Citation
Successful determinants of pay for performance in an educational setting

Material Information

Title:
Successful determinants of pay for performance in an educational setting
Creator:
Emrich, Phyllis Ann
Publication Date:
Language:
English
Physical Description:
xiii, 216 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Teachers -- Salaries, etc ( lcsh )
Merit pay ( lcsh )
Performance awards ( lcsh )
Incentive awards ( lcsh )
Incentive awards ( fast )
Merit pay ( fast )
Performance awards ( fast )
Teachers -- Salaries, etc ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 208-216).
General Note:
School of Education and Human Development
Statement of Responsibility:
by Phyllis Ann Emrich.

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:
50727256 ( OCLC )
ocm50727256
Classification:
LD1190.E3 2002d .E57 ( lcc )

Full Text
SUCCESSFUL DETERMINANTS OF PAY FOR PERFORMANCE IN AN
EDUCATIONAL SETTING
by
Phyllis Ann Emrich
B.A., Oklahoma State University, 1980
M. Ed., University of Central Oklahoma, 1991
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Educational Leadership and Innovation


This thesis for the Doctor of Philosophy
degree by
Phyllis A. Emrich
has been approved
by
Kenneth L. Bettenhausen
tML
u



Emrich, Phyllis Ann (Ph.D., Educational Leadership and Innovation)
Successful Determinants of Pay-for-performance in an Educational Setting
Thesis directed by Professor Michael Martin
ABSTRACT
This purpose of this case study was to discover if some or all of eleven
determinants of success derived from literature were manifested in four pay-
for-performance plans, whether or not this information was useful in
understanding these plans, and how the plans were perceived for success by
the employees. This information was gathered to determine why some plans
were successful while others were not, and why some plans continued to
exist while others were terminated. Finally, the data were used to add to our
understanding of the workability of pay-for-performance plans in the areas of
education being studied.
The pay-for-performance plans that were implemented in the district
being studied were: (a) an attendance plan for transportation employees,
(b) a gainsharing plan for fleet maintenance employees, and (c) both an
individual incentive plan (merit pay) and a small group bonus plan for
administrators and managers.
The case study proceeded in two phases. In the first phase the
employees were interviewed about their perceptions of the pay plans, based
in


on questions aligned with successful determinants of pay-for-performance
programs. These interviews continued throughout a five-year span. In the
second phase, employees were given a questionnaire to complete, which was
also based on successful determinants of pay-for-performance programs.
Results indicate that certain determinants of success must be in place
in order to achieve a successful pay-for-performance plan in an educational
setting. Organizational trust and ongoing communication of the pay plan are
imperative factors in creating a successful pay plan.
Additionally, support of the plan by managers and adequate financial
support to accurately track pay-for-performance programs are required. The
success of the plans that continue to exist was attributed to the continual
evolution of the plan to meet the needs of the employees and of the
organization.
Implications for successful pay-for-performance programs in the areas
of education, including classified and administrative employees, are
presented.
This abstract accurately represents the content of the candidates thesis.
recommend its publication.
Signed
Michael Martin
IV


ACKNOWLEDGEMENT
I wish to express my gratitude to Dr. Michael Martin for his guidance
and support as graduate advisor and thesis committee chairman. His advice
helped steer my thinking in the right direction and his support was always
there and allowed me to place this challenge in perspective. I would also like
to thank Dr. Connie Fulmer, Dr. Rod Muth, and Dr. Michael Murphy for their
support and suggestions regarding this study. I also thank Dr. Kenneth
Bettenhausen for adding the perspective of the college of business in this
study.
I thank the many classified employees and administrators for their time
in completing questionnaires and their time in conversations around the topic
being studied. I would especially like to thank the managers who revisited this
topic with me on several occasions. I am also grateful for the time and effort
of Ms. Laura Rothenfeld, who proofread each rendition as it was written.
Most importantly, I would like to thank my husband Clark, and my
children, John Alan, age 7, and Kasey, age 4, for their love, support,
encouragement, and personal sacrifice that they experienced throughout this
time. Without their constant support, this endeavor would not have been
possible.


CONTENTS
Tables.........................................................xii
CHAPTER
1. INTRODUCTION...........................................1
Background of Pay-for-Performance...................1
Reasons for Pay-for-Performance Implementation ... 3
Alignment of Pay to the Organizations Strategic
Plan..........................................5
Private-Versus Public-Sector Organizations....6
Background of Study.................................8
Nature of Problem...................................9
Research-based Framework...........................12
Statement of Problem...............................14
Definition of Terms................................16
Methodology........................................17
Delimitations of Study.............................21
Summary............................................22
2. REVIEW OF THE LITERATURE..............................24
Pay for Performance: Definition and Background.....24
VI


Increased Use of Performance Pay...............25
Reasons for Increased Use of Performance Pay...26
Total Quality Management.......................27
Reasons for Pay-for-Performance Implementation.. 28
Concerns Regarding Performance Pay.............30
Individual Incentive Pay..............................31
Downside of Individual Incentive Pay...........33
Merit Pay.............................................34
Debate of Merit Pay Effectiveness..............37
Downside of Merit Pay..........................40
Group Incentive Pay................................. 42
Increase in Group Incentive Programs...........43
Group Size.....................................45
Drawbacks of Group Incentive Plans.............46
Gainsharing ..........................................47
Increase in Gainsharing Plans..................49
Reasons for Increased Use of Gainsharing Plans... 49
Factors of Successful Gainsharing Programs.....51
Results of Successful Gainsharing Programs.....52
Barriers to Successful Gainsharing Programs....53
VII


Determinants of Success for Pay for Performance......54
1. Employee Participation in the Implementation
Process.....................................55
2. Link Between Pay and Performance............57
3. Line of Sight...............................58
4. Organizational Trust........................ 58
5. Aligned to the Strategic Plan...............59
6. Decentralization............................60
7. Appraisal Process...........................61
8. Manager Flexibility.........................63
9. Adequate Financial Resources................63
10. Employee Perception of Fairness............64
11. Education and Communication of Plan........66
Diffusion: Public-Private Sector Differences..........66
Amorphous Jobs........................................68
Research is Needed....................................71
Summary...............................................73
3. METHODOLOGY............................................. 75
Research Questions....................................75
Research Methods......................................76
Design................................................81
viii


Design Validity................................82
Description of Study.................................85
Methods of Data Collection...........................87
Subjects.......................................88
Response Rate..................................92
Dependent Variable...................................93
Independent Variables................................93
Measurement of the Determinants of Success....94
Methods of Data Analysis.............................98
Coding.........................................98
Summary.............................................101
4. DESCRIPTION OF FINDINGS................................103
Research Questions 1 and 2..........................104
Determinant of Success 1......................105
Determinant of Success 2......................108
Determinant of Success 3......................112
Determinant of Success 4......................116
Determinant of Success 5......................123
Determinant of Success 6......................125
Determinant of Success 7......................125
ix


Determinant of Success 8.
129
Determinant of Success 9.....................131
Determinant of Success 10....................136
Determinant of Success 11....................141
Summary of Findings Research Question 1....144
Summary of Findings Research Question 2....145
Research Question 3................................147
Transportation...............................148
Fleet Maintenance............................153
Administrators and Managers..................159
Research Question 4................................166
Findings.....................................167
Summary of Findings................................169
Research Question 1......................... 169
Research Question 2..........................169
Research Question 3..........................170
Research Question 4..........................171
5. FINDINGS, CONCLUSIONS, AND IMPLICATIONS...............172
Purpose of Study.................................. 172
Findings and Conclusions...........................173
x


Research Questions 1 and 2: Findings........173
Research Questions 1 and 2: Conclusions.....177
Summary of Research Questions 1 and 2........177
Research Question 3: Findings................178
Summary of Research Question 3...............192
Research Question 4: Findings................192
Summary of Research Question 4...............194
Limitations of the Study.......................... 195
Implications for Further Research..................196
Summary of Chapter 5...............................198
APPENDIX
A. Interview Questions..................................200
B. Letter to Employees..................................201
C. Questionnaire for Administrators.....................202
BIBLIOGRAPHY...................................................208
XI


TABLES
Table
1 Employee Perceptions of Merit Pay................................41
2 Percentage of Productivity Sharing Firms Affirming Each
Non-monetary Benefit............................................53
3.1 Employee Participation in the Planning and Implementation
Stages Five Point Scale......................................106
3.2 Employee Participation in the Planning and Implementation
Stages Seven Point Scale.....................................106
4 Link Between Pay and Performance................................109
5 Line of Sight...................................................113
6 Organizational Trust........................................... 117
7 Program is Linked to the Strategic Plan.........................124
8 Appraisal Process...............................................127
9 Flexibility of Manager......................................... 130
10 Adequate Financial Resources....................................132
11 Fairness........................................................139
12 Education and Communication of Plan.............................142
13 Perception of Employees From Survey Are Determinants of
Success in Place?..............................................145
14 Total of Determinants of Success in Place.....................146
xii


15 Success and Existence of Plan
146
16 Number of Shares by Category..................................149
17 Transportation Employees Attendance Plan......................150
18 Gainsharing Payment.......................................... 155
19 Individual Incentive Payment................................. 161
13 Perception of Employees From Survey Are Determinants of
Success in Place?............................................ 175
20 Determinants of Success in Place............................. 176
xiii


CHAPTER 1
INTRODUCTION
The study examines the relationship between eleven research-based
determinants of success in pay-for-performance plans and to measure the
effectiveness of these plans, designed and implemented for administrators
and classified employees in a large suburban school district. This study also
examines four incentive programs that were designed and implemented
within three departments in this particular in the school district being studied.
If educational leaders are able to establish the extent to which the
determinants of success are implemented into the pay-for-performance plans,
then this study should likely predict the outcome of pay-for-performance
implementation. This study also will add to our understanding of the viability
of pay-for-performance in the areas of education being studied.
Background of Pay for Performance
Pay for performance has been a popular strategy used to increase
employee performance and the overall performance of organizations during
the last two decades. In fact, private industry and businesses have been
experimenting with several new pay practices such as variable pay (Lawler,
1


1990), group incentives (Cable & Welbourne, 1995), and skill-based pay
(Johnson, 1993; Schuster & Zingheim, 2000). These are now included with
gainsharing plans (Balkin, Gomez-Mejia, & Welbourne, 1995), which is a
reward system designed to share financial gains when improvement in
performance over a historical base is realized, and individual incentive
programs (Eskew, Fisher, & Heneman, 1996), both of which have been used
for more than two generations (Balkin, et al.).
In a review of private-sector pay practices, Milkovich and Wigdor
(1991) found that firms frequently used merit pay for their managerial and
professional staff (95%), and that variable pay plans were used less
frequently (16-40 percent, depending on the type of pay plan). However, they
stated that with increased worldwide competition, variable pay programs are
beginning to become more common.
When an organization makes a decision to implement a pay-for-
performance program, Lawler (1990) contends that a participative system
should be employed for successful plan implementation and must be used
for the full potential of the plan to be utilized (p. 114). Lawler (1990) and
Schuster and Zingheim (1992) argue that if individuals are involved with the
design of the new pay plan, they will be more committed to the plan and will
put more effort into ensuring its success. However, Gomez-Mejia and
Welbourne (1995) warned that such conclusions about employee participation
2


in the design phase of pay-for-performance plans are not tightly structured
and should be studied more thoroughly.
Reasons for Pav-for-Performance Implementation
Money can be a powerful motivator for employees, if used in the right
way. Organizations that align their compensation system with the values and
culture of their company can have a positive effect on the employees self-
esteem. According to Lawler (1987), productivity increases of 15-35% have
been reported when incentive pay systems are put into place.
In addition to using pay for performance as a motivator and a vehicle to
increase productivity, pay-for-performance programs are often propelled by
strong political will, and are used as a symbolic effort in holding employees
accountable for their productivity (Flannery, Hofrichter, & Platten, 1996; Perry
& Rainey, 1988). For instance, the implementation of pay-for-performance
programs can give stakeholders of the organization the impression that
employees are being held accountable for their work. In fact, Milkovich and
Wigdor (1991) assert that organizations often use performance-based pay
plans to make their management decisions appear more legitimate to both
employees and other organization stakeholders (p. 120). Further, Flannery
et. al (1996) contend that the perception of the public is that public employees
simply earn pay increases through longevity.
3


Additionally, Schuster and Zingheim (1992) argue that the largest
employer, the public sector, should have new pay practices legislated, instead
of being allowed to use current merit pay, in order to get pay focused on
results achieved and value received by the taxpayer (p.310). They believe
that merit pay in the public sector is just another term for an automatic pay
increase based on longevity in the organization. These arguments place
pressure on public organizations to adopt new compensation strategies that
hold employees accountable for their performance, rather than paying
employees increases simply for the sake of longevity.
Organizations implement new pay strategies to increase productivity,
retain and attract employees, increase employees skills and motivation, and
implement the organizations strategic plan. In both the private and public
sector, linking performance to financial incentives has been shown to
motivate employees to increase productivity and performance (Lawler, 1990;
Milkovich & Wigdor, 1991). This reward system can also provide incentives
that will attract new employees and entice current employees to remain in the
organization (Lawler, 1994; Milkovich & Wigdor, 1991). Studies also show
that when it comes to compensating employees, how you pay, that is, what
combination of incentive and base pay you pay, may be even more important
than the amount you pay (Milkovich & Milkovich, 1992, p. 53). Schuster and
Zingheim (1992) argue that only those organizations that do not compete for
4


business and customers are appropriate in their choice to use traditional pay
programs. With many of these reasons in mind, pay for performance
programs were developed and implemented in the school district being
studied.
Alignment of Pay to the Organizations Strategic Plan
Without the pressures to implement pay-for-performance programs
from stakeholders of organizations, new compensation systems are often
overlooked. In these days of organizational change and restructuring, many
compensation systems have been untouched and do not reflect the changing
values, vision, and climate of the organization itself. Pay needs to be aligned
to the strategic plan of the organization and cannot be seen as an expense
but as a long-term investment linked to the success of the company (Flannery
etal., 1996).
When aligning the compensation system of the organization to its
strategic plan the changing role of the employee must be considered. Todays
employees are asked to take on more risks and responsibility, acquire more
skills, and work as a team instead of in isolation (Heneman & Von Hippel,
1995; Sarnmer, 1996). Compensation needs to be modified to meet and
reward these needs.
5


A properly developed compensation plan, which includes proven
strategies for the successful implementation of performance pay programs
can result in a plan aligned with the organizations strategic plan, values, and
culture. As Flannery et al. (1996) state, An effectively designed, carefully
aligned compensation strategy certainly wont make all of an organizations
employees happy or satisfied ... [but it] will go a long way in improving
performance and results (p. 5).
Private- Versus Public-sector Organizations
An issue pertinent to this study is that private and public sector
organizations differ in their funding capabilities. For instance, if a private
sector firm sees an increase in productivity, then results usually include
increased profit that can be filtered down to the employees in the form of
performance pay; however, in a public organization, the organization may
increase productivity, such as gains in student achievement, yet realize no
gain in funding due to a variety of circumstances beyond their control.
Pay-for-performance programs are increasingly being reintroduced to
school districts throughout the United States. Although pay-for-performance
systems have been implemented at high rates in the private sector (95% of
companies reporting some type of pay-for-performance program), public-
6


sector organizations, such as school districts, attempt these programs on a
less frequent basis (Milkovich & Wigdor, 1991).
Often, private-sector pay-for-performance programs are transferred
directly into public-sector settings. Ingraham (1993a) described a case study
of the effectiveness of a pay-for-performance plan in a public transit authority
twelve years following the implementation of the program. Key decision-
makers about pay-for-performance plans assumed that private-sector pay
practices would have the same effect on the public sector. The study proved
that the diffusion model, taking practices from the private sector and
transferring these to the public sector, was not successful. According to
Ingraham (1993a), the findings demonstrated that many of the untested and
frequently unstated assumptions held by key decisions makers about pay-for-
performance were without foundation (p. 348). The Organization for
Economic and Community Development (as cited in Ingraham, 1993a)
maintains that in general, systematic analysis of incentive programs in public-
sector organizations is not pervasive.
Another issue concerning the diffusion of private-sector pay practices
to the public-sector organizations is that of manager flexibility in personnel
issues. In a survey by the National Research Council (as cited in Milkovich &
Wigdor, 1991), it was noted that personnel managers in private-sector
7


organizations had more flexibility than those in public-sector organizations in
promoting top performers and dismissing poor performers.
Background of Study
Organizations are challenging the practice of using performance pay
only for employees in management positions and are realizing that it can be a
system-wide program. They now are including their classified and support
staff in incentive programs, especially in organization-wide profit sharing and
gainsharing plans such as the gainsharing plan adopted by the school district
being studied (Kessler, 1994; Marsden & Richardson, 1994).
If worker productivity and performance is indeed increased through
new compensation programs in the school district, which was studied, the
implementation of these new plans could benefit the entire educational
process. For instance, if the gainsharing plan used in the maintenance
department results in saving district funds, these funds could then be applied
to lowering class size by hiring additional teachers. Further, if central-based
and building-based administrators set goals that help the school district in
their efforts to accomplish the districts strategic plan, the results could
increase student achievement.
The large suburban school district which was studied is not unlike
many school districts throughout the United States that find themselves faced
8


with the community seeking accountability in job performance. A catalyst for
this particular districts discussion of new compensation strategies was a
request by local chambers of commerce for implementation of some type of
system that would pay employees according to their performance. The
support of new compensation strategies continued following a finding of a
task force chosen from the community to study the finances of the school
district. To their astonishment, they found that teachers, administrators, and
classified personnel earned pay raises simply for accruing years of
experience.
The phenomenon described is not unique to this school district, and
school districts and other public bodies would benefit from a study that
evaluates these new plans to determine the effectiveness of the group and
individual incentive programs implemented in this school district. It would also
benefit other school districts by adding to their understanding of the
workability of pay-for-performance plans in the areas of education being
studied.
Nature of Problem
Considerable research examines the broad spectrum of new
compensation plans; however, this research often lacks studies regarding
unsuccessful pay plans. For instance, in a review of research on gainsharing,
9


Gomez-Mejia and Welboume (1995) discovered that the case study literature
was not balanced in terms of the number of successful and unsuccessful
programs reported. They are not sure if this is because most gainsharing
plans are successful or if most gainsharing plans that are failures are not
reported. They request that future studies focus on unsuccessful gainsharing
programs rather than on successful gainsharing programs.
Group-incentive programs are being implemented at an increasing
rate. In fact, many organizations are redirecting their pay programs toward
group incentives, rather than individual incentives. This is occurring because
employers are seeing substantial increases in productivity gains and
decreases in the number of grievances filed (Schuster, 1984) and an
increasing number of organizations are moving toward team-oriented work
design (Gomez Mejia & Welboume, 1995).
Most of the research conducted on group incentive plans has focused
on the outcomes of the programs instead of the actual process that must be
implemented to achieve success. More extensive research on the process of
implementing group incentive plans should be attempted (Cable &
Welboume, 1995; Weber, 1994). In addition, much of the research is
generated from a blatant pro- or anti-group incentive bias (Weber, 1994, p.
1.1), suggesting that more objective and unbiased evaluations of group
incentive plans should be completed. Further, most of the organizational
10


justice literature about group and individual incentive plans includes studies of
university students in lab settings, which has resulted in Webers request for
more extensive research to be completed in an authentic setting.
Cable & Welbourne (1995) suggest that studies are abundant in the
area of individual based-criteria as determinants of pay satisfaction [but]
studies on the effect of group-based incentives on pay satisfaction are fairly
scarce (p. 713). Although individual incentive programs seem to be
thoroughly researched, the bulk of the existing data concerns jobs that are
more concrete than the typical managerial position. Managerial performance
is not easily quantified because the nature of the job is amorphous and most
tasks are not directly observable (Milkovich & Wigdor, 1991).
Additional research needs to be completed in public-sector
organizations, such as school districts, that are attempting to respond to
public concerns about holding employees accountable for their performance.
A greater critical analysis that describes both successful and unsuccessful
attempts of pay-for-performance plans should be undertaken so that an
accurate reporting of these programs is completed. In addition, data should
be gathered and analyzed that would add to the current research regarding
determinants of success, found in the literature, of successful pay-for-
performance programs, such as the effects of employee participation during
the developmental stage of the new pay plan. The determinants of success of
11


pay-for-performance plans involved in the actual process of implementation,
and the effectiveness of group and individual incentives on productivity were
also be analyzed to add to the literature on pay-for-performance programs in
educational settings.
Research-based Framework
A combination of perspectives, those of Lawler (1990), Ingraham
(1993a, 1993b), and others, were used in this study. Ingraham has studied
the policy diffusion process of pay-for-performance programs from private-
sector organizations to public-sector organizations and identified factors that
must be included in public-sector organizations for successful pay-for-
performance programs. In addition, Lawler (1990) has defined determinants
of success of pay-for-performance programs as a result of his extensive
research. The perspectives of Ingraham (1993a, 1993b), Lawler (1990) and
others, such as Gomez-Mejia and Welboume (1995), Milkovich and Wigdor
(1991), and Schuster and Zingheim (1992) on pay for performance helped
explain the successes and failures of the pay-for-performance plans
implemented in the school district being studied. The following eleven
determinants of success were utilized as the framework for this study:
1. A participative system during the planning and implementation stage is
needed for most incentive systems to be successful (Balkin &
12


Bannister, 1993; GAO (General Accounting Office), 1986, 1991;
Jenkins & Lawler, 1981; Johnson, 1993; Kessler, 1994; Lawler, 1990;
Schuster & Zingheim, 1992).
2. The employees must perceive a link between their pay and their
performance in order for any pay change to impact their motivation
(Eskewetal., 1996; Lawler, 1987, 1990; Marsden & Richardson, 1994;
Milkovich & Wigdor, 1991).
3. The measures must be seen as credible and susceptible to influence
(line of sight) (Bennett, 1997; Lawler, 1990; Milkovich & Wigdor, 1991).
4. The climate of the organization should include organizational trust
between supervisors and employees, based on shared values and
goals (Ingraham, 1993a, 1993b; Jenkins & Lawler, 1981; Vest, 1988).
5. The pay system should be aligned with the organizations strategic
plan (Gomez-Mejia & Welbourne, 1988a; Hays, 1999; Ingraham,
1993b; Lawler, 1990).
6. Each department or unit should institute its own pay system (Bennett,
1997; Lawler, 1990; Milkovich & Milkovich, 1992; Milkovich & Wigdor,
1991).
7. The appraisal process should focus on observable behavior and be a.
quantitative measure that can be translated into the pay system
13


(Eskew, Fisher, & Heneman, 1996; Gaertner & Gaertner, 1984;
Greenberg, 1986; Lawler, 1987, 1990; Milkovich & Wigdor, 1991).
8. Managers should have the flexibility to recognize and adequately
reward or terminate employees (Betz, 1987; Ingraham, 1993a;
Milkovich & Wigdor, 1991).
9. Financial resources should be adequate to make incentives meaningful
and to fund the program on a continued basis (GAO 1991;
Greenberger, Miceli, & Near, 1991; Ingraham, 1993a; Montemayor,
1995).
10. The employees must perceive the plan as fair (Balkin et al., 1995;
Eskew etal., 1996; GAO, 1991; Gomez-Mejia & Welbourne1988b;
Milkovich & Newman, 1995; Milkovich & Wigdor, 1991; Welbourne,
1998).
11. Educating employees about the new compensation plan and
communicating with them during each phase of its development is
imperative (Flasch & Imberman, 1990; Imberman, 1995, 1996;
Schuster &Zingheim, 1992).
Statement of Problem
Generally, pay-for-performance research fits into two major categories:
pay for performance in the private sector and pay for performance in the
14


public sector. Because the organization being studied is a public school
district, it fits into the public-sector category.
This study investigates the effects of group and individual incentive
programs on group and individual performance in the public sector. It consists
of three case studies, each a separate department within a large suburban
school district, including transportation, fleet maintenance, administrators, and
managers. The following incentive programs were implemented:
(a) transportations individual incentive attendance plan, (b) fleet
maintenances gainsharing plan, (c) administrators and managers small
group incentive plan, and (d) administrators and managers individual
incentive merit pay plan. The effects of these programs, whether successful
or not, were reported.
This study establishes which determinants of success in pay-for-
performance programs were manifested in the four pay-for-performance plans
in the school district being studied. It also determines if the information was
useful or not in understanding these plans and if employees perceived these
plans to be successful. Although eleven determinants of success are found in
the literature and used regarding the pay-for-performance programs in this
study, the literature does not tell us if all of the determinants of success need
to be in place to produce successful pay-for-performance programs. This
study should help determine why some plans'were successful or not and why
15


some plans continued to exist while others were terminated. Finally, the
viability of the pay-for-performance plans in the areas of education being
studied was determined.
Definition of Terms
Gainsharina-a program that is based on historical data (Schuster & Zingheim,
1992) and is designed to reward all employees in a business unit for
organizational improvement (Bullock & Lawler, 1984).
Individual Incentive Plan-a program that ties the compensation to individual
performance based on quantitative performance measures (Schuster &
Zingheim, 1992).
Attendance Plan-a program that ties the employees attendance to a reward
system.
Merit Pav-a pay program that links annual wage and salary increases to
employee performance over the prior year (Risher, 1999).
Small Group Incentive Plan-a pay program designed to pay a uniform reward
based upon the group performance level of a predetermined goal. Each
member of the group shares the responsibility for the work output (Johnson,
1993) .
Measures of Productivitv-these were unique to each group. An attendance
plan was used for the transportation employees. The difference between
16


costs and revenues was used for fleet maintenance employees. School
improvement was used for administrators.
Mixed-The term used to indicate if varied responses resulted from interview
and questionnaire answers that were slanted towards affirming that the
determinants of success were in place.
Neutral-The term used to indicate if questionnaire responses scored near 3
on the Likert scale and similar verbal responses from the interview were the
result for the determinant of success.
Success of Plan-The plan must still exist and must meet the goals of the
specific employee group.
Methodology
This study consists of three case studies using qualitative and
quantitative research methods in which interviews, surveys, observations, and
information on the employee productivity were used for data collection.
Interviews were used as the primary means for evaluating the effectiveness of
the new compensation plans. The purpose of the survey was to complement
the information gathered through interviews. Observations of the classified
employee group meetings which focused on various compensation plans
were helpful in order to better understand the process of planning and
implementing the new compensation plans. Data were also collected on
17


changes in employee productivity in the transportation and fleet maintenance
departments as one method of determining the success of each plan.
The interviews provided qualitative data about employee perceptions.
Research has shown that collecting hard data in the area of new
compensation plans is very difficult. In fact, Eskew, Fox, and Heneman (1998)
provide a rationale for utilizing employee attitude surveys to measure the
impact of new compensation strategies. They suggest that it is difficult to find
objective measures of profitability, that external factors often influence a
change in productivity, and that it is challenging to be able to distinguish
between the effects of a new compensation system and that of other
organizational changes that are often occurring simultaneously.
Eskew et al. (1998) acknowledge that studies lacking hard data may
be misleading, but they contend that even studies that do include hard data
may show changes of productivity that result from cost cutting rather than
from the new program itself.
In addition to the participant interviews, the supervisors of each
department were interviewed about changes in productivity within their
department. They supplied hard data through annual reports that consisted of
the changes in productivity that took place. Also, observations of monthly
Classified School Employees Association (CSEA) committee meetings were
conducted. The CSEA pay-for-performance committee was made up of
18


representatives from each classified employee group in the district being
studied. The members discussed their particular departments progress in the
process of pay-for-performance implementation. This provided the opportunity
to observe first hand, the process of the design and implementation phase of
the pay-for-performance plans.
The interviews were used to gather information about the employees
perceptions of the new compensation plan. Data from the interviews were
used in conjunction with research-based survey questions from pay-for-
performance literature, in order to develop a questionnaire. The interview
questions were a brief, but more in-depth, version of the survey. Surveys
were disseminated to all employees participating in one of the four pay-for-
performance programs being studied. Survey questions were taken from Vest
(1988), Weber (1994), Lawler (1981), and from specific concerns concerning
the compensation plans that employees voiced during their interviews.
Data such as minutes and observations of meetings of the various
groups being studied, reports showing changes in productivity, and results of
the new pay plans, were compiled.
The samples of interview subjects were compiled in various ways. A
probability sample (Krathwohl, 1993) was used for fleet maintenance and
transportation. The supervisor of fleet maintenance and transportation was
asked to supply names of employees who were opponents, proponents,
19


and fence sitters of the pay-for-performance program. The classified
employee groups (fleet maintenance and transportation) involved many of
their employees in the process of developing their new compensation plan.
This process has been ongoing for over three years. Conversely, the central-
and building-based administrators have not included as many of their
employees in the process so purposive sampling (Krathwohl, 1993) was
used. For the purpose of this study, the key decision-makers were identified,
then two administrators from each of the following categories were randomly
chosen: (a) elementary principals, (b) middle school principals, (c) high school
assistant principals, and (d) high school principals.
The school district selected as a site for this study met two compelling
criteria: (a) departments were at the beginning stages of implementing new
compensation plans, and (b) these programs were site-based and were
offered to both classified and administrative employees.
The interviews were transcribed and coded. The codes were aligned
with the determinants of success by means of the program Non-numerical
Unstructured Data Indexing Searching and Theorizing (QSR NUD*IST). The
information gathered from the supervisors concerning any changes in
productivity was also incorporated into QSR NUD*IST for analysis.
The questionnaire was developed following the interviews so that
concerns that needed to be clarified could be integrated into the survey.
20


SPSS (1994) was utilized to analyze the survey data and to report the mean
responses to each question, which were grouped by determinant of success.
Delimitations of Study
The following statements define the limits of the study:
1. Quantitative data collection was used to complement the qualitative data,
and therefore advanced statistical analyses techniques beyond mean scores
are not used.
2. To enhance the qualitative nature of this study, various manager levels in
all four departments being studied were combined into one category.
3. The standards for any successful pay program in this study were rigorous.
For instance, the program must still be in existence in order to be deemed
successful. This study differs from many less comprehensive studies that are
completed shortly after implementation. In fact, Doherty, McAdams, and Nord
(1989) call for longitudinal studies of programs that begin early in the design
process and continue through the implementation and maintenance phase.
4. An intensive case study of one school district was purposively selected in
order to obtain enriched qualitative data over time as suggested by the
research (Doherty et al., 1989).
21


Summary
Although pay-for-performance programs are found in most private-
sector organizations, they are now beginning to appear more frequently in
public-sector organizations (Ingraham, 1992). Despite the growth in pay-for-
performance in the public sector, very little research has been completed.
This study should provide valuable insight into factors influencing increased
productivity in pay-for-performance programs in a public-sector organization,
such as this school district.
Chapter 2 reviews research literature relevant to the proposed study.
The determinants of success for pay-for-performance plans are covered as
well as research questions. Chapter 3 provides a detailed discussion of the
research methodology utilized to investigate the proposed relationships
between the determinants of success and the success of the new pay
programs. Topics addressed include the research design, methods of data
collection, subjects, methods of data analysis, description of the study, and
display of the data. Chapter 4 presents the data and findings. It includes the
findings regarding the eleven determinants of success and factors that
attributed to the termination or continued existence of each pay-for-
performance plan. Chapter 4 concludes with findings concerning the
workability of pay-for-performance in educational settings. Chapter 5 provides
a summary of the findings, conclusions drawn from the findings, and
22


limitations of the study, while addressing the implications of the study for
practice and future research.
23


CHAPTER 2
REVIEW OF THE LITERATURE
This chapter contains a literature review of pay-for-performance
programs, emphasizing the history of individual incentive pay, merit pay,
group incentive pay, and gainsharing programs. The chapter continues with
the description of eleven determinants of success for pay-for-performance
programs, and explanation of the diffusion process, a description of the
amorphous nature of management jobs, and concludes with research needed
and research methods. Research questions are included throughout the
review of literature.
Pay for Performance: Definition and Background
The label, pay-for-performance, encompasses a broad spectrum of
compensation systems including skill-based pay (Johnson, 1993), merit pay
(EskewS Heneman, 1996), variable pay (Lawler, 1990), incentive pay
(Eskew & Heneman), gainsharing (Balkin et al., 1996), and performance-
contingent pay (Milkovich & Wigdor, 1991). Further, some of these are
separated into categories of individual- or group-based pay. Pay-for-
performance can also be defined as broadly as any compensation system
24


that links pay and performance (Milkovich & Wigdor, 1991). For this paper,
pay-for-performance includes compensation systems identified as merit pay,
group incentive pav. gainsharing, and individual incentive pav. because these
are the pay programs being studied and implemented in the school district
being studied.
Increased Use of Performance Pav
Pay-for-performance is becoming more common worldwide. For
instance, Marsden and Richardson (1994) report that performance-related
pay has been on the increase in the UK, in both the public and private
sectors. Their research, along with that of Kessler (1994) also reveals that
although performance-related pay has long been a feature of managerial pay,
it is now being extended to a much wider range of occupations: skilled and
semi-skilled, clerical, and manufacturing. In 1995, almost 50% of all
companies in a survey by Towers Perrin had developed some sort of pay-for-
performance plan for non-executives, which is almost double of what was
reported four years earlier (Schine, Smith, & Zellner, 1996). Schine et al. also
reported that IBM increased workers pay an average of 8% in 1996, mostly
as a result of performance tied to their divisions performance. As early as
1991, 95% of U.S. private-sector firms were using merit plans, and many (16-
40%) firms were using variable pay plans (Milkovich.& Wigdor, 1991). This
25


phenomenon occurred as a result of peaked interest in pay-for-performance
plans following increased competition worldwide.
Reasons for Increased Use of Performance Pay
The increase in the use of pay-for-performance can be attributed to
several factors. One of the major goals of organizations is to increase the
performance level of employees in the workforce. In their literature search of
employee productivity for the Public Agenda Foundation, Immerwahr and
Yankelvich (1983) discovered that the productive capabilities of the U.S.
could be substantially improved if we could tap what the report calls
discretionary efforts. Employees responses to their survey resulted in
findings that: (1) 23% were not working to their potential, (2) 44% did not put
much more effort into their jobs over and above what was required to keep
the jobs, (3) 75% said they could be significantly more effective in their jobs,
(4) 73% said the quality and amount of effort they put into their jobs had very
little to do with how much they were paid, and (5) 73% believed the absence
of a clear connection between pay and performance was one of the main
reasons why their work efforts had deteriorated.
Pay-for-performance is used by managers to motivate individuals to
perform more effectively, to allow organizations to develop performance-
oriented cultures, and to increase the rate of attraction and retention of high
26


performers (Lawler, 1990; 1995; Milkovich & Wigdor, 1991). Meyer (1975)
argues that pay-for-performance would be a powerful motivator for most
people since the principle of this compensation plan is logically based. In
other words, If two people are hired to perform the same job and one
performs at a substantially higher level than the other, surely he should be
paid more for his superior contribution (p. 39). Bretz, Milkovich, and Read
(1992) add that managers also indicate that they utilize pay-for-performance
as a vehicle for improving future performance, making pay distribution
decisions, and communicating expectations regarding future performance
(p. 334). In addition, literature has also found that an added benefit of pay-for-
performance is the increased socialization and communication among
employees and supervisors (Milkovich and Wigdor, 1991; Milkovich &
Milkovich, 1992; Ingraham, 1993a).
Total Quality Management
Prior to the widespread use of pay-for-performance programs was the
introduction and use of total quality management (TQM); however, the TQM
movement is floundering (Imberman, 1996). In a review of the state of TQM,
Imberman reported on several studies. One of those studies was completed
by Ernst & Young. This study finds that many U.S. companies are not being
successful in their implementation of TQM and that many of the quality
27


management plans are too amorphous to generate better products and
services. Another Imberman study completed in 1977 reported that poor
quality costs in manufacturing averaged 5.8% of sales. This means that for
every $1,000,000 in sales, $58,000 was wasted on poor quality. In 1994, a
follow-up study found that poor quality costs in manufacturing averaged 6.1%,
or $61,000 out of each million. This evidence shows that the quality output
has not increased or improved over the years.
Since the TQM movement, pay-for-performance plans have become
very popular. Organizations often implement these programs for political
reasons (Kessler, 1994). One prominent reason is to enhance the legitimacy
of the organization as perceived by internal and external constituents
(Milkovich & Wigdor, 1991). Perry (1986) observes that from a political
perspective of supervisors, administrators, and even the public, that
implementing pay-for-performance asserts control over the bureaucracy.
Perry (as cited in Ingraham, 1993b) contends that pay-for-performance gives
the impression that the public or administration is in control and things are as
they should be (p. 199).
Reasons for Pav-for-Performance Implementation
Research studies show that effective compensation systems can
increase the motivation of employees up to 30% (Lawler, 1990); however,
28


much of this research was completed in lab settings using college students as
subjects (Milkovich & Wigdor, 1991). In addition, Schuster and Zingheim
(1992) report that organizations with moderately competitive base pay levels
and variable pay opportunities actually outperformed organizations with highly
competitive base pay levels, by a significant margin. Lawler (1990) states that
it is the anticipation of high pay that is motivational rather than the high pay
itself and that a high paying organization does not necessarily have a
motivated workforce. This suggests that organizations need to look at
incorporating some type of pay-for-performance program in order to achieve
increased performance among the employees. Gomez-Mejia, Welbourne, and
Wiseman (2000) warn, however, that these pay-for-performance plans must,
Infuse enough risk in the compensation package to create a common
fate between employees and the firm, without shifting so much risk to
them that they become overly cautious in their behavior and decision
making, resulting in even lower performance than would otherwise be
observed in the absence of such plans (p. 494).
Pay for performance has been used successfully in many
organizations. Imberman (1995) reported that many companies offer
incentives that result in reducing costs attributed to poor quality. These plans
are self-funded in that a portion of the savings earned from decreasing the
poor quality costs is passed on to the employees. Kaiser Aluminum increased
its productivity by 80% and decreased its poor quality costs by 70% within a
five-year period. General Tims, a plant in Illinois, paid out 20 million dollars to
29


its employees in bonuses after they boosted productivity instigated by a pay-
for-performance plan. The company pocketed an additional 10 million dollars
in profit besides the bonuses paid to employees. Jostens, a class ring
manufacturer, implemented pay-for-performance in which employees earned
bonuses for boosting their good-ring output. Prior to the plan, the average
employee produced sixteen good rings in ten calendar days beginning with
the work order and continuing on to production and shipment. This amount
increased within one year to 36 rings being produced by each worker in the
same time frame (Imberman, 1995).
Concerns Regarding Performance Pay
Although the above testimonials are success stories of pay-for-
performance programs, there are some problems that must be taken into
account when implementing such programs. Organizations, private and
public, have encountered problems with supervisors overrating employees on
their evaluations, inadequate funds being available for rewards, failure to link
appraisal to reward (Immerwahr & Yankelovich, 1983), and managerial
reluctance to deal with poor performers (Milkovich & Wigdor, 1991; Ingraham,
1993a). In addition to these downfalls of pay-for-performance plans, individual
incentives may create other concerns, such as increasing competition among
employees whose success depends on working interdependent^ with their
30


colleagues. There are many examples of the negative effects of individual
incentive plans that have been documented in the area of retail sales, for
instance, employees not restocking merchandise because it is not part of the
incentive program. Also, retail clerks have even been known to hide the most
sought after merchandise from each other. As an example, Sears auto
centers were accused of selling unnecessary auto repairs to customers so
that they would earn bonuses from their pay-for-performance plan (Milkovich
&Milkovich, 1992).
Pay-for-performance programs are not for every organization.
According to Lawler (1981), performance pay should not be used if trust level
in the organization is low, high performers are not able to be compensated
substantially more than low performers, and performance can not be validly
and objectively measured.
Individual Incentive Pay
Excluding a merit pay program, which is discussed later in this chapter,
individual incentive pay is used in one department in this study. This
department implemented an individual incentive plan that based
compensation on attendance.
Typically, individual incentive programs tie the compensation to
individual performance based on quantitative performance measures. It is
31


believed that employees see this type of pay as more doable because they
can directly affect their pay through their individual performance. This is in
contrast with group incentive plans or merit plans, which tend to tie pay to the
performance of the group or organizational goals and use more qualitative
type performance measures (Meyer, 1975).
Meyer (1975) points out that individual incentive plans are usually
more successful in environments that include: 1) simple, structured jobs in
which employees are relatively autonomous, 2) work settings in which
employees trust management to set fair performance goals, and 3) a stable
economic environment (p. 153). Individual incentives are offered to motivate
employees to perform at higher levels, to retain the best employees, and to
encourage the lower performing employees to leave the organization or
increase their productivity (Lawler, 1990; Weiss, 1987). Although the use of
individual incentives seems to be widespread, it appears that they are
declining in use (GAO, 1991).
In a study by Farr (as cited in Dickinson, Honeywell, & Poling, 1997), it
was shown that individual incentive pay was an effective means of
compensation. They compared the effectiveness of hourly pay, individual
incentive pay, and group incentive pay on the performance level of small
groups in a lab setting. In the individual group, each person received
compensation based on his individual work, whereas, in the equal group-pay,
32


each member received the same amount of pay as the others, based on the
performance of the group. Finally, the individuals in the differentiated group
received differentiated pay based on their performance and based on the
group performance. Individual and group incentives resulted in significantly
higher performance levels than in hourly paid compensation.
Downside of Individual Incentive Pav
The downside of individual incentive programs is that it can result in
decreasing teamwork and increasing competition among coworkers. A bonus
program that pays only to the highest performer places employees in
competition with one another. In addition, it is often difficult to determine what
is contributed by each worker (Hansen, 1997; Schuster & Zingheim, 1992).
Lawler (1990) adds that individual incentive plans can lead employees to
neglect aspects of their jobs that are not specified in the incentive plan, report
invalid data in order to receive the incentive, and not work toward the group
norms of the organization. In most complex organizations, success depends
on interdependency, cooperation, and teamwork (Schuster & Zingheim,
1992). If the organization needs people to work interdependently in order to
be successful, then using individual incentives may create an internally
competitive climate that does not lead to group cooperation to insure the
success of the organization (Schuster & Zingheim, 1992; Meyer, 1975).
33


In addition, it has been shown that using individual incentives with jobs
that are complex, interdependent, and have amorphous goals can result in
employees ignoring important aspects of their job in order to meet the
appraisal goals (Meyer, 1975).
There are numerous studies suggesting that individual incentive pay
does not increase productivity and that it may be harmful to the work
environment, especially if the jobs require interdependence and teamwork for
success. A study by Konovsky and Podsakoff (as cited in Perry, 1986) found
that individual incentives had no impact on performance in an interdependent
lab task and that the contingent-pay actually decreased employee
performance where interdependent work was required. Balkin and Gomez-
Mejia (1989) completed a study based on a sample of 175 scientists and
engineers. The study of the Research and Development (R & D) employees
shows that individual-based rewards (either merit pay or individual bonuses)
are perceived as less effective than group-based incentives. The pay
effectiveness measures included individual performance, propensity to leave
the organization, pay satisfaction, and project performance.
Merit Pay
Merit pay is one of the two types of pay plans used by the
administrators in this study. Administrators earn one of three of the following
34


merit pay increases according to their performance appraisal: (a) employees
meeting the standard, 1 % increase; (b) employees meeting or exceeding in at
least one area, 2% increase; and (c) employees who exceed the standards in
all areas, an additional .5% increase. This last category is a lump sum
payment, not added into the base pay amount.
The term merit pay generally refers to a pay program that, links
annual wage and salary increases to employee performance over the prior
year (Risher, 1999). The payments are usually added on to the base pay and
are awarded to employees on an individual basis, depending on the level of
performance or their performance rating (Milkovich & Wigdor, 1991;
Montemayor, 1995). Surveys show that merit pay is the most commonly used
method of determining pay for exempt employees; however, the use of merit
pay is declining in favor of other types of pay-for-performance (Heneman,
1992). For instance more than half of the companies in the ACA survey said
that they now use variable pay plans (Sammer, 1996).
In their review of research on merit pay, Milkovich and Milkovich (1992)
found that there was little or no research on the effects of employee behavior
and organization performance even though merit pay is the most common
(95%) pay strategy used by organizations. One of the reasons stated for the
popularity of merit pay is that it recognizes the individual and shows him that
he, as an individual, matters and is compensated on the difference he makes
35


to the organization. It is also believed that merit pay can motivate an
employees job performance and can increase organizational effectiveness
(Lawler, 1987). Lawler states that productivity increases of 15%-35% have
occurred when incentive systems such as merit pay are used. If an
organization is to use merit pay successfully it needs to ensure that there are
available funds for this pay strategy. Individuals are more likely to be satisfied
with their merit pay if they perceive it to be large, in comparison to the
average merit increase, and the larger the increase, the more satisfied the
employee will be (Vest, 1988). In a study completed by Montemayor, (1995)
he found that pay-for-performance norms become more egalitarian when the
budget is limited and become more meritocratic when the budget is ample.
The literature on fit suggests that merit pay works well with steady-
state organizations, while group incentives are better matched with those
organizations that are innovative and entrepreneurial (Milkovich & Wigdor,
1991).
Performance evaluation plays a large role in the success of merit pay.
Pearce and Perry (1983) found that the acceptance of merit pay depends on
the employee perceiving that the appraisals are accurate and reflect true
performance. If this does not occur, the employee will then seek only high
ratings instead of good performance. In his study of a public transit authority
merit pay program, Vest (1988) found that the perceived performance
36


appraisal accuracy is a factor for employees who receive performance
appraisals that are lower than expected, whereas this is not a factor to those
employees who receive their expected or higher than expected performance
appraisal. He further states,
For the group of individuals who receive a lower than expected
performance appraisal, the more accurate an individual perceives their
performance appraisal to be, the more likely the individual is to believe
that pay is tied to performance (p. 146).
In addition, the perception of an adequate link between pay and
performance must be apparent for the merit pay system to appear legitimate
(Montemayor, 1995). Unfortunately, researchers have found that employees
involved in merit pay systems only perceive a moderate link between their
pay and performance (Heneman, 1992; Montemayor, 1996).
Debate of Merit Pay Effectiveness
Although merit pay is the most common pay strategy used as a pay-
for-performance program, there is continuing debate whether or not merit pay
is a successful pay strategy. A decrease in the effectiveness of merit pay was
reported by Eskew et al. (1996). They found in a survey of senior
compensation professionals in 72 organizations conducted to assess the
effectiveness of merit pay in achieving organizational objectives, that merit
pay was marginally successful. Eskew et al. compare this to the same type of
37


study completed ten years earlier, which resulted in a merit pay rating of
moderately successful, leading to the conclusion that merit pay is decreasing
in effectiveness. Pearce, Perry, and Stevenson (1985) find that there was no
relationship between merit pay and organizational performance in the
organization that they studied. They conclude that merit pay may be an
inappropriate method of improving organizational performance.
Another factor that some researchers believe may contribute to the
failure of merit pay is the managers ability to reward and evaluate the
employees. The manager must be able to give ample awards to the
employees and accurately evaluate the employee. In their longitudinal study
of the Civil Service Reform Act of 1978, Pearce and Perry, (1983) find that,
most behavioral scientists believe in the merit pay principle and attribute the
frequent failures of merit pay to inadequacies in its implementation (p. 316).
In their study, Pearce and Perry utilized the Federal Employee Survey semi-
structured interviews with a stratified random sample of employees,
managers, and union representatives, and archival data. This collection of
data allowed them to use mean response, t-test probability, and correlations
in determining their conclusions. One of the conclusions is that failures of
merit pay systems are often attributed to the fact that managers do not make
large enough salary discriminations between subordinates. This is reiterated
by Meyer (1975), who adds that managers should be more thoroughly trained
38


in the merit pay process. In addition to understanding the importance of the
discrepancy of pay, they need to be trained in accurate and objective
performance evaluation. Furthermore, Meyer suggests that another problem
is that the employees realize they are dependent on the supervisor for their
reward.
Pearce and Perry, (1983) identified several problems concerning
diffusion, the transfer of the compensation system from the private sector to
the public sector. They speculate that in addition to the problems that occur in
the private sector with merit pay, the public sector also has to face the
problems of an ambiguous performance environment, tight budgetary
restraints, freedom of information about the individuals salaries, diffuse
authority for implementation, a major managerial succession, and significant
changes in organizational goals (p. 324).
Another factor that is difficult to overcome is employee dissatisfaction
with pay. Few people are happy with the way merit pay plans are structured.
In Meyers (1975) search of pay literature, he found that surveys of
employees covered by merit pay have consistently shown high levels of
dissatisfaction with pay (p. 40).
39


Downside of Merit Pay
According to Marsden and Richardson (1994), merit pay may actually
be demotivating in nature. In their case study they interpreted data collected
from a questionnaire whereby 2000 employees and managers expressed
opinions about merit pay. They concluded that, Although the scheme was
thought by staff to have a number of virtues, it was very unlikely to raise
employee motivation appreciably .. . (p. 243). They also suggest that if
motivation were not increased, certainly employee performance would not
have increased. Their reasoning for using management and employee
attitudes and opinions from a questionnaire was because merit pay is based
on the employer performance evaluation and this specific one included no
less than thirteen elements, some which were not applicable to all employees,
it would be difficult to monitor behavior on these elements. The basis of the
questionnaire was to measure any change in motivation by asking the
employees questions about themselves and their colleagues. The survey
recorded the employees opinions and attitudes toward the merit pay plan two
years following the implementation of the plan, which gave workers time to
adjust to the plan. It becomes apparent from the data collected that few
employees felt that merit pay had provided them with incentive, or a
motivation, to change their behavior at work at all significantly (p. 250-51).
40


For instance, they were asked if they agreed or disagreed that performance
pay (PP) had led them to:
Table 1
Employee Perceptions of Merit Pay
Agree Disagree
Employees Improve quality of work .12 .... 80
Work harder .. 9 .... 78
Be more effective in dealing with the public . 9 .. .. 68
From Managers PP has helped to increase the quality of work of many staff... .10. ... . ... 82
PP has made many staff more committed to their work . . .12 . . . . 79
All Staff PP has helped to undermine morale .55 . . . ... 25
PP has caused jealousies between staff ..62 . . . ... 21
(p. 251)
Finally, the last and perhaps most serious factor in determining the
success of merit pay, is in regard to employee self-esteem. In Meyers (1975)
review of pay-for-performance literature, he finds that merit pay can actually
threaten the employees self-esteem because they will not receive rewards
that they feel their performance justifies (p. 42). For example, Meyer
contends that 90 percent of the managers at General Electric rated
themselves above average. Meyer points out that almost everyone thinks
they are above-average performers, so when their salary increase does not
41


reflect their perception, the "effects of the actual pay increases on motivation
are likely to be more negative than positive (p. 44). In his study of a large
transit authority, Vest (1988) found that, merit increase satisfaction was
found to exhibit a significant positive correlation (r = 0.46, p, 0.0001) with
instrumentality beliefs (p. 148) (the belief that pay is tied to performance).
This suggests that the higher the merit increase, the more likely the employee
is to believe that pay is tied to his performance. With these findings in mind,
many companies have sought other pay strategies such as variable pay and
alternative pay strategies to replace merit pay (Sammer, 1996).
Group Incentive Pay
A group incentive program is a pay method designed to pay a uniform
reward based upon the group performance level of a predetermined goal.
Each member of the group shares the responsibility for the work output
(Johnson, 1993). Group incentive pay is increasing in use mainly because of
the greater emphasis on team-based work (Sammer, 1996; Heneman & Von
Hippel, 1995) and the interdependency required of todays workforce. The
other advantage to group incentive programs is that the incentive is based on
company performance, therefore, when company performance declines,
payments also decline (Sammer, 1996; Milkovich & Newman, 1993).
42


Increase in Group Incentive Programs
Group incentive programs are increasing in number as merit plans are
decreasing. Sammer (1996) reported that more than half of the companies in
an American Compensation Association (ACA) survey had reported
establishing variable pay budgets for group incentives. In addition to this,
nonmanagement employees are also being included into these group
incentive plans (Zitaner, 1992). Heneman and Von Hippel (1995) maintain
that, Critics of traditional merit pay plans charge that they decrease
cooperation among group members because they reinforce the
accomplishment of individual goals at the expense of group goals (p. 64).
Group-based incentives overcome this criticism because they compensate
the employees for their contributions to the group. This trend is occurring
because group incentive plans have boosted productivity for employers
(Schuster, 1984), and resulted in fewer grievances and improved labor
management relations (GAO, 1981). In fact, Milkovich and Wigdor (1991)
suggest that
The adoption of group incentive plans may provide a way to
accommodate the complexity and interdependence of jobs, the
need for work group cooperation, and the existence of work group
performance norms and still offer the motivational potential of
clear goals, clear pay-for-performance links, and relatively large
pay increases (p. 86).
43


The reasons that most organizations implement group incentive pay
are to: (a) foster product quality improvement (Milkovich & Wigdor, 1991;
Johnson, 1993), (b) improve productivity and financial outcomes (Milkovich &
Wigdor, 1991; Zitaner, 1992; Johnson, 1993), (c) revitalize the organization
consistent with business strategy (Milkovich & Wigdor, 1991), (d) more easily
implement a pay strategy that is not necessary to identify individual
contributors (Balkin & Gomez-Mejia, 1989), (e) increase employee
involvement in organizational goals (Milkovich & Wigdor, 1991; Zitaner,
1992), and (f) encourage cooperation among members of the same group
(Johnson, 1993; Weiss, 1987). Although organizations implement group
incentives for the above reasons, they acknowledge the ... importance of
contextual factors such as employee involvement, information sharing, and
ongoing marketing and communication to the employees covered (Milkovich
& Wigdor, 1991, p. 158).
Group incentives can change the productivity level of the employees.
Hansen (1997) completed a study within a large U.S. financial corporation
concerning employees responses to a group incentive program. He used
individual-level data collected on a monthly basis. The group incentive
program was based on reducing the amount of time callers had to wait in
order to receive service from the financial advisor manning a trading line or a
service line. His findings indicate that the incentive plan caused the
44


performance of the employees to converge to a standard. The least
productive workers improved greatly, whereas the performance of the most
productive workers did not change. The evidence suggests that the overall
productivity of the group increased. In addition to this study is a review of
group incentive research by Weber (1994) who found that more recent
research on the productivity effects of group incentives [results in] enhance[d]
productivity (p. 164).
Group Size
The effectiveness of group incentives may depend on the size of the
group. The literature concerning group incentives reports more success with
those programs using small-group incentives or team-based incentives as to
those using large group incentives. Large group incentives, such as profit
sharing or stock-based plans, increase the retention of employees, but they
do not increase employee performance or pay satisfaction (Balkin & Gomez-
Mejia, 1989). In small groups, workers have the opportunity to influence the
groups productivity and, in turn, influence their pay (Johnson, 1993; Nickel &
ONeal, 1990). This is also termed line of sight by Lawler (1990), who adds
that the workers influenceabilitv or line of sight can become obscured as the
size of the group becomes larger.
45


Drawbacks of Group Incentive Plans
The literature concerning the drawbacks of group incentive systems
speaks to the problems of motivation, social loafing, and line of sight.
Although group incentives avoid many of the pitfalls of individual incentives, it
is sometimes at the expense of less motivation for high effort and lower
rewards to high performance workers (Weiss, 1987, p. 138) and it often
creates the opportunity for lower performing employees to earn the same pay
as the higher performing employees (Heneman & Von Hippel, 1995).
Social loafing may occur when lower performing employees withdraw
their efforts in the hopes that the other group members will achieve the
groups goals (Heneman & Von Hippel, 1995). However, this situation can be
corrected through peer pressure of the group members on those social
loafers. Unfortunately, this may result in animosity among the group members
and may work against the supportive culture that groups need to operate
effectively (p. 64).
Line of sight, the ability to influence your pay, is an important variable
that needs to be considered in any kind of contingent pay system. When an
individual does not have control over his individual earnings, and his pay is
dependent on the other members of the group, line of sight may be
hampered. This can result in workers being less productive than in an
individual incentive pay setting (Dickinson et al., 1997). Organizations that
46


implement group incentive systems need to foster the employees beliefs
about their ability to influence the group performance measures (Milkovich &
Wigdor, 1991). Milkovich and Wigdor add that two of the required factors of
line of sight, that goals be doable and that the link between employee
performance and pay be clear" (p. 86), are not well satisfied in group
incentive plans.
Gainsharing
Gainsharing bonus systems are designed to reward all employees in a
business unit for organizational improvement (Bullock & Lawler, 1984; Dow
Scott, Little, & Markham, 1992; GAO, 1986) based on historical data
(Schuster & Zingheim, 1992). Gainsharing is a group bonus program
(Gomez-Mejia & Welbourne, 1988b) that attempts to encourage cooperative
behaviors in situations in which interdependency is critical in attaining desired
outcomes (Balkin et al.1995, p. 881). These programs are designed to
involve all employees in making decisions on saving costs, decreasing waste,
increasing productivity and quality control and then giving a portion of these
savings to the employees (Bullock & Lawler, 1984). In a study completed by
Bullock and Lawler, the typical payout percentage given to the employees
was 60-75%; however, others report that a 50/50% split is common to these
plans (Imberman, 1995; Lawler, 1990).
47


The primary objective of gainsharing plans is to create conditions in
which employees and managers benefit by moving on parallel paths
towards the common goal of improving productivity (GAO, 1986, p.8). These
programs are replacing programs based on individual or small group
productivity measurements, and are achieving high success rates (GAO,
1986). Gomez-Mejia and Welboume (1988b) assert that more companies
have implemented gainsharing in the last five years than did in the last fifty
years. Possible reasons for the increase in gainsharing are: (a) the union
stance of general dissatisfaction with merit pay and their search for a more
objective measure that would be perceived asfair by employees (Gomez-
Mejia & Welboume, 1988b, 1995), (b) organizations realizing the need for
flexibility, cooperation, and teamwork, (Bullock & Lawler, 1984; Gomez-Mejia
& Welboume, 1988b; Lawler, 1990), (c) the reality of shrinking middle
management (Gomez Mejia & Welboume, 1995), and (d) the plan is easier to
sell to top management since this type of program is self-funded (Gomez-
Mejia & Welboume, 1988). The success of gainsharing has been reported in
companies ranging from those that are extremely profitable to those on the
verge of bankruptcy (Gomez-Mejia & Welboume, 1988).
48


Increase in Gainsharing Plans
Gainsharing programs have grown dramatically in numbers. Balkin et
al. (1995) report that gainsharing has been in existence for more than two
generations, yet its use has grown dramatically in recent years (p. 881).
Lawler and Cohen (as cited in Balkin et al., 1995) found that in a survey of
1000 fortune 500 companies, 30% of these companies relied on some sort of
gainsharing program. This is a one-third increase in just three years.
Imberman, (as cited in Gomez Mejia & Welbourne, 1995) reported that at
least 2,000 firms are now using gainsharing, and it has now spread to all
industries, including the private sector (Welbourne, 1998; Zitaner, 1992).
Reasons for Increased Use of Gainsharing Plans
Gainsharing programs can significantly increase productivity. In a
study by Cooke (1994) of 841 manufacturing firms, it was found that
gainsharing increased individual and team performance by 18-20%. Two
studies of gainsharing that were described by Milkovich and Milkovich (1.992)
resulted in data suggesting both positive and negative implications of
gainsharing. The first study, completed by Schuster, (as cited in Milkovich &
Milkovich, 1992) consisted of a five-year study of the effect of gainsharing on
28 plants. Of those plants, over half posted significant gains in productivity,
and those that didnt were hampered by several variables, including
49


infrequent bonus payouts, poor relations, and a lack of employee input into
the plan design. Milkovich did, however, also cite some negative evidence in
the second study, which consisted of a gainsharing plan at AT&T where it
was found that individuals became less productive. The individuals who were
least productive prior to the plan improved in their efforts, but the most
productive employees reduced their efforts.
Another success described by Schuster (1984) is a nine-year
longitudinal analysis of a Scanlon plan implemented in a large manufacturing
plant. He combined a case-study approach with an interrupted time series
design. The productivity level and voluntary turnover was measured on a
monthly basis. During this time the plants employment turnover remained the
same, which ran counter to the industry nationwide. Productivity increased
and 70% of the 2,477 suggestions by employees were implemented.
Bullock & Lawler (1984) reviewed case study literature on gainsharing
programs and found that two-thirds of the plans were successful. They
measured success in terms of improvement in productivity, cost savings,
individual attitudes, ideas and suggestions, pay, and bonuses. Although not
all studies reported success in all areas of Bullocks study, most of them
reported success in at least some of the areas.
Gainsharing programs have increased in popularity for many reasons.
One reason is the data described above, which is increased productivity
50


resulting from gainsharing implementation. Another common explanation is
that gainsharing requires employee involvement and participation (Bullock &
Lawler, 1984; Lawler, 1990). Bullock & Lawler (1984) describe this as a
change in the culture of the organization that occurs after gainsharing
implementation. Employees are transformed from individuals who were
accustomed to working on their own tasks, unaware of how their jobs
interfaced with others, to a group that gained an understanding of the total
organization and its success.
Factors of Successful Gainsharing Programs
In order for gainsharing programs to be successful, they must include
several factors. Some of these factors are: (a) employee involvement in plan
design (Bullock, 1990; Imberman, 1995; Kim, 1999; Lawler, 1990); (b) use of
outside practitioners, expert guidance (Bullock & Tubbs, 1990; Flasch &
Imberman, 1990; Imberman, 1995); (c) employee favorability toward the plan
(Bullock & Tubbs, 1990; Schuster, 1984); (d) the committed and continued
involvement and support of managers (Flasch & Imberman, 1990; GAO,
1986); (e) supervisors trained on how to be receptive to suggestions, to be
cooperative, and to work in a participative environment (Flasch & Imberman,
1990; Lawler, 1990; Schuster, 1984); (f) trust in management (Gomez-Mejia
& Welbourne, 1988); (g) the pay-out formula must be reasonable and doable
51


(Imberman, 1995); (h) a program of employee education about gainsharing
must take place (Imberman, 1995); (i) the historical base must remain the
same throughout the duration of the plan (Lawler, 1990); and (j) continued
communication of the plan to employees must take place (Lawler, 1990).
In addition to the factors listed above, some consultants recommend
that employees vote on the plan so that they will obtain buy-in, thus reducing
the potential sabotage. If buy-in does not occur, managers will often complain
that there is no employee ownership of the plan (Dow Scott et al., 1992).
Managers will also have to change their leadership style from autocratic to
open and participatory if the gainsharing program is to be successful (GAO,
1986).
Results of Successful Gainsharing Programs
Successful gainsharing programs benefit organizations in multiple
ways, such as increasing productivity and enhancing the organizations
culture. A study based on a random sample of 845 companies completed by
the GAO (1986) resulted in the following responses to the question of whether
or not their gainsharing plans have resulted in non-monetary benefits.
52


Table 2
Percentage of Productivity Sharing Firms
Affirming Each Ncn-monetarv Benefit:
Improved labor management relations.......80.6%
Fewer grievances..........................47.2%
Less absenteeism..........................36.1%
Reduced turnover..........................36.1%
Other.................................... 47.2% (p. 18)
Successful gainsharing programs also enhance teamwork and
cooperation among employees (GAO, 1991; Lawler, 1990; Milkovich &
Wigdor, 1991), increase job satisfaction (GAO, 1981), improve productivity
(Doherty, McAdams, & Nord, 1989; Milkovich & Wigdor, 1991), increase
attendance (Doherty et al., 1989; GAO, 1981), result in employee suggestions
of cost saving ideas (Doherty et al., 1989; Lawler, 1990), increase employees
awareness of organizational goals (Doherty et al., 1989), result in more
innovation (Lawler, 1990), improve the climate of the management-labor
relations (GAO, 1981), result in less resistance to change (GAO, 1981), and
increase vertical and lateral communication (Doherty et al., 1989).
Barriers to Successful Gainsharing Programs
A variety of factors have been identified that might prevent gainsharing
programs from being successful. Preivisch (as cited in Doherty et al., 1989)
lists some of these factors as changing business conditions, design problems
(such as a poor accounting system), and the implementation process.
53


Doherty et al. (1989) adds that failing to achieve the full support of the
management is also a detriment to the success of a gainsharing program. In
addition, Flasch and Imberman (1990) argue that a lack of management
credibility will decrease the chances of a company developing a successful
gainsharing program. They report that cynicism [will] set in when
management reserve[s] the sole right to interpret and announce the official
results (p. 2). If the organization consistently changes the bonus calculation,
the management and system will lose credibility (Gomez-Mejia & Welboume,
1988). Imberman (1995) suggests that paying out small bonuses or assuming
that large improvements will occur immediately, results in failure of the plan. If
any of these factors are present during the design process or implementation
stage, or throughout the duration of the program, it will be difficult to introduce
or maintain a successful gainsharing program.
Determinants of Success for Pav for Performance
There are many determining factors that either ensure the success or
demise of pay-for-performance programs. In line with the various groups in
this study (transportation, fleet maintenance, and administrators), the
following determinants of success have been identified as integral to the
outcome of the pay-for-performance plans.
54


1. Employee Participation in the Implementation
Process
It is believed by many pay-for-performance consultants and experts
that pay decisions require involvement from employee groups that are
affected by the decision in the design and implementation phase of the plan
(Balkin & Bannister, 1993; GAO, 1986, 1991; Johnson, 1993; Kessler, 1994).
In a study completed by Jenkins and Lawler (1981), data were collected on
the effects of having employees participate in the design phase of a base-pay
plan for a small manufacturing firm. The results of attitudinal and behavioral
data of 58 employees collected before and after the development and
implementation of the plan were in favor of employee participation. The
understanding of the plan, job satisfaction, pay satisfaction and organizational
trust all increased. It was concluded in this study that employee participation
in pay plan development resulted in a better overall relationship between the
employees and their organization and created a more effective pay plan.
Another reason to involve individuals in the design of new pay plans is
that employees will be more likely to accept the new system and be more
committed to seeing that it operates effectively. The participative decision-
making process also helps the employee gain a better understanding of the
program. During the process, individuals from a variety of areas within the
organization can offer information that others do not have, which results in
55


better data being brought to the process (Lawler, 1990). With participation of
employees integrated into the process, widespread ownership of the pay
program becomes the responsibility of everyone in the organization to see
that the program is a success. If the system is not participative, employees
and managers often imagine ways that the program could have been
improved if they had been in charge (Lawler, 1987).
Employees from each level of the organization should be included in
the design, implementation, and evaluation phase of the process (GAO, 1991;
Lawler, 1990). They should be given parameters of involvement before the
actual involvement takes place, so the employees will feel that their
expectations were met (Schuster and Zingheim, 1992). This is much more
productive than assembling a group of employees without providing
parameters. This can often result in a product that the management will not
accept, further frustrating the process and employees.
Opposing the stance of employee participation is Welboume (1998),
who completed a study of two companies that had implemented gainsharing
plans. Welboumes study focused on procedural fairness. In her findings, she
suggested that employee participation in the planning and implementation
phase might have led to the demise of the high technology firms program.
She explained that the firm spent so much time convincing employees that
the plan was fair and effective, that by the time it was finally implemented, the
56


employees rebelled against the management. One employee even stated to
Welbourne, This is another management joke (p. 9). On the other hand,
employees in the other organization being studied took a wait and see
attitude when their plan, developed by management, was implemented.
Welbourne encourages constant and honest communication of the plan to
employees instead of early participation of employees in the development of
the plan.
2. Link Between Pay and Performance
Employees must perceive a clear link between pay increases and
performance (Milkovich & Wigdor, 1991; Eskew et al., 1996; Lawler, 1990;
Marsden & Richardson, 1994). Although pay-for-performance plans have
increased in numbers, there is considerable evidence that most organizations
fail to create a perceived relationship between pay and performance
(Lawler, 1987). In a study completed by Immerwahr (1983), it became
apparent that employees do not perceive a link between their performance
and their pay. Almost three quarters of the jobholders felt that employees who
put in the extra effort are not rewarded, and that same number say that the
quality and effort that they place in their jobs is not a determining factor in
their salary.
57


3. Line of Sight
The measures to determine pay increases must be seen as credible
and susceptible to influence by the employee. Lawler terms this line of sight.
The employee must be able to affect his or her performance measure through
his work behavior (Lawler, 1990). In a study by Bennet (1997) of various sites
that have implemented pay-for-performance, he found that if the pay-for-
performance program were to extend beyond the employees site boundary
that it is very hard for workers to feel that they can affect the performance
measures. This is directly attributed to line of sight because the employees
lose the perception that they can directly influence the performance
measures. The employees must be able to perceive that the task is doable,
given their own abilities and skills (Milkovich & Wigdor, 1991). Line of sight
must be apparent if the pay-for-performance plan is to be successful.
4. Organizational Trust
Organizational trust must be present prior to the implementation
process of pay-for-performance programs. The organizational climate should
be one that is based on shared goals and values between managers and
employees (Ingraham, 1993a). In Ingrahams (1993b) study of private-sector
organizations, five characteristics were found to be common among the
successful pay-for-performance programs. The first characteristic on this list
58


was employees and managers have high levels of organizational trust and
commitment (p. 198). This seems to be a common theme in successful
programs, as Vest (1988) argues in his study, that individuals are more likely
to perceive that their pay is based on their performance if they are satisfied
with their pay increase and they have trust in the top management. The top
management needs to place some time and emphasis on creating trust
among its employee groups. One way to achieve this is to consistently
communicate the pay plan to the employees so they have an understanding
of the pay plan. This communication most often results in employee
commitment to the plan (Jenkins & Lawler, 1981).
5, Aligned to the Strategic Plan
The compensation system of an organization is used as an integrating
system to ensure that the employees efforts are directed toward the
organizations strategic plan. When the plan is designed properly, the pay
system then becomes an integral part of the effectiveness of the organization
(Gomez-Mejia & Welbourne, 1988a; Ingraham, 1993b). Hays (1994)
discusses the benefits and pitfalls of pay-for-performance programs and
suggests that successful pay-for-performance plans help support the
organizations goals in areas such as profit margins and customer
satisfaction. Furthermore, if an organization has a clear sense of its strategic
59


plan, then it can place an emphasis on the types of skills needed to achieve
these goals. It can also offer incentives that target these skills and objectives
to create a competitive advantage in the marketplace (Lawler, 1990).
Although Schuster and Zingheim (2000) argue that few businesses design
their compensation plans to be strategic, companies that do, address pay
and rewards as a lead element of change will gain strategic advantage (p. 15)
6. Decentralization
If the organization is structured in such a way that decentralized units
are required for performance effectiveness, then each unit should have its
own reward system (Lawler, 1990). Bennet (1997) contends that successful
pay-for-performance programs are sensitive to the site location. Almost three-
quarters of employees surveyed felt that they have a large impact on their
specific site but did not feel that they can impact performance measures once
they cross the site boundaries. It is strongly recommended that decentralizing
organizations pay-for-performance plans occur in order for these units to
have the flexibility to develop programs that fit their unique qualities (Kim,
1999; Lawler, 1990; Milkovich & Milkovich, 1992; Milkovich & Wigdor, 1991).
60


7. Appraisal Process
The appraisal process should be focused on observable behavior, be a
quantitative measure that can be translated into the pay system, and be
perceived as fair by the employee. In a review of research on pay-for-
performance programs, Milkovich and Wigdor (1991) found that performance
evaluation was critical to the pay-for-performance plans in order to improve
performance and to be accepted as fair and credible. In order for effective
pay-for-performance programs to exist, the appraisal process must include
credible comprehensive performance measures (Lawler, 1987, 1990). If this
requirement is not met, then the pay plan will not effectively relate pay to
performance in a motivating way. The organization must first be able to
identify the behaviors that it wants to motivate. The performance appraisal
should then focus on these observable behaviors and results (Lawler, 1990).
Part of the performance evaluation in a merit pay or variable pay plan
should include performance goals or standards as a basis for the evaluation
(Gaertner & Gaertner, 1984). This will benefit the employee and the
organization by placing the focus on challenging but attainable goals for the
individual, while at the same time meeting the needs of the organization.
The appraisal process should be perceived by the employee as fair. In
their review of literature on performance evaluation, Milkovich and Wigdor
(1991) and Greenberg (1986) found that studies suggested if the employee
61


were given the opportunity for input into the performance appraisals, then it
would lead to the perception of fairness in the evaluation process.
Furthermore, perceptions of fairness in evaluations will be enhanced if the
supervisors will clearly explain why certain ratings were given and what the
employees can do to improve these ratings over time (Eskew et al., 1996).
Milkovich and Wigdor (1991) also found that it is key to the success of
appraisals that the employee perceive that the appraisal represents a
reasonable estimate of the employees performance, that the evaluator has
expertise in the area being evaluated, and that there is trust between the
evaluator and the employee. Finally, the measurement formula for the
performance must be understandable by the employees, simple, and under
the employees control (GAO, 1986). For a successful appraisal process of
employees involved in pay-for-performance programs to take place, the
following must occur: (a) the process should focus on observable behavior
and be quantitative in nature, (b) the employee should be allowed input into
the appraisal, (c) a clear explanation and honest feedback should be given to
the employee, (d) a perception that the supervisor has expertise must exist,
(e) a perceived trust between supervisor and employee must exist, and (f) an
understandable and practical measure of performance must be simple and
under the employees control.
62


8, Manager Flexibility
Managers should have the discretion and authority to hire, recognize,
reward, and terminate employees if pay-for-performance programs are going
to be successful (Ingraham, 1993a). Giving managers this authority does not
ensure success. Managers have consistently failed to recommend large pay
/
differentials for their employees when large performance differences occur
(Betz, 1987). Although some managers do attempt to award salary increases
to top performers, they are not given this flexibility in the public sector
organization (Milkovich & Wigdor, 1991).
9. Adequate Financial Resources
Research and experience have shown that financial resources must be
available so that employees consider the rewards to be meaningful (GAO,
1991; Ingraham, 1993a). The credibility of the pay-for-performance program
can only be maintained if the base pay remains the same and is not
manipulated to free funds for the new pay program. In addition, the pay-for-
performance program must be properly funded and not required to be budget
neutral (GAO, 1991). This must occur so that pay increases are adequate and
awarded to all deserving employees (Greenberger et al., 1991).
In a study by Montemayor (1995), he found that many researchers
argue that the meritocratic principle of pay-for-performance programs will not
63


operate effectively unless adequate resources are available to increase an
employees pay above the rate of inflation. In his study, he hypothesized that
pay-for-performance norms would be more 'meritocratic (larger differences in
raises across performance levels) (p. 1) when adequate funds are available.
The results did indeed show that when there is a scarcity of funds, pay-for-
performance norms become somewhat egalitarian, in contrast to when there
is ample funding and the payouts are more meritocratic.
10. Employee Perception of Fairness
Employees must perceive the pay-for-performance plan as being fair in
terms of the procedures used for the plan (Eskew et al., 1996). The actual
procedures or process used to administer the pay plan constitute procedural
justice, while the actual payout or the ends achieved is distributive justice. In
a review of literature on organizational justice, Balkin et al. (1995) find that the
perceived fairness of procedures may be more important to agents than the
perceived fairness of outcomes (p. 16). Studies have shown that the
employees perception of procedural fairness significantly influences his or
her impression of the fairness of the pay-for-performance plan. In addition,
employees and supervisors are more likely to accept low pay if they believe
the process to determine the outcome was fair (Milkovich & Newman, 1995).
The research completed by Milkovich and Newman also suggested four
64


ingredients for a pay plan to be perceived as fair. These are: (a) if they are
consistently applied to all employees, (b) if employee participation and/or
representation is included, (c) if appeals procedures are available, and (d) if
the data used are accurate (p. 49). Welboume (1998) argues the point
established above, that employee participation is needed for successful
program implementation. She states that, although procedural fairness is
crucial to pay programs, especially with low payouts, this may not necessarily
require early employee participation, but does require ongoing and honest
communication. Gomez -Mejia and Welbourne (1988b) add to this research in
their findings that procedural justice is only achieved if explanations to
employees are adequately reasoned and sincerely communicated (p. 412).
Pay-for-performance programs must ensure that all employees are
treated fairly. Supervisors need to establish trust and respect among their
employees and should not use non-performance-related factors in their
consideration of ratings (GAO, 1991). If the employees do perceive the
process as fair with regard to pay distribution, then this will result in increasing
their job satisfaction and their trust in management (Milkovich and Wigdor,
1991).
65


11. Education and Communication of Plan
Educating employees about the new compensation plan and
communicating with them during each phase of its development is imperative
for the success of the plan (Imberman, 1996). Educating the employees about
the plan is not simply a matter of telling them what the plan is, but it is
educating them about the plan with a question and answer period and follow-
up meetings to ensure understanding of the plan (Imberman, 1995). As
Schuster and Zingheim (1992) found, it is also important to communicate the
pay plan clearly and to make this communication specific to the employee,
while at the same time, not try[ing] to sell the program to the employees (p.
76).
In conclusion, the determinants of success are theoretically based, and
if these determinants of success are attained, one would assume that the
pay-for-performance program being implemented would have a good chance
of success.
Diffusion: Public-Private Sector Differences
Ingraham (1993a, 1993b) has completed several studies regarding the
diffusion model, transferring processes such as pay-for-performance
programs from the private sector to public-sector organizations. She
questions to what extent these techniques are transferable and concludes
66


that public-sector organizations do not share the same attributes as their
counterparts, private-sector organizations. Her findings suggest that key
decision-makers often assume the successful transferability of pay-for-
performance programs, but these assumptions are, as she stated, without
foundation (p. 348). In fact, Ingraham, Jacobson, and Selden (2001) argue
that the acceptance by political leaders is, that private-sector management
practices are good, effective, and efficient, while public-sector ones are bad.
ineffective, and inefficient (p. 598).
A classic example of what Ingraham describes as problematic of
diffusion is apparent in her example involving the chief architect of the federal
Civil Service Reform Act of 1978. The architect was asked ten years after the
I
failure of the reform, ... why the act had not provided for experimentation
and evaluation of pay-for-performance prior to government-wide
implementation. He replied, I saw no need. It was my perception that it
worked fine in the private sector (U.S. Senate, 1988, p. 28 as cited in
Ingraham, 1993a).
Often, public-sector employees are combined into a category labeled
public service, and this label is then treated as a synonym for all government
employees. Elmer Staats (as cited in Perry & Recascino-Wise, 1990)
describes public service as ... a concept, and attitude, a sense of duty-yes,
even a sense of public morality (p. 368). This statement more closely depicts
67


the definition of public-sector employees and organizations, than the too
commonly used term of government worker. Unfortunately, the attributes of
the public-sector organization are much more difficult to overcome. Several
studies show that, although public-sector managers support the philosophy of
pay-for-performance plans, they often fall back on their traditional
bureaucratic system that requires them to support equal pay for all people in
the same salary grades and steps (Milkovich and Wigdor, 1991).
Key to achieving success in pay-for-performance implementation is the
ability to overcome obstacles that often appear in private-sector organizations
but not public-sector organizations, such as: (a) managers having discretion
to recognize employee achievements, reward workers adequately, and
terminate them if necessary (Ingraham, 1992); (b) adequate resources being
available (Ingraham, 1992; Pearce & Perry, 1993); and (c) not sharing
information about all wage and salary increases (Milkovich and Wigdor, 1991;
Pearce & Perry, 1993).
Amorphous Jobs
More jobs are becoming that of knowledge work (Lawler, 2000).
Organizations are taking on increasingly complex challenges and for
knowledge-based organizations people are a critical and valuable asset that
must be better understood, invested in, and strategically directed toward the
68


achievement of identified organizational results (Ingraham, Moyihan, &
Selden, 2000, p. 57). However, it is more difficult to measure the individual
employee in the area of service, knowledge work, and jobs that require
interdependent skills (Lawler, 1990). Further, the managers job is most often
amorphous. It does not lend itself to quantifiable measures, simply due to the
many tasks performed by managers, which are not directly observable
(Milkovich & Wigdor, 1991). Managerial jobs are complex and uncertain and
require a wide range of skills to meet the daily demands the job requires
(Perry, 1986). Milkovich and Wigdor (1991) stipulate that the bulk of existing
research on performance-pay relies on data gathered from jobs that are more
concrete and have direct measures of performance. Gaertner and Gaertner
(1984) add that as a result of the difficulty of tying managers jobs to work-
related behavior, the actual administration of pay in many organizations is
more seniority-based than performance-based, lip service about merit pay
notwithstanding (p. 89).
An example of the difficulty of using performance measures in the
knowledge-work area is described in a study completed by Balkin and
Gomez-Mejia (1992). They describe the nature of the work of university
professors as nonprogrammable, or not easily measurable. Because
programmable tasks allow the supervisors to specify which tasks are easily
observed and the university professors jobs lacked this ability, the
69


measurement often used for performance evaluation is simply the number of
publications that one produces in a given year.
Another unique feature of the managerial position is that objective
measures of the job may not be specifiable in advance. This presents a
problem with the common practice of paying managers for their attainment of
goals. In their review of goal-setting research, Latham and Yukl (1975) found
that goal-setting programs for managers were not as successful as those
programs for simple jobs. They suggest that the complexity of the managerial
jobs is the reason for the lack of consistent success of goal-setting programs
for managers. In their study of merit pay programs, Pearce et al. (1985)
argues that managerial jobs may often include performance that is not
specifiable in advance. To avoid risking failure of the goal-setting plan, they
suggest that a combination of objective and subjective performance
measures be used in the managerial performance evaluation.
The administrators pay-for-performance plan being studied herein
resembles that of many managerial variable pay programs of managers.
Therefore, the concerns stated above mirror the concerns for the
administrators pay-for-performance plan in this study.
70


Research is Needed
Much research is needed in the area of performance-related pay
schemes, whether it be in gainsharing or group incentives, the process of
implementation, or measuring the acceptance of the plan by employees
(Marsden & Richardson, 1994; Milkovich & Wigdor, 1991; Weber, 1994).
Studies that directly measure the perceived fairness of pay-for-performance
plans are scarce (Milkovich & Wigdor, 1991). In fact, Ingraham (1993a) states
that the American Political Science Association reports that personnel ranks
at the very bottom of the research interests for those political scientists who
have an interest in public administration (p. 355). She adds that this is a
concern since effective governance is also dependent upon a sound
compensation program. In addition to the limitations of the data that are
collected on pay-for-performance programs, is the focus on blatant pro- or
anti-group incentive bias (Weber, 1994, p. 11). The most commonly cited
studies have been those that examine successful programs. It seems that few
organizations spend the time and funds to study unsuccessful compensation
programs.
The design of the surveys, more specifically, which employees are
surveyed, is also a concern that Bretz et al. (1992) emphasize in their
suggestions for future research. Bretz et al. ask that instead of sending a
survey to be completed by one individual within an organization, this
71


questionnaire should be sent to multiple people at various organizational
levels in order to more accurately represent the entire organization.
Despite the popularity of a long history of gainsharing programs,
research on their effectiveness is limited. Reasons for this are: (a) the
research is atheoretical (Balkin et al., 1995); (b) most studies are completed
by authors who take an advocacy position, therefore, gainsharing failures are
not adequately represented (Balkin et al., 1995; Gomez-Mejia & Welbourne,
1988b; Welbourne, 1998); (c) much of the research relies on testimonials
instead of empirical analysis (Gomez-Mejia & Welbourne, 1988b); and
(d) most research has been completed in the manufacturing sector with little
done in the service sector (OBannon & Pearce, 1999). In addition, Doherty et
al. (1989) call for longitudinal studies of programs that begin early in the
design process and continue through the implementation phase. They see the
need for research to be completed regarding the process, implementation,
and maintenance of the new programs. For gainsharing plans to be effectively
studied, the research should include empirical evidence of the success or
failure of the plan and the perceptions of the employees on the acceptance
and fairness of the program.
Although the use of group incentive programs is escalating, little is
known about the success of these programs. Much of the research that has
been conducted has emphasized the outcome of these programs, rather than
72


the process required to achieve a successful plan (Weber, 1994). Weber
adds to this by saying that because there are only a small number of firms
that use group incentive programs, there are ample case studies completed,
but survey research on group incentive systems is lacking.
More research is needed to measure the perceived fairness of pay-for-
performance plans, to add to the literature that is limited due to blatant pro-or
anti-group incentive bias, and to add to the validity of survey studies by
collecting data of more than one employee per organization. More research of
pay-for-performance programs needs to be based on empirical analysis and
include more longitudinal studies in this area of research. Finally, the process
used for the implementation of pay-for-performance plans should be
addressed, as this is the basis for the pay-for-performance plan and can, in
the end, predict the success or demise of the plan.
Summary
In this chapter, individual incentive pay, group incentive pay, merit pay,
and gainsharing programs have been presented. Through this literature and
research, determinants of success were identified. It was argued that if these
determinants of success are in place in the new pay plans, then the plans will
be successful. On the other hand if the determinants of success are not
infused into the pay plans, then the chances of new pay plans being
73


successfully established will be lessened. The determinants of success that
must be apparent in new compensation plans are: (a) employee participation
in the implementation process, (b) a perceived link between pay and
performance, (c) line of sight, (d) organizational trust, (e) alignment to the
strategic plan, (f) decentralization of the plan, (g) an appraisal process that
focuses on observable behavior and is based upon a quantitative measure,
(h) manager flexibility in hiring and rewarding employees, (i) adequate
financial resources, (j) perception of fairness by the employee, and (k) the
education and communication of the plan.
Chapter 3 provides a detailed discussion of the research methodology
utilized to investigate the proposed relationships between the determinants of
success and the success of the new pay programs being studied. Topics
addressed include the research design, methods of data collection, subjects,
methods of data analysis, description of the study, and display of the data.
Chapter 4 presents the data and findings. Chapter 5 provides a summary of
the findings, conclusions drawn from the findings, and limitations of the study,
while addressing the implications of the study for practice and future
research.
74


CHAPTER 3
METHODOLOGY
This chapter presents the research methods, design, and description
of the study. Data collection procedures and methods of data analyses will
also be presented.
The purpose of this study was to determine if the four pay-for-
performance plans were successful and to discover if all or some of the
determinants of success were manifested in the four pay-for-performance
plans and if the information was useful or not in understanding the
employees perception of success of their plans. This information should help
one to understand why some of the plans were successful or not and why
some plans were terminated, while some still exist. This study will add to our
understanding of the workability of pay-for-performance in some areas of
education.
Research Questions
The following research questions will be answered in order to
accomplish these objectives:
75


Research Question 1
How are the determinants of success derived from literature
manifested in the four pay-for-performance plans?
Research Question 2
How are the determinants of success derived form literature
useful or not useful in understanding the four pay-for-performance
plans and in how they are perceived for success by the employees?
Research Question 3
Why were some of the four pay-for-performance plans terminated
while others continue to exist?
Research Question 4
What does this study add to our understanding of the workability
of pay-for-performance plans in some areas of education?
Research Methods
This study consisted of three individual case studies and included both
qualitative and quantitative research methods. Case study research is
appropriate when investigators desire to (a) define topics broadly and not
narrowly, (b) cover contextual conditions and not just the phenomenon of the
study, and (c) rely on multiple and not singular sources of evidence (Yin,
1993, p. xi). The topics that were addressed were broad in nature, with each
pay program being comprised of many variables that might affect the plan.
The context of each pay program was different in the way it was implemented
76


and in the composition of each department. For instance, some departments
consisted of white-collar workers whereas other departments included only
blue-collar workers. Finally, as Yin suggests, multiple sources of evidence are
indeed used as strategies for this study.
Yin (1993) encourages the case study researcher to identify units of
analysis, which are the pay-for-performance plans, and the context, which are
the three departments (transportation, fleet maintenance, administrators and
managers). Because the context includes an embedded design of three
departments, Yin suggests that different measurements be used for the
various contexts. Each employee group received a questionnaire unique to its
department. Interview questions were similarly developed to address specific
needs within each pay-for-performance plan.
Yin (1993) also cited the need for the researcher to decide on (a) a
single or multiple case study, (b) a site selection process, and (c) a data
collection method, whether it be a one-time data collection or an extended
period data collection. This study was comprised of three single case studies.
The site was selected because of its accessibility and its feasibility and for the
variety of plans being developed and implemented. A one-time data
collection, comprised of a questionnaire, was planned; however, after
interviewing and re-interviewing subjects over the years, I extended the data
collection period, in order to add to the validity of the study.
77


This study was largely qualitative in nature because quantifiable
measures of the variables are not always possible. According to Liebscher
(1998), qualitative methods are appropriate when the phenomena being
studied are social in nature and cannot be measured through quantification.
He adds that many researchers combine methods in order to gain greater
insights, which is my intention for this study.
The primary purpose in using qualitative procedures was to help
formulate types of questions for the employee survey and to provide
contextual meaning to the quantitative assessment. It is also suggested by
Yin (1993) that case-study research should involve quantitative research
methods. The quantitative results were used to supplement the qualitative
data.
Another major reason that I chose to use both qualitative and
quantitative methods is that this is in line with the literature in social science
research methods that advocates the use of multiple methods, termed
trianaulation (Jick, 1979; Yin, 1993). Jick underscores the importance of the
researchers use of both methodologies. His premise is that when quantitative
methods are used in addition to qualitative methods, it allows the researcher
to systematize observations, to utilize sampling techniques, and to develop
quantifiable schemes for coding complex data sets (p. 604). In addition, Jick
says that surveys become more meaningful when coupled with qualitative
78


information. Although the literature suggests that qualitative measures are
legitimate as the sole method of research, survey research will add to the
validity of theresults.
A between method triangulation, as labeled by Denzin (as cited in Jick,
1979) was used as a triangulation method. Denzin explains the between
method as one that is largely a vehicle for cross validation when two or more
distinct methods are found to be congruent and yield comparable data (p.
602). Jicks example of how multiple methods are used is that of measuring
the effectiveness of a leader. For instance, the effectiveness of the leader is
measured by interviewing the leader, observing his or her behavior, and
evaluating performance records. The type of data collection may vary, but the
focus is still on the leaders effectiveness. This is similar to this study, as it
focused on the effects of individual and group incentive plans on employee
productivity, as measured by interviewing, surveying, and collecting employee
performance records. In addition, data were collected from multiple
viewpoints, some direct and indirect. In other words, not only were employees
asked in interviews and through the questionnaire about their attitudes and
opinions of the success of the incentive plan, but supervisors were also asked
about their perceptions of what the employees attitudes were and if
employee performance levels changed.
79


An attitude survey was used as one method to measure the success of
the individual and group incentive programs. Experts in the field of
compensation have used this same method to collect data on employee
attitudes in order to evaluate new pay programs. In fact, Eskew et al. (1998)
conducted surveys to gather data on employee attitudes toward the pay
package they were studying. Their stance on using surveys as a measure of
the effectiveness of pay plans is that it is often difficult to distinguish between
the effects of a new pay program and changes that take place in an
organization. In their review of research, they found that positive employee
attitudes toward the pay system are linked to reduced turnover, improved
attendance, and a decreased probability of unionization.
The software program QSR NUD*IST was used as a tool to code the
data sets in this study. Software packages used for analyzing qualitative data
are becoming widespread. These are termed Computer Assisted Qualitative
Data Analysis Software (CAQDAS) and are being adopted by key members
of the qualitative research community (Atkinson, Coffey, & Holbrook, 1996). I
utilized the program QSR NUD*IST to support the analytical process of
qualitative research. Although QSR NUD*IST has provided another technique
to test some of my conclusions to the research questions, it has not replaced
my role as a qualitative researcher in understanding the meaning of interview
text, meeting observations, and open ended questions from the survey. I have
80


heeded the advice of Catterall and Maclaran (1997), who argue that
interpreting the text in interviews needs to be completed in the context of the
conversation and not just in coded statements. They add that one must be
cautious in taking segments of text out of context and ignoring the full
meaning of the text as it is aligned to the concept.
The data collected through the questionnaire and analyzed through
SPSS (1994) were used to supplement and validate the data collected
through the interview process. This validation was accomplished by providing
the mean response to each statement on the questionnaire and classifying
these responses by determinant of success.
Design
Both qualitative and quantitative methods have been utilized for these
three case studies. Newman and Benz (1998) contend that in qualitative
research, the data is collected and analyzed, which results in the conclusion,
hypothesis, and finally theory. This is almost reverse of what is practiced by
quantitative researchers who begin their studies with the theory, review of
literature, and hypothesis, and then continue with data collection and the
steps for qualitative research. The qualitative-quantitative philosophy of
educational research suggested by Newman and Benz allows the researcher
to utilize self-correcting feedback "loops. For instance, the first step of this
81


study was to interview subjects (data collection), analyze the data (analysis),
prepare and disseminate a questionnaire from literature and interview
responses (literature review, data), to analyze the data (analysis), and draw
conclusions (conclusions) that answered research questions (hypothesis) and
which resulted in implications of the research (theory). This method allowed
me to begin with the first step of a qualitative researcher but to loop back
after data were collected in order to collect quantitative data.
Design Validity
Yin (1993) argues that case-study research should involve quantitative
data where relevant, and the rigor of case study research should include
criteria for validity. Newman and Benz (1998) have identified several criteria
that are a framework for establishing design validity. They contend that their
list is not exhaustive, but if most of these criteria are in place, it will add to the
validity of the qualitative study.
82


1. Neutrality. The data were objective in the sense that employees with
various opinions regarding the new pay-for-performance plan (opponents,
proponents, fence sitters) were interviewed. Further, 100% of all employees
who were eligible to participate in the plan were given a questionnaire
focused on their perceptions of the new pay plan.
2. Prolonged Engagement On-site. The first interview took place in
1996, and more interviews occurred each year through the year 2002. In
addition, questionnaires were disseminated to each employee involved in the
pay-for-performance program in the year 2000.
3. Persistent (Consistent) Observation. Many employees were
interviewed twice between the years 1997 and 2000. One supervisor who
was immersed in these plans was interviewed on four different occasions in
1996 through 2002. These various interviews provided data that indeed,
resulted in consistent information.
4. Trianaulation. A variety of data sources was used in this study,
including interviews, a questionnaire, observations of pay-for-performance
committee meetings, data regarding employee productivity levels, and data
collected from both employees and supervisors of employees.
5. Member Checking. Those subjects who were interviewed more than
one time briefly reviewed their initial interview to check for accuracy. Two
83


subjects, who provided a large amount of information due to their positions in
the organization, were frequently asked to confirm the data content they
provided.
6. Theoretical Sampling. Following the interviews, additional
statements originating from the interviewed employees were added to the
questionnaires. These statements included their concerns and issues
regarding the pay-for-performance plans.
7. Leaving an Audit Trail. Other researchers can replicate this study for
the purpose of developing or evaluating new pay-for-performance plans in
their organization by using the same questionnaire and the identical
researched based framework.
8. Generalizabilitv. Although Newman and Benz (1998) contend that
the nature of the qualitative paradigm does not include generalizability, they
assert that, The following concepts, applicability and transferability (context
limited), are examples of the kinds of questions to ask to improve those
efforts (p. 54).
Applicability. The samples being studied were described in enough
detail with literature cited that is specific to each group so that another
researcher could judge whether or not his group of employees was similar
enough to apply the same results.
84


Context Limited (Transferability). The determinants of successful pay-
for-performance programs that were utilized in this study were those that
spilled over into four different types of pay-for-performance plans and were
not specific or unique to any one of the plans.
9. Truth Value. Many of the criteria for design validity, identified by
Newman and Benz (1998), were apparent in this study. Similarly, truth value,
which is the result of the extent that the above components exist in the study,
was also inherent in the study.
Description of Study
This study consisted of three case studies, one for each employee
group (transportation employees, fleet maintenance employees, and
administrators). Case study research is a design that fits under the qualitative
research method. Benz and Newman (1998) identified the sources of data
collection for case studies as ... interviews, observations, documents, and
historical records (p. 66), all of which were data collection strategies utilized
in this study.
Validity of case study methodology is increased by the fact that
multiple data collection techniques were used, such as interviews,
observations of pay-for-performance committee meetings, and documents
85


with data regarding changes in employee productivity (Benz & Newman,
1998).
The first step in this study was to interview subjects on their
perceptions of the new pay plan. I sought approval from the Human Research
Committee (HRC) at the University of Colorado, Denver, in order to interview
these subjects. The study was approved and given the protocol number of
250. This qualitative approach of interviewing employees allowed me to
explore, in depth, the employees perceptions regarding the new pay plan. It
also permited me to grasp the entire process or to have a holistic picture of
the phenomenon, as Krathwohl (1993, p. 352) describes. The study was
based on the process of implementation of pay-for-performance programs,
rather than on the performance program itself, and qualitative methods were
the appropriate tool to accomplish this task.
The quantitative method utilized in this study established the mean of
responses to each questionnaire statement that was associated with the
determinants of success found in literature on pay-for-performance programs.
The questions were categorized by each of the eleven determinants of
success, and the mean of each question was tabulated separately, with the
responses shown in table form.
86


Methods of Data Collection
Data were collected through in-depth interviews, observations,
questionnaires, and employee performance records. The questionnaires were
developed following the interviews so that employee concerns expressed
during the interview could be placed on the questionnaire. For instance,
several employees in fleet maintenance mentioned their perception that
management manipulated the bonus program during the first year. The bonus
manipulation was meant to ensure that employees would vote for the
gainsharing program that was to be implemented following the pilot year. A
question, specific to this perception, was then placed in the questionnaire.
The interviews included employees in all three employee groups
(transportation workers, fleet maintenance crews, and administrators and
managers). Thirty minutes were allotted for each audio taped interview. Prior
to the interview, interviewees gave me permission to record the conversation.
The interviews were semi-structured, as the questions were pre-determined,
open-ended and placed in a specific order. These conversations were then
transcribed and coded with the application program QSR NUD*IST (1997).
Either a paid clerical worker or I transcribed the information. To attain
accurate information, I again proofed all transcribed work. In addition, I
completed all coding. The interview questions are found in Appendix A.
87