs, which thereby produce more agreement.
Conversely, a message falling within the receiver's latitude of
rejection is contrasted and viewed as being further removed from
the speaker's actual position than it really is, which theoret-
ically should produce less, or possibly even negative, attitude
change. On the basis of this theoretical perspective, therefore,
a disagreeable message which is clearly stated should fall in
the receivers latitude of rejection. However, a disagreeable
message which is vaguely encoded should create a wider range of
interpretations and more likely fall within the receiver's latitude
of noncommitment.
Williams' study examined the topic "What should be included
in a liberal college education." By changing to vague terminology
the terminology of item content found disagreeable to one random
group, Williams found that the social judgement theory held
"When faced with a vague message, the receiver has more
flexibility of interpretation," says Williams, "and appears to
assign meanings congruent with his/her own frame of reference
so as to achieve some degree of cognitive balance."
The results have several implications for the rhetorical
use of deliberate vagueness. Vagueness provides the speaker with
an effective means of avoiding premature exposure of his innermost
feeling, leaves the receiver with a neutral to moderately favorable

disposition, minimizes the chances of recalling the disagreeable
issues and avoids negative connotations which might jeopardize
future persuasive attempts. Williams hastens to clarify, however,
that "deliberate vagueness should not be perceived as a dominant
strategy for the persuader but rather as one which under center
circumstances can be useful in a long-term persuasive effort.
As an end in itself, the consistent use of vagueness is question-
able, since it avoids considering the specific Issues which should
be addressed. It is a stalling strategy so, at a later time,
we 11-conceived attitude change strategy could be used."
Campaign consultants can benefit by recognizing the
temporary advantages to using deliberate vagueness in order to
buy time until a more long-term persuasive strategy can be
developed. Even on a temporary basis, allowing the voter to read
into the message what he would prefer to see has advantages for
the consultant. Negative reactions can be sidestepped and the
consultant can have a better chance at influencing the voter in
the direction advocated.
Can-1 nformat ?on Camna lens be Per.suas ive?
Earlier in this discussion of designing campaign messages,
Ken Bailey, a former Republican National Committee field represen-
tative, said of candidate races, "The campaign that seeks to
educate loses." This philosophy, however, may not be entirely
true for campaigns dealing with concepts like "indifference" or,
conceivably, with policy decisions as in referenda.
In response to the question, "Can effective information

campaigns be designed?" Harold Mendelsohn concluded that the use
of research can guide media people to prepare effective information
campaigns if they:
1) Plan around the assumption that most of the public
will be only either mildly interested or not at all interested
in what is communicated.
2) Set middle range goals as specific objectives.
3) Set specific targets In terms of demographic and
psychological attributes, life-style and belief systems and mass
media habits. Here, determine the scope of prior indifference
and uncover its "roots" as well.
Examples of successful information campaigns that followed
the above guidelines include "The National Drivers Test" and "A
Snort History." The National Drivers Test was preceded by a
massive publicity campaign. All in all, some fifty million
official test answer forms were distributed via newspapers,
magazines, and petroleum dealers prior to air-time. The National
Drivers Test sought to overcome both prior public apathy and over-
confidence in its ability to operate motor vehicles correctly
by allowing drivers to measure their own abilities and then to
decide for themselves whether they were proficient. Local driver
improvement programs were available for those who wished to improve
their abilities.
To demonstrate the effect of alcohol and driving, "A Snort
History" was created. To overcome previous bias against blood
and warped-steel-on-cement programs, the quiet, six-minute film

combined live action and animated sequences. Blue-grass music
was used in the background, but not a word was said. The firm,
which won industry awards, was serious in its Jive sequences and
humorous in its animated ones.
Each film tackled "public indifference" in a different
way and showed, according to Mendelsohn, that effective information
campaigns could be created.
Campaign consultants for referenda may wish to consider
whether their subject areas lend themselves to information
campaigns. It must be noted that even in the examples given of
effective information campaigns that each still influenced the
viewer in order to be effective. It seems appropriate to caution
consultants about the necessity of influencing through the Informa-
tion campaign. Information alone Is not the end goal of a refer-
endum campaign. According to William Butcher of Butcher-Forde
consulting agency, "When it comes to political consulting, the
name of the game is winning on Election Day.
Imp.! i cat ions, for.Campa ion Consultants
The campaign consultant must consider numerous factors
in designing a message which will influence the voter. He must
be able to analyze and convey the attributes of the referenda:
relative advantage, compatibility, complexity, observability,
and trialablIIty. As the consultant advertises the message, he
must seek to overcome the selective filtering of the voter as
the referendum's message competes with untold others for atten-
tion. Viewing the message from the voter's perspective may also

help the consultant select an angle which will enable him to avoid
the voter's filtering because of prior attitudes.
Repetition of the message has been found to be an effective
strategy to counter selective filtering, lending support to
frequent spot advertising.
The Baus and Ross agency suggests general strategy for
referenda campaigns, including the possibility of a "tasteful"
fear appeal based on the potential consequences of the referenda's
defeat. Other research supports the effective use of strong or
minimal fear appeals under certain circumstances.
Ambiguous messages have also been found to be desirable
as a delaying tactic, especially through the use of vague rather
than specific terms.
Any successful campaign, including Information campaigns,
must also influence the voter. As Edward Whetmore said, "No matter
how clever a message campaign may be, if it fails to Influence,
it fails" a challenge to every campaign consultant.
Use, of. Measurement .Tools to Assess. Cl.lent Attitudes
and to, Devel og. Camoa iguv..Strategy
Prior to the development of a persuasive messageand
even prior to the development of a campaign strategythe campaign
consultant must be able to assess characteristics and attitudes
of the client population or his future campaign runs the risk
of not being appropriate for the voters involved. In literature
of planned change, "empathy" Is stressed as a key element in
persuasion. The consultant, In order to be able to "Put himself

in another person's shoes" and design a campaign strategy accord-
ingly, must understand that person's perception of reality.
In this context, Jon Winterton. Professor of Communication
at the University of Colorado, has said it is often more important
to understand what individuals "think" reality is than to dwell
on reality itself. He takes this position because he argues that
it is their "reality" which will guide their behavior.
Probing the client's mind to understand his perception
of reality has mushroomed through public opinion polling. Dan
Nimmo explains: "Contemporary campaigners turn to behavioral
scientists for valid data on public issues and voter attitudes.
By offering systematic inquiry instead of intuition, educated
estimates instead of haphazard guesses, research consultants play
an increasingly significant role in...elections."
Tvoes of Surveys
To assess client attitudes, five different kinds of surveys
are commonly used by campaign consultants today, according to
political scientist Larry Sabato, each with its own special
requirements and each done for specific purposes. In a major
campaign, he says, all five types may be employed one or more
times. ^
First in the series is the BENCHMARK POLL, also called
a "baseline" survey; which is constructed early in the pre-campaign
period (often a year or more ahead of the election date). The
goal is to assess the state's (or special district or other social
system's) mood and the public's general perception of the strengths

and weaknesses of the Issue a+ hand. The poll usually Is conducted
personally by Interviewers visiting the homes of respondents, and
the interview is quite lengthy. The sample size can be unusually
large. Lately, because of rising costs, many benchmark polls have
been conducted over the telephone. In any case, says Sabato, ''the
benchmark is a planning, not a predictive document.... As British
Prime Minister Harold Wilson once remarked, 'A week is an eternity
in politics,1 and a poll that attempts to predict an election
result months in advance Is a colossal waste of money."
"The second phase of polling is a refinement and extension
of the first," says Sabato. "Once a complete analysis of the
benchmark poll has been completed and areas of additional study
proposed, one or more FOLLOW-UP SURVEYS are conducted to probe
more deeply into topics of concern." Perhaps some actions
have been taken In response to the benchmark poll and the follow-
up sample can gauge the effectiveness of the efforts. Specific
segments of the population may be oversampled to give a clearer
indication, within a narrower margin of error, of their views.
Sabato says that the sample size for a follow-up survey is smaller
than for a benchmark and Is more frequently done on the phone.
Two or more are sometimes conducted six months or so prior to
the election. Closer to the election, "brushfire" surveys, short
polls with small samples focusing on issues that arise during
the campaign, are usually taken.
PANEL SURVEYS constitute a third major kind of political
polling. They involve the re interviewing of up to half the

respondents of a previous follow-up or benchmark poll after the
passage of several months to measure opinion shifts. This type
of poll, Sabato says, has built-in difficulties, even though it
is conducted by telephone, quite brief, and seeks well-defined
Information. Problems include the wider margin of error because
of the smaller sample size and the fact that previously interviewed
Individuals may not be as representative of the total population
as they once were (interviewing has been found to have a measurable
impact on political awareness and participation levels of respond-
ents). As a result it is necessary to set up a panel of new
respondents who are asked the same questions as the repeat
. 160
"The most critical kind of election polling comes last,
in the TRACKING PHASE," says Sabato. "As the election campaign
actually begins, shifts in sentiment among voters are traced
by calling fifty to one hundred or more individuals each night,
with a half a dozen very specific questions asked (usually about
the latest issues, changing perceptions, ...the effect of media
advertising, etc.). Not enough interviews are conducted each
evening to comprise a worthwhile sample, of course, but a system
of moving averages' is instituted. If one hundred individuals
are interviewed each night, then the last five consecutive nights'
interviews constitute the total sample, and with each new evening's
additional one hundred, the oldest nightly accumulation is
dropped."*6* Charting a trend in this way can be especially useful
to the campaign consultant as he attempts to make crucial decisions

in the closing days of the campaign.
"Tracking has been credited with assisting wins, as in
the November 7, 1982, front-page Washington Post article by
columnist David Broder entitled "TRACKIMG; Daily Polls Help GOP
Keep Senate Edge."
"A fifth category of tool used by pollsters," explains
Sabato, "is actually not a random sample poll at all. Called
a FOCUS GROUP, it is a small collection of individuals (a dozen
or less) selected nonrandom Iy to reflect age, sex, race, economic,
or life-style characteristics, usually representative of the larger
client group, brought together under the leadership of a trained
I &*?
discussion leader to talk generally about the campaign." As
the discussion unfolds, observers behind a two-way mirror make
extensive notes on the comments and connections among Issues,
images and advertisements. The result is similar to an open-ended
question polI.
"Focus groups are impossible to quantify," notes pollster
Patrick Caddell, "but their advantage Is that you get insights
that go beyond your numbers. You get a linear thinking process
that explores all sorts of unexpected dimensions to the
Weaknesses.of Pol-l.lna
In deciding on the types of surveys to be taken for a
given campaign, the campaign consultant must be aware of the
limitations and potential errors which can occur in polling.
Because of these I imitations and potential errors,

professional reactions to polling vary. The largest warning is
not to expect too much-or to rely too much~on polls. Richard
Wirthlin, head of Decision Making Information, a large polling
firm, "describes polling as the science of ABCalmost being
Sabato believes that "polling is oversold as a 'scientific'
profession and as a campaign tonic. Its weaknesses are at least
as apparent as its strengths, and there is a frightfully large
number of entry points for error in the polling process. Many
polls are just plain wrongpoorly done or poorly interpreted."
Several sources of potential error pointed out include prior
prejudices and question bias; the construction, analysis, and
interpretation of the polling instrument and results, including
the sampling and respondent screening process; and a host of
problems with interviewers and respondents.
Dan Nimmo suggests other sources of possible error, princi-
pally what to do about nonvoters (pollsters often incorporate
"filter questions" to determine If the individual has voted in
recent elections, is registered, or plans to vote) and the
undecided. Nimmo says that about five to ten percent are usually
undecided in their views on issues or perceptions of the campaign.
If a disproportionate number of the undecideds vote In the same
manner, rather than dividing as the decideds do, the poll will
not reflect this.*66
Another problem, says Nimmo, stems from the pollster's
recognition that any opinion expressed in an interview Is a

response both to the question asked (the stimulus object) and
to the situation (the interviewer, the campaign, and the entire
political environment), "Astute campaign managers interpret
opinion polls with an eye to the context within which the surveys
were taken," Nimmo says.
Moreover, the hypothetical of "if the election were held
today" (it is NOT being held today) is only one aspect of the
limitations on the ability to profile the mind of the electorate,
says Nimmo, Also, "there is no code of ethics despite periodic
I £ Q
calls for one by respectable firms,"
The least reliable source of representative attitudes
comes from the campaign volunteers, personal contacts, reaction
of the press, and warmth of reception at meetings. These can
be overly optimistic," Nimmo says.
This researcher is reminded of a favorite comment attri-
buted to a defeated candidate who, reflecting upon his decision
to enter the political race, lamented, "I thought I heard the
roar of the crowd, but it was really just the applause of fifty
of my best friends."
Use of Information from Pol-ls
Polling information is used by the campaign consultant
to develop campaign strategy, identify categories of voters,
identify issues, test commercials, and track changing public
The first step in developing campaign strategy is to
identify characteristics of voters. Polling information, plus

registration and voting records from county election offices or
the Secretary of States office, can aid in this task. Nimmo
uses this Information to:
1) Identify stable vote patterns. "One rule of election
campaigning is Go hunting where the ducks are, he says. To
find the ducks, he suggests analyzing voter turnout and analyzing
past election history.
2) Identify demographic correlates of voting. At this
point, group characteristics can be correlated with area voting
patterns. Demographic characteristics of residents learned from
polling can be supplemented by U.S. Bureau of the Census data,
which Is available either from the Government Printing Office
or from data service businesses such as Datamatics Corporation.
This information allows demographic data such as social ranking
(based on various occupational and educational levels), urbaniza-
tion ranking (based on types of housing, proportion of women in
the labor force, and fertility rates), and ethnic ratings (based
on the proportion of blacks, foreign-born and other minorities)
to be correlated with voting patterns. Whether analyzed by
computer or by volunteers, the end result is a systematic descrip-
tion of the constituencies.
3) Check residences and voter registration. (In Colorado
registered voter lists are available from computer services which
have secured tapes listing registered votersas issued by the
Secretary of State's office or by specific county election
offices.) Also door-to-door surveys can be done to determine

such information as age, registration, status, occupation.*7^
The use of such information can be endless. For instance,
with occupational information, "personalized letters can be
prepared by computer to reach special interest groups (doctors,
Insurance salesmen, recent migrants to Colorado from another state,
etc.). Polling information can identify issues or concerns
important to each subgroup so that any communication targeted
to members of the subgroup can have maximum effect. The campaign
personnel can, after correlation of demographic, voting pattern,
voter registration and attitudes/issues Information plan campaign
strategy and determine the best allocation of campaign resources.
Patrick Caddell, President Carter's pollster in 1980,
says, "We use survey data as an Information bank and that Informa-
tion has the basic function to drive the campaign." His polling
Is used not only to shape the themes of campaigns but also to
dictate the allocation of resources~"where people should go,
how much money should be spent where, which groups we target and
the kinds of money and messages we use," says Caddell.*7*
Pollster Robert Teeter states, "Polling data tend to cut
across a lot of areas and clearly contribute Important input to
scheduling, speech writing, advertising, and strategy." Coordi-
nating advertising decisions, target areas and speeches often
become the joint responsibility of the poll interpreters and
campaign strategists.
A Richard Wirthlin aide, pollster Richard Beal, cautions,
"Only data defining specific kind of voters are of much use in

planning. The Nation as a whole never votes, so knowing how it
thinks isn't all that important."
Dan Nimmo suggests a Typology of Campaign Targets," based
on time of voting decision and degree of voter interest. This
typology, coupled with specific polling and data information can
provide a guideline to overall strategy.
Polling and surveys will help to categorize voters into
the four segments and provide detailed descriptions of characteris-
tics and attitudes of voters within each category. Campaign
strategists, with this specific knowledge, can determine how to
campaign effectively to each group (and subgroup) within each
According to Nimmo, I) Early deciders who pay close atten-
tion to the election campaign must be reinforced in their decision;
2) Campaign strategy for Group Two is to mobilize favorables and
avoid antagonizing opposition In such a way that they are shaken
from their lethargy and stimulated to vote; 3) Voters who pay
close attention to the campaign yet delay their choices until
late in the period are prime targets for campaigners. Most are
probably undergoing an attitude conflict. The goal of the
Attentive Indifferent
Voter Voter
Reinforce attitudes 1 Mob i11ze or i mmob i11ze voters 2
Change attitudes and mobi1ize 3 Motivate turnout 4

campaigner is to resolve those conflicts favorably and mobilize
those voters to prevent them from withdrawing from any decision;
4) Unless captivated by a critical issue, these late deciders
are so Indifferent to politics that they will not vote at ail.
They are critical to campaigners who must (a) motivate them to
take a last minute interest in the election and (b) secure their
Use of Information in Pol itical, Advertisino
Planning political advertisements is a vital part of
strategy and use of information gained from polling. The "man-
in-the-street advertisement illustrates one use of polling and
data Information. "Man-in-the-street depicts filmed or recorded
interviews with people representative of subgroups within the
constituency to be reached. First used in politics by BBD&O agency
for Governor Thomas E. Deweys 1950 re-election campaign in New
York, it has been found to be unusually effective, especially
as a reinforcement tool. Generating the highest audience recall
of all the standard formats, according to Sabato, it is helpful
in communicating with targeted subgroups.
"If one's polling results indicate that black v/omen over
forty years of age are a swing group, for example, then such an
Individual, saying just the right thing, is Included In the commer-
cial, giving the targeted viewer someone with whom to identify,
says Sabato.
Polling can indicate issues which need to be addressed
and give clues as to style of presentation which might be most

effective within other popular advertisement formats, such as
"talking head," where an Individual speaks directly into the
camera, or "minidocumentaries" and "press conferences."
A specific example of polling leading to the use of a
particular advertising format occurred in the Carter-Reagan 1980
Presidential campaign. Polling showed that ninety percent of
the country outside California had a perception of Reagan as an
actor. Credibility had to be built into the campaign to combat
this image. The result: a documentary film called the Reagan
Record. which aired continually in the early campaign.
Not only can polling guide in the development of advertise-
ments, it can pre-test ads to determine their effectiveness.
The polling technique of focus groups was used extensively to
test Reagan commercials head-to-head with Carter commercials.
Although focus groups are usually used for qualitative testing,
not quantitative testing, the Reagan campaign attempted to quantify
some results through their chief pollster. By comparing the magni-
tude and shape of vote shift curves In all the cities used for
focus groups, the strategists know they were doing the right
things. y
Rafshoon, advertising coordinator for Carter, cautions
about using focus groups for negative spots, saying the "dirty
pool" effect appears upon first showing the commercial and that
the negative spots true effect come after repeated media exposures
and as the information becomes slowly internalized by the
v i ewer

In addition to using polling techniques to pre-test
advertisements, polls can also be used to determine effects of
strategies and advertisements upon the electorate and provide
quantifiable feedback to campaign consultants. Follow-up, panel,
and tracking polls are especially effective for this purpose.
The campaign consultant then can determine whether his advertising
strategy is sound or whether changes in the strategy need to be
Formats for Collecting Information
In addition to determining the types of polls to be used
and the use to which the information will be put, campaign consul-
tants must also determine the most cost-effective way of gathering
information through polling.
With the rising cost of pollingand the especially high
comparative cost of in-home polIing for benchmark surveysresearch
has been done to compare data collected in household Interviews
to that collected by telephone. Using the equated group method
in which respondents were selected by random samples within groups
and the results statistically weighted where necessary, Jordan,
Marcus, and Reeder examined the hypothesis that data collected
by telephone surveys is comparable to that collected in household
Interviews. Whereas content of data was comparable, specific
differences were noted. Income data was missing more In telephone
surveys (21$ compared to 12$ in household interviews). Telephone
respondents had a greater tendency to agree, to Omit responses,
and to use more extreme categories (VERY strongly agree, for

instance), this latter difference significant at well beyond the
.001 level. With absence of nonverbal cues, quality of data was
not as good as that obtained in household interviews. Inter-
viewers, for example, could not tell as well when to probe to
allow more complete answers. Dan Nimmo's comment may be a
reflection of the importance of the interviewer in such a situa-
tion. "An opinion survey is no more reliable than the inter-
| on
viewer," he says.
Mail surveys are common tools for gathering information
also, but nonresponse is a problem with this technique. Research
analyzing "Nonresponse in Mail Surveys: Access Failure or Respon-
dent Resistance" showed, through random sample polling, that three
out of five nonrespondents cited lack of time or interest as the
reason for not answering. "Low priority to participating" In
this type of survey was the key explanation of the nonresponse
I mo U cat Lons, for-Campaign Consultants
The campaign consultant who wishes to assess voter atti-
tudes before determining a campaign strategy needs to be aware
of five different types of polls, each with its own specific
purpose: benchmark poll, follow-up survey, panel survey, tracking
poll, and focus groups.
Such information can be used to develop campaign strategy,
identify categories of voters, identify Issues, develop commer-
cials, and track changing public opinion.
Polls can also be used to pre-test advertisements or

post-test their effectiveness. Consultants are warned to avoid
focus group pretesting of negative commercials, however, as the
'dirty pool effect Is to be expected at the first showing.
A comparison of telephone surveying to in-home interviews
shows that telephone surveys are comparable, but that certain
drawbacks need to be noted; in telephone surveys, more income
data will be missing. Moreover, respondents have a greater
tendency to agree and to use extreme categories.
Should consultants be tempted to use mail surveys, they
need to be aware that nonresponse is a big problem and can be
attributed to the respondents' lack of time or interest.
Despite the important role polling information pays, often
campaign consultants expect too much from polls. Also, problems
can occur with the polling instrument itself, the sampling process,
or the interviewers' implementation. The consultant must be aware
of these potential problems if he is to effectively use measurement
tools to assess client attitudes and to develop campaign strategy.
With the demand for referenda campaign consultants out-
stripping the supply, the education of new people interested in
this field is essential. Knowledge drawn from the subject areas
of diffusion of innovations, communication, and marketing will
be helpful to them and enable them to more effectively devise
campaign strategies. The problem of insufficient information
has been addressed in Chapter 2, which conveniently assembles

pertinent studies and synopses of literature for these new
This thesis also proposes that effective referenda
campaigns contain the key diffusion variables, about which informa-
tion has been assembled. These variables are use of credible
change agents, use of opinion leaders, use of a persuasive message,
use of appropriate channels of communication and use of measurement
tools to assess client attitudes and to develop campaign strategy.
The overview at the beginning of this chapter presented
a decision-making model which identifies each stage of the advance-
ment of an innovation requiring a collective decision. It enables
referenda consultants to make certain that each stage has received
appropriate attention and has been developed properly. The over-
view also stresses that the public's perception of being involved
in the process Is more important than the actual participation.
In addition to possessing knowledge about the process,
a consultant should know specific information about each of the
five variables discussed in this chapter. Generalizations and
highlights from each of the discussions follow.
First, consultants need to identify credible change
agents. Trustv/orthiness and Iikemindedness with the voters enhance
the change agent's credibility. His effect may differ depending
on the personality traits of the receivers. Credibility of the
source of Information (often the change agent) becomes less
Important the more time elapses between the delivery of the message
and Election Day, this phenomenon being known as the "sleeper

Change agents need to legitimize the innovation through
the selection and involvement of opinion leaders. Most charac-
teristics of leadership are widely acknowledged in the literature
though more than one method can be used to identify opinion
leaders. One of the most Important characteristics Is that the
leaders hold the same norms as the voters.
Change agents can speed the diffusion process by working
with opinion leader throughout the social structure. Moreover,
it is Important to involve opinion leaders because they may be
inclined to try to scuttle the proposal if they are ignored.
Opinion leadership may be limited to particular issues,
indicating a person could be a leader in one area and not in
another. Also, women do not seem to be as widely acknowledged
as leaders on political Issues as men.
In addition to being certain that opinion leaders legiti-
mize the referendum, campaign consultants must choose the channel
of communication the campaigns messages will travel.
Research supports the use of Interpersonal channels for
persuasion and mass media channels for information. It is
Important for the voter to be exposed to the right channel at
the right phase of the decision process. There are, however,
certain individuals who prefer interpersonal networks for informa
tion and for whom mass media can be persuasive.
BelievabiIity of the mass medium is important in the
campaign consultants selection of one channel over another.

Television v/as identified by almost half the respondents in a
study as the "most believable" and less biased. Specific groups
may attribute higher credibility to one source than other groups;
farmers, for instance, identify radio as highly credible.
Mass media use may be most effective with the marginally-
interested voters, who respond favorably to newspapers, especially.
Mass media channels can also be effective as a reinforcer of the
voters' existing attitudes.
Some mass media can be used for selective appeals to build
morale and name identification (i.e., billboards, yard signs,
etc.) and to turn out the vote. The telephone, as well as print
media, Is effective for mobilizing the vote, but the taking of
time for advocacy messages during this endeavor has not proven
For channels of communication to play an important part
in a campaign, a message needs to be developed to be sent to the
voter. The purpose of the message may be to change the voter's
attitudes or to merely have the voter learn the messages and modes
of behavior acceptable to the persuader without being converted.
Regardless of the guiding philosophy behind the creation
of the persuasive message, the campaign consultant must convey
to the voters why the Innovation will benefit them. These most
important characteristics of the Innovation are known as perceived
attributesrelative advantage, compatibility, complexity, observ-
ability, and trialabiIity.
Advertising expresses these attributes to the voter.

So much advertising bombards the voter (i.e., "clutter") that
repetition of a message is often used to transcend the selective
attention barrier. Spot advertising is often used for this
purpose. Such advertising is also more effective with minimally-
informed voters.
The Baus and Ross agency suggests basic strategy for refer-
enda campaigns which will guide the creation of messages. It
also introduces the possibility of using fear as a motivational
appeal. Whereas minima I-fear appeals are more commonly considered
effective, strong-fear appeals can be used safely when the topic
is unfamiliar to the voter or when a genuine danger actually
confronts a community.
Whereas fear can be a long-range strategy, deliberate
vagueness (i.e., ambiguity) can be an effective short-term
Developing both a short-term and a long-term campaign
strategy is a function performed by the campaign consultant.
He often needs to assess client attitudes through the use of
measurement tools (I.e., polling) in order to devise an effective
strategy. He must know the voters1 perceptions of reality as
those perceptions will guide their behavior, not reality itself.
Different types of polls, each serving a different purpose,
may be used: benchmark poll, follow-up survey, panel surveys,
tracking polls, and focus groups. The Information gathered from
such polling can be used to identify issues, test commercials,
track changing public opinion, and identify categories of voters.

Identifying categories of voters will, for instance, allow the
consultant to I) Identify stable vote patterns, 2) identify demo-
graphic correlation of voting, and 3) check residences and voter
registration. From this Information, personalized letters can
be targeted to reach special interest groups within the population,
campaign material on specific issues can be geared to those most
concerned about those issues, and campaign resources can be
properly allocated.
The development of effective mass media advertising can
be a direct result of polling information; moreover, such adver-
tisements can be pre-tested through focus groups, preventing
campaign dollars from being spent unwisely. Only negative" adver-
tising is difficult to pretest, as groups and individuals have
a "dirty pool" reaction upon the first showing.
To save campaign dollars, a campaign consultant can recom-
mend that a telephone survey be substituted for an In-home survey
so long as he anticipates that certain specific differences may
occur. Though easy to administer, mail surveys are subject to
nonresponse and thus present problems for the researcher.
Having reviewed the five campaign variables in depth,
this researcher will examine the importance of these variables
through a comparative analysis of two referenda on the same
subject, one successful, one unsuccessful. Should the analysis
indicate that the successful campaign showed the effective use
of each variable, consultants will know that their time is well-
spent in gaining knowledge about each of the variables.

Everett M. Rogers and F. Floyd Shoemaker, Communication
of. Innovations. 2nd ed. (New York: The Free Press, 1971), p. 275.
^ Rogers and Shoemaker, pp. 276-281.
^ Rogers and Shoemaker, p. 39.
^ Lewis Donohew and Edward R. Springer, "Information
Seeking versus Information Diffusion: Implications for the Change
Agent of an Alternate Paradigm," Community,Development Journal.
15, No. 3 (1980), 208-213.
^Rogers and Shoemaker, pp. 277-278.
Rogers and Shoemaker, pp. 278-280.
^ Rogers and Shoemaker, pp. 229-230.
Jon Winterton, Professor of Communication, UCD,
Lecture, September I, 1982.
^ John W. Gartrell and C. David Gartrell, "Status, Knowl-
edge, and Innovation," Rural .Sociology. 44, No. I (1979), 91.
Rogers and Shoemaker, p. 280.
** Rogers and Shoemaker, p. 281.
Rogers and Shoemaker, p. 286.
Rogers and Shoemaker, p. 286.
^ Rogers and Shoemaker, p. 291.
Rogers and Shoemaker, p. 245.
Mjchael Burgoon, Michael D. Miller, Marshall Cohen,
and Charles Montgomery, "An Empirical Test of a Model of Resistance
to Persuasion," Human Communications Research. 5. No. I (1978),
Herbert C. Kelman, "Process of Opinion Change," Public
Qpi.nlon..Quarterly, 25 (1961), 68.

*8 Virginia Andreoli and Stephen Worchel, "Effects of
Media, Communicator, and Message Position on Attitude Change,
Public Opinion Quarterly. 42, No. I (1978), 60.
^ James Jacard, "Toward Theories of Persuasion and Belief
Change, Journal of Personality and Social Psychology. 40 (1981),
^ Brian Sternthal, Lynn W. Phillips, and Ruby Dholakia,
"The Persuasive Effect of Source Credibility: A Situational
Analysis," Public Opinion Quarterly. 42, No. 3 (1978), 290.
Sternthal, Phillips, and Dholakia, pp. 288-289.
^ Rogers and Shoemaker, p. 241.
Rogers and Shoemaker, p. 212.
Sternthal, Phillips, and Dholakia, p. 307.
^ Sternthal, Phillips, and Dholakia, p. 292.
^ Sternthal, Phillips, and Dholakia, p. 296.
Gary I. Schulman and Chrysoula Worrall, "Salience
Patterns, Source Credibility, and the Sleeper Effect," Public
Opinion.,Quarterly. 34, No. 2 (1970), 371-382.
Virginia P. Richmond and James C. McCroskey, "Whose
Opinion Do You Trust?" Journal., of. Communication. 25, No. 3 (1975),
Richmond and McCroskey, p. 44.
^ Rogers and Shoemaker, p. 35.
Rogers and Shoemaker, p. 281.
Rogers and Shoemaker, p. 35.
^ Rogers and Shoemaker, p. 35.
^ Paul F. Lazarsfeld and Herbert Menzel, "Mass Media
and Personal Influence," in The Science of Human Communication,
ed. Wilbur Schramm (New York: Basic Books, 1963), p. 98.
^ Lazarsfeld and Menzel, p. 108.
^ Rogers and Shoemaker, p. 274.
^ Lazarsfeld and Menzel, p. 97.

Lazarsfeld and Menzel, p. 97.
Rogers and Shoemaker, p. 274.
John W. Kingdon, "Opinion Leaders in the Electorate,"
Public, Opinion Quarterly. 34, Mo. 2 (1970), 258.
Kingdon, p. 358.
^ Richmond and McCroskey, p. 46.
^ Richmond and McCroskey, p. 48.
^ Richmond and McCroskey, p. 48.
Lazarsfeld and Menzel, pp. 110III.
Rogers and Shoemaker, p. 280.
^ Rogers and Shoemaker, p. 281.
H Rogers and Shoemaker, p. 38.
H Rogers and Shoemaker, p. 14.
Rogers and Shoemaker, p. 212.
Rogers and Shoemaker, p. 24.
Rogers and Shoemaker, p. 207-208.
Rogers and Shoemaker, p. 382.
J Lazarsfeld and Menzel, p. 106.
Kenneth Starck, "Values and Information Source Prefer-
ences," Journal, of .Communication. 23, No. I (1973), 78-80.
Lazarsfeld and Menzel, p. 96.
Lazarsfeld and Menzel, p. 96.
Thomas S. Robertson, "The Process of Information and
the Diffusion of Innovation," Journal of.. Marketing. 31 (1967),
Lazarsfeld and Menzel, p. 105.
^ James H. Copp, Maurice Sill, and Emory J. Brown, "The
Function of Information in the Farm Practice Adoption Process,"
Rural Sociology. 23 (1958), 146-157.

Frederick Williams, Herbert S. Dordick, and Frederick
Hors+mann, "Where Citizens Go for Information," Journal of Communi-
cation. 27, No. I (1977), 95-99.
^ Lazarsfeld and Menzel, p. 95.
^ Dan Nimmo, The Political Persuaders; The Techniques
of. Modern Election Campaigns (New Jersey: Prentice-Hall, 1970),
pp. 113-114.
^ Larry J. Sabato, The Rise of Political Consultants:
New Wavs of Winning Elections (New York: Basic Books, 1981),
p. 117.
65 Nimmo, p. 115.
66 Nimmo, pp. 117-118.
7 Andreoli and Worchel, p. 60.
Andreoli and Worchel, p. 69.
69 Edward Jay Whetmore, Med Lamer t ea: Form. Content,
and Consequence of Mass Communication (Belmont, California: Wads
worth Publishing Co., 1979), pp. 268-269.
70 Nimmo, p. 133.
7* Nimmo, pp. 119-125.
7^ Nimmo, pp. 124-125.
7^ Sabato, p. 118.
7^ Nimmo, p. 161.
J E. Terrence Jones and Joan Saunders, "Persuading an
Urban Public: The St. Louis Privacy Campaign," Journalism Quar-
ter t-v. 54 (1977), 669.
Andreoli and Worchel, p. 69.
77 Nimmo, p. 133.
7 Whetmore, p. 31.
Lazarsfeld and Menzel, p. 96.
Whetmore, p. 180.
Nimmo, p. 128.

82 Nimmo, pp. 113-114.
83 Charles K. Atkin, John Galloway, and Oguz B. Nayman,
"News Media Exposure, Political Knowledge and Campaign Interest,
Journal ism Quarterly. 53 (1976), 236.
8^ Nimmo, p. 134.
83 Nimmo, p. 133.
88 William C. Adams and Dennis J. Smith, "Effects of
Telephone Canvassing on Turnout and Preferences: A Field Experi-
ment, PubIic Opinion Quarterly. 43, No. 3 (1979), 394.
87 Ithiel de Sola Pool, "The Effect of Communication
on Voting Behavior," in The Science.of Human Communication,
ed. Wilbur Schramm, (new York: Basic Books, 1963), p. 133.
00 Jones and Saunders, p. 673.
As shown by Lazarsfeld and Menzel, "People overwhelm-
ingly select for their attention statements of those opinions
with which they already agree. News and opinions about an issue
are paid most attention to by those who are most interested In
the issueand that usually means those whose minds are already
made up. A paradoxical result of this is the fact that those
who read most and hear most about an issue are the ones whose
opinions and intentions are least likely to change." Lazarsfeld
and Menzel, p. 96.
Jones and Saunders, p. 673.
1 Jones and Saunders, p. 673.
92 Victor D. Wall, Jr., and John A. Boyd, "Channel Varia-
tion and Attitude Change," JournaI.of Communication. 21, No. 4
(1971), 363.
93 Wall and Boyd, p. 367.
9^ Sabato, p. 111.
93 Sabato, p. 3.
98 Sabato, p. III.
97 Nimmo, p. 193.
98 Pool, p. 133.
Whetmore, p. 269.

Whetmore, p. 249,
Nimmo, p. 164.
Nimmo, p. 173.
Nimmo, p. 181.
Nimmo, p. 182.
The "ludenic" theory of mass communication centers
around a "play" concept. Nimmo, p. 182.
Nimmo, p. 186.
Dan Nimmo, Polttical Communication and Pub Iic Opinion
in America (Santa Monica, California: Goodyear Publishing Co.,
1978), p. 119.
Rogers and Shoemaker, p. 138.
Rogers and Shoemaker, p. 142.
Rogers and Shoemaker, p. 167.
Rogers and Shoemaker, p. 167.
Rogers and Shoemaker, p. 145.
Rogers and Shoemaker, p. 167.
Rogers and Shoemaker, p. 168.
Rogers and Shoemaker, p. 155.
Rogers and Shoemaker, p. 168.
Charles K. Atkin, Lawrence Bov/ern, Oguz B. Nayman
and Kenneth G. Sheinkopf, "Quality versus Quantity in Televised
Political Ads," Public Opinion Quarterly. 42, No. I (1978), 210.
I 18
Joseph T. Plummer, "A Theoretical View of Advertising
Communication," Journal, of Communication. 21, No. 4 (1971), 317.
Plummer, p. 317.
Nimmo, Political Persuaders, p. 32.