$215531:
£13"

. ‘2
rain“:
.
. v . I
133?...ﬂdul T
ﬁst... z...
. . ~b|z .
T3.
.IlL uvqu _ Q
. .R— 3.11. .
V I n!
n3 was a
.3.
h... .
v 1.: Qty... In,
$5.3...
{ ~

 

.115. ‘1. ... ii c
v.73. A .
‘ . a.“ :l
.. I! u . \
. 1
.. isﬂnwai
. I .133
u .-

 

$1.-
r x. .2... ,

 

k mﬁamwﬁhﬂw

 

THESIS

£002.

 

 

LIBRA—FTY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

ALLEVIATING THE UNINTENDED CONSEQUENCES OF
AFFIRMATIVE ACTION:
ACHIEVING ATTITUDE CHANGE BY
INVOKING ALTERNATIVE ETHICAL FRAMEWORKS

presented by

EDWARD I . FUBARA

has been accepted towards fulﬁllment
of the requirements for

PhD degree in Business Administration

 

 

ﬂ/zf%

/ Major professor

 

Date /2.// 21/0/

MSU is an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 alClRC/DateDuepGﬁ-p. 15

 

ALLEVIATING THE UNINTENDED CONSEQUENCES OF
AFFIRMATIVE ACTION:
ACHIEVING ATTITUDE CHANGE BY
INVOKING ALTERNATIVE ETHICAL FRAMEWORKS
By

Edward I. Fubara

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Management

2001

ABSTRACT
ALLEVIATING THE UNINTENDED CONSEQUENCES OF
AFFIRMATIVE ACTION:
ACHIEVING ATTITUDE CHANGE BY
INVOKING ALTERNATIVE ETHICAL FRAMEWORKS
By

Edward I. Fubara

In spite of the laudable goals of afﬁrmative action and afﬁrmative action
programs, they have met with considerable opposition, even from those whom they are
designed to help. One of the sources of such opposition, and other unintended outcomes,
‘ is the fact that afﬁrmative action is often seen as unfair, unjust, or unethical. This study
attempts to clarify the relationship between individuals’ attitudes toward afﬁrmative
action and their ethics and justice frameworks. It also attempts to reduce the negative
attitudes associated with afﬁrmative action by altering the ethical frameworks with which
individuals evaluate the initiative.

The study revealed a link between attitude toward afﬁrmative action and subjects’
ethics frameworks. Speciﬁcally, individuals using a more consequentialist ethics
approach were more negatively disposed toward afﬁrmative action. This supports and
clariﬁes some of the anecdotal evidence about the source of opposition to afﬁrmative
action.

Although the experimental manipulation did not have a signiﬁcant effect on

subjects’ attitudes toward afﬁrmative action it holds promise for future research. The

ﬁndings suggest that such an ethics training manipulation may impact some individuals,
depending on their previously held ethics frameworks. Based on the study’s ﬁndings,
changes in individuals’ ethics frameworks can be expected to impact their attitudes
toward afﬁrmative action.

This study also revealed some interesting ﬁndings about the relationship between
the consequentialism and nonconsequentialism ethical frameworks. These two ethical
frameworks were found to be best represented as opposite ends of a continuum. The
existing measures of the constructs were rejected in favor of a single, unidimensional
consequentialism-nonconsequentialism measure.

Finally, the study showed a clear relationship between attitudes toward afﬁrmative

action and the Religious Conviction and Categorical Imperative ethical frameworks.

ACKNOWLEDGMENTS

. . .I have learned, in whatsoever state I am, therewith to be content. I know

both how to be abased, and I know how to abound: every where and in all

things I am instructed both to be full and to be hungry, both to abound and

to suffer need. I can do all things through Christ which Strengtheneth me

(Philippians 4:11-13, KJV).

This passage was one that I put up on the wall of my ofﬁce during one of the more
trying periods of my career as a doctoral student. I suppose this dissertation marks the
tangible proof of its truth. However, since I know the Author of the passage, I didn’t
really need proof. His publications need no references, no human subjects approval and
His conclusions are always correct, even when the editors disagree. I am extremely and
eternally thankful to the Lord Jesus for making it all possible.

I wish to acknowledge my chair, John Hollenbeck, who has been a role model,
teacher, adviser and supporter for over a decade. Thanks for your patience. Thanks to my
committee members — Georgia Chao, Michael Lambert, and Ray Noe for great ideas,
constructive feedback and more patience. Thanks to the other faculty members who in
one way or the other propped up my faltering sense of self conﬁdence. Thanks also to my
fellow doctoral students, especially Kok Yee Ng — you were an answer to prayer.

I would like to thank my friends, family and church family for prayerful support
and encouragement — you had more conﬁdence in me than I had in myself. Special thanks
to Rev. Melvin T. Jones for the unwavering support and to Dr. Ernest Fubara for setting
the example.

I thank my three sons Abiye’, Priye’ and Adonye’ for giving up time with Daddy.

Finally, thanks to LaVeme, for making my life better. I love you so much.

iv

TABLE OF CONTENTS

List of Tables ....................................................................................... viii

List of Figures ....................................................................................... ix

INTRODUCTION ................................................................................... 1

CHAPTER 1

ALLEVIATING THE UNINT ENDED CONSEQUENCES OF AFFIRMATIVE

ACTION ............................................................................................... 3
Unintended consequences of afﬁrmative action ......................................... 3
Alleviating the unintended consequences — focus on attitudes ........................ 6
Deﬁning key concepts ..................................................................... 10
History of afﬁrmative action ............................................................ .13
Deﬁnition of afﬁrmative action .......................................................... 17
Afﬁrmative action target groups ......................................................... 20
Operationalizing afﬁrmative action - afﬁrmative action programs ................. 20
Attitudes toward afﬁrmative action ..................................................... 25
Alleviating negative attitudes - using alternative ethical frameworks .............. 28
Alleviating negative attitudes — the role of justice and fairness norms ............. 39
Summary and conceptual model ......................................................... 59

CHAPTER 2

METHOD ............................................................................................ 62
Sample ....................................................................................... 62
Power analysis .............................................................................. 62
Study design ................................................................................. 62

Pretest ....................................................................................... 64

Attitudes toward afﬁrmative action ...................................................... 64
Attitudinal and behavioral outcomes associated with afﬁrmative action ........... 65
Ethical frameworks ........................................................................ 65
Justice frameworks ........................................................................ 67
Training manipulations .................................................................... 68
Data analysis ................................................................................ 71

CHAPTER 3

RESULTS ........................................................................................... 73
Descriptive statistics ....................................................................... 73
Factor analysis of dependent variables .................................................. 73
Hypotheses tests ............................................................................ 87
Supplemental analyses ..................................................................... 97

CHAPTER 4

DISCUSSION ...................................................................................... 105

Discussion of the effects of ethics variables on
attitudes toward afﬁrmative action ..................................................... 105

Discussion of the effects of justice variables on

attitudes toward afﬁrmative action ..................................................... 107

Reexamining ethics measurement ...................................................... 108

Limitations and future research ......................................................... 109
APPENDIX A: Consent Form ................................................................... 112
APPENDD( B: Measurement Instruments ..................................................... 115
APPENDIX C: Scale Sources, Reliability and Validity ..................................... 135

vi

APPENDIX D: Training Script ................................................................. 139

vii

LIST OF TABLES

Table 1 - Summary of Hypotheses ................................................................ 60
Table 2 - Study Conditions ........................................................................ 63
Table 3 - Variable List .............................................................................. 72
Table 4a - Pretest Descriptives ..................................................................... 75
Table 4b - Posttest Descriptives ................................................................... 77
Table 5a - Pretest Correlations .................................................................... 79

Table 5b — Posttest Correlations .................................................................. 82
Table 6 - Scale Reliabilities (Internal Consistency Estimates) ................................ 85
Table 7 - Rotated Factor Matrix — Dependent Variables ....................................... 86
Table 8a - Hypotheses Tests (Zero-Order Intercorrelations) .................................. 88
Table 8b - Hypotheses Tests (Zero-Order Intercorrelations)

with Revised Dependent Variables ............................................................... 89
Table 9 - Moderated Hypotheses Tests — Hypothesis 1 ........................................ 90
Table 10 - Moderated Hypotheses Tests - Hypothesis 5 ....................................... 92
Table 11 - Moderated Hypotheses Tests — Hypothesis 6 ....................................... 93
Table 12 - Effect of Video Manipulation (Zero-Order Intercorrelations) .................... 96
Table 13 - Factor Matrix — Ethics Framework Items ........................................... 99
Table 14 - Factor Matrix -— New Consequentialism Scale Items ............................ 101

Table 15 - New Consequentialism Scale ....................................................... 102

Table 16 — Revised Hypothesis Test (Zero-Order Intercorrelations)
with Unidimensional Consequentialism Scale ................................................. 103

viii

LIST OF FIGURES

Figure 1 - Model of Attitudes toward Afﬁrmative Action ..................................... 61
Figure 2 - Study Design ............................................................................ 63
Figure 3 -— Basic Experimental Design ........................................................... 7O

INTRODUCTION

Simply deﬁned, afﬁrmative action refers to the general concept of proactive
efforts to achieve equal employment opportunity. The general concept is implemented in
organizations through a variety of speciﬁc afﬁrmative action programs. These and other
identity-conscious1 organizational initiatives are arguably, the most challenging,
controversial, and conﬂict-ridden facets of work in organizations (Chang, 1996; Opotow,
1996). They provide difﬁcult challenges for a variety of organizational stakeholders
including employees, managers, researchers, regulators, and even society at large. In
spite of their intentions and perceived beneﬁts, afﬁrmative action efforts attract ﬁerce
opposition even from those who are expected to beneﬁt from them.

As with any other organizational initiative, afﬁrmative action programs (AAPS)
have their shortcomings. However, unlike other organizational efforts (such as training
and development, beneﬁts administration, or wellness and safety campaigns) afﬁrmative
action and other identity-conscious programs are associated with a signiﬁcant number of
unintended, often negative consequences. These consequences tend to generate a level of
conflict and controversy not often associated with other human resource management
initiatives (Chang, 1996; Opotow, 1996; Tierney, 1997).

The current study serves as the foundation for a program of research on alleviating
these unintended consequences. AS delineated below, a number of unintended
consequences have been linked to afﬁrmative action, either through empirical study or

conceptual propositions. The consequences are said to impact multiple parties, including

 

IThe term “identity-conscious” refers to activities that explicitly consider the demographic characteristics of
organizational members (Konrad & Linnehan, 1995). The term is described in more detail later in this

chapter.

 

the intended beneﬁciaries of afﬁrmative action efforts, those who perceive themselves to
have been denied opportunities as a result of these initiatives, and more neutral observers.
The consequences also span a range of outcomes including attitudes, behavior, and
results.

This study addresses the attitudinal bases that undergird other aspects of
individuals’ reactions to afﬁrmative action and afﬁrmative action programs. The
underlying assumption is that individuals’ attitudes toward afﬁrmative action are rooted
in conceptions of whether or not afﬁrmative action is normatively right or fair. I attempt
to change individuals’ attitudes toward afﬁrmative action by presenting them with an
alternative framework for distinguishing right from wrong. I expect that this alternative
framework will foster more positive attitudinal reactions to afﬁrmative action (see
Figures 1 & 3). This change in attitude serves as a foundation for more extensive changes
in reactions and behavior. Thus, this study serves as a foundation for future research and
potentially, organizational policies directed toward reducing the negative outcomes

associated with afﬁrmative actionz.

 

2 This study does not attempt to make a normative statement regarding the use or non-use of afﬁrmative
action or afﬁrmative action programs. Rather, it is a theory-based attempt to address a very real concern
facing contemporary organizations. In spite of some vocal opposition, afﬁrmative action and related
identity conscious efforts are expected to remain a part of the American organizational landscape for some
time to come (Kravitz & Platania, 1993).

Chapter 1
ALLEVIATING THE UNINTENDED CONSEQUENCES OF
AFFIRMATIVE ACTION
Unintended consequences of afﬁrmative action

In her 1994 review, Madeline Heilman provides an excellent summary of both
conceptual and empirical work on the unintended consequences of afﬁrmative action
programs. These unintended consequences are predominantly seen as negative, or
counterproductive, and often represent a signiﬁcant challenge for managers. They include
reduced self assessment of competence among afﬁrmative action beneﬁciaries;
counterproductive work behavior; adverse reactions to others in the work setting; adverse
reactions to the work setting itself; a perception that selection processes are unfair; stress;
negative affect; stigmas of incompetence (ascribed to afﬁrmative action beneﬁciaries by
their observers); negative characterizations of afﬁrmative action beneﬁciaries (non-
competence related characterizations); negative perceptions of the jobs in which
afﬁrmative action beneﬁciaries work; and perceptions of injustice on the part of those not
selected.

Although Heilman’s work focuses on women, afﬁrmative action programs tend to
be quite similar, regardless of whether they are directed toward women, or members of
various racial/ethnic minority groups (referred to from here on as minorities). However,
programs directed toward the handicapped tend to take on a different form, and thus to
evoke unique attitudinal and behavioral responses, even though they are also designed to
promote equality of employment for a group of persons. Most equal employment

Opportunity efforts directed toward the handicapped are based on the Americans with

Disabilities Act (ADA) of 1992. The key difference between these and other equal
employment opportunity programs is that employers may have legitimate, objective
concerns about the abilities of individuals with certain disabilities to perform Speciﬁc
jobs (Klimoski & Palmer, 1994). Thus, one aspect of EEO initiatives for the disabled is
that they often involve the elimination of barriers (physical or otherwise), modiﬁcations
to jobs, and/or the provision of speciﬁc accommodations for individual persons, rather
than a class of persons per se. AS mentioned previously, programs for the handicapped do
not generally evoke the same kind of attitudinal reactions as those directed toward women
and minorities. Likewise, the reactions to veterans’ employment preferences rarely
produce the adverse unintended consequences outlined earlier. These distinctions are
signiﬁcant enough that they suggest distinct streams of inquiry. Afﬁrmative action efforts
directed toward veterans and the handicapped are outside of the focus of this
investigation. No further mention will be made of them.

In spite of the similarity between afﬁrmative action efforts directed toward
women and those targeting minority group members, the affective and behavioral
reactions to programs directed toward these different groups are not identical. Some
studies have uncovered Similar reactions to programs directed at these different groups
(Heilman, Block & Lucas, 1992 [Study 2]), others have stressed differences. Fletcher and
Chalmers (1991), for example focus on the effects of contextual factors in determining
attitudes toward afﬁrmative action. Their study, conducted in Canada, found that support
for afﬁrmative action programs was inﬂuenced by the beneﬁciary group in question.
Programs directed toward women received greater support than those directed toward

French Canadians (an ethnic minority) and Native Canadians (a racial and ethnic

minority). This result held true for both men and women. Note, however, that French
Canadians demonstrated equal support for programs directed toward the Francophone
minority and those directed toward women. Overall, their study provides further
justiﬁcation for research designed to focus on afﬁrmative action efforts directed toward
minority group members. These and other studies (Parker, Baltes, & Christiansen, 1997)
suggest that there is a need to further investigate the underlying processes associated with
programs directed toward different beneﬁciary groups. In fact, some scholars have
explicitly recommended that more research attention be directed toward minority group
members (Kravitz, 1995; Kravitz & Platania, 1993).

In spite of some very vocal opposition, including that of many prominent political
ﬁgures, afﬁrmative action programs are expected to exist in one form or another in
American organizations for some time to come. Both the history of the American
workplace, and the contemporary dynamics of the workforce suggest that afﬁrmative
action programs have not outlived their usefulness (Johnston & Packer, 1987; Kravitz,
1995; Kravitz & Platania, 1993). Afﬁrmative Action and related initiatives also have
potential application in a number of countries outside of the United States where
multiculturalism and racial reconciliation are important issues. The breadth and impact of
these efforts indicates that their study is an important problem both for scholars and
practicing managers -- a problem that merits considerable and careful investigation.
Given that these programs are expected to exist for some time to come, it is logical for the
research agenda to include the investigation of ways to make them more palatable to

those affected by them. This concern is the primary justiﬁcation for the current study.

Alleviating the unintended consequences -- focus on attitudes

As introduced above, the emphasis of the current study is on impacting
individuals’ attitudes toward afﬁrmative action. These attitudinal reactions are expected
to underlie and predict the behavioral consequences discussed above. The study also
focuses on afﬁrmative action as a concept, rather than Speciﬁc afﬁrmative action
programs. A number of existing studies have already examined individuals’ differential
reactions to various speciﬁc afﬁrmative action programs (Heilman, Battle, Keller, & Lee,
1998, Kravitz, 1995; Summers, 1995; Taylor-Carter, Doverspike, & Cook, 1995).
However, other investigators have shown that individuals tend to enact their own pictures
of what afﬁrmative action entails. AS such, their reactions are not necessarily directed at
the Speciﬁc, actual programs in operation in their organizations but at their personal
conceptions of what the policies entail. Often, these conceptions are quite erroneous,
including in them program elements that are impracticable, or illegal. In general, there is
signiﬁcant confusion about what afﬁrmative action entails (Tomasson, Crosby, &
Herzberger, 1996).

The confusion among employees, is often compounded by the fact that employers
may do a poor job of implementing afﬁrmative action programs. Thus, employees may
have a clear understanding of their organizations’ afﬁrmative action programs and still be
negatively disposed toward them. For example, in spite of the fact that they are, in most
cases, illegal, many organizations still use some form of quota in afﬁrmative action
decision making. Employees observing such a practice may assume that this is a normal
component of afﬁrmative action and thus extend their negative attitudes toward their

organizations’ speciﬁc programs to the general concept of afﬁrmative action.

The limited emphasis of this study is also valuable in that it may provide
theoretical support for the investigation of other related organizational programs. In
some instances, reactions to afﬁrmative action may generalize to other identity-conscious
organizational efforts, such as equal employment opportunity programs, or management
of diversity.

Extant research has attributed negative responses to afﬁrmative action to a
number of psychological explanations, many of which are rooted in deep-seated
individual values, including modern racism, belief in the dominant ideology of
opportunity, and simple self—interest (Kluegel & Smith, 1983); fairness judgments
(Nacoste, 1990); and ethics (Blackstone & Heslep, 1977). The current study focuses on
two possible causal explanations for these negative reactions to afﬁrmative action, and in
turn, attempts to use these explanations as approaches to altering individuals’ reactions.

The ﬁrst approach to addressing individuals’ responses to afﬁrmative action is
rooted in the domain of ethics. A rich tradition of ethics research has produced a
typology of ethical frameworks or theories with which individuals make determinations
regarding right and wrong. The second analytical perspective, is drawn from the
organizational justice literature. Here, individuals’ justice frameworks or theories-in-use
determine how they determine what is fair or unfair. Each analytical approach is used as
a predictor of attitudes toward afﬁrmative action. Additionally the ethics frameworks
serve as the foundation for an attempt to alter individuals’ negative affective reactions.
Note that these analytical tools do not necessarily determine the normative outcomes of
individuals’ judgments. They do however affect the processes through which individuals

arrive at normative conclusions.

One of the initial challenges facing this study is that it is an attempt to manipulate
perspectives and attitudes, which are often deeply ingrained and difﬁcult to alter. Both
the independent variables (ethics and justice frameworks) and the dependent variable
(attitude toward afﬁrmative action) are deep-seated, value-laden characteristics and are
resistant to change. However, research on both ethics and justice indicates that within-
person differences may be observed for the various frameworks/perspectives. In other
words, there are Situational and temporal considerations that affect individuals’ dominant
frameworks. There are also indications that these frameworks respond to training and
educational interventions. For example Iordanis Kavathatzopoulos (1993) conducted a
study examining the malleability of individuals’ ethical frameworks. In this study, a
training manipulation was used to attempt to move subjects toward the autonomous moral
function, that is, toward addressing moral problems using concrete, situation-speciﬁc
information, rather than abstract moral principles.3 The dependent variable was the moral
function or ethical phase revealed through the explanations given by individuals for their
solutions to business ethics problems. The educational manipulation had a robust effect
on individuals’ moral functions, moving most subjects from one phase to another.

Similarly, the direct malleability of afﬁrmative action attitudes has also been
demonstrated: Taylor-Carter, Doverspike, and Alexander (1995) assessed the effect of
pro- versus anti-afﬁrmative action organizational communications on individuals’
fairness perceptions. In this context, fairness perceptions were taken to be a form of
attitude, with the potential to predict behavioral responses. This study, which focused on

the reactions of women toward programs designed to beneﬁt women, demonstrated a

 

Note that In some theoretical traditions, this form of moral reasoning is seen as more advanced and more

signiﬁcant impact of organizational communication on fairness perceptions. Fairness
perceptions were also signiﬁcantly influenced by individuals’ cognitive responses: this
refers to the number of positive, negative, or neutral afﬁrmative action-related thoughts
the individuals generated and recalled, immediately following their exposure to the
organizational communication.

Change in attitudes toward afﬁrmative action has also been demonstrated by
Fletcher and Chalmers (1991). They directly manipulated individuals’ attitudes towards
afﬁrmative action, asking them to reconsider their previously held Views. Unfortunately,
this approach has the potential to introduce a very powerful priming effect and is unlikely
to induce, and in tum assess the potential for long-term change. Further it was conducted
in Canada, where afﬁrmative action is an accepted and institutionalized means of
redressing historic discrimination. In that context, prevention of future discrimination
and remedying of past discrimination are seen as inextricably intertwined. Thus, Fletcher
and Chalmers’ study provides further justiﬁcation for the indirect approach of the current
paper, particularly for the use of deeply held frameworks as levers for inducing change.
In their words: “there is room for persuasion, however, and this lies not only at the level
of policy preference but also at the level of deeply held values” (Fletcher & Chalmers,
1991 :91).

In contrast, other research indicates that attitudes about race in general are less
malleable. In a longitudinal study, Kinder and Rhodebeck (1982) examined changes in
Whites’ support for racial equality (operationalized in terms of their acceptance of

changes in the racial status quo, racial integration of neighborhoods and busing).

 

effective. However, such considerations are outside of the domain of this study.

Speciﬁcally, they examined the effects of personal economic conditions, exposure to
crime and violence, and the perception that African-Americans threaten their private
interests. The most powerful effect was for exposure to crime and violence. However,
none of these effects was of practical signiﬁcance. The most striking result of the study
was that racial opinions appear to be quite consistent over time, regardless of experience.
These ﬁndings also lend support to the approach of this study, that is focusing on more
deeply seated individual differences as predictors of responses to afﬁrmative action.

Again, my intention in this study is to capitalize on within-person differences in
ethics/justice frameworks, and related within-person differences in attitudes toward
afﬁrmative action. Admittedly, these are difﬁcult manipulations to accomplish; however,
in such a setting even small effects are practically important (Prentice & Miller, 1992).
Deﬁning key concepts

At this stage, it will be meaningful to provide a brief explication of some of the
central concepts used in this study:

Afﬁrmative action and afﬁrmative action programs. Afﬁrmative action is one of
the central concepts of this study and as such will be deﬁned in much greater detail in the
next chapter. However, one legal deﬁnition of afﬁrmative action is contained in the
Uniform Guidelines on Employee Selection Procedures (U.G.E.S.P., 1978). The
Guidelines deﬁne afﬁrmative action as “those actions appropriate to overcome the effects
of past or present practices, policies, or other barriers to equal employment Opportunity
(29 CFR part 1608 section 1C) .”

Afﬁrmative action was originally conceived for use by Federal contractors, who

were instructed to make good faith efforts toward achieving equal employment

10

opportunity for women and members of ethnic minority groups. As mentioned earlier,
the deﬁnition, along with the history and scope of afﬁrmative action will be discussed in
greater detail in the next chapter. However, it is valuable at this point to make the
distinction between afﬁrmative action and afﬁrmative action proggams: afﬁrmative action
refers to the general concept of proactive efforts to achieve equal employment
opportunity. Afﬁrmative action programs, on the other hand, are speciﬁc organizational
initiatives designed to implement afﬁrmative action policy. Much of the existing research
on attitudes toward afﬁrmative action has focused on reactions to speciﬁc programs as
opposed to the general concept. As discussed earlier, the distinction between these two
approaches is consequential for both research and practice.

Identity-conscious organizational proggams. Also referred to as identity conscious
management programs refers to organizational initiatives that explicitly consider the
demographic status of organization members (Konrad & Linnehan, 1995). The most
commonly used demographic characteristics are race, gender, and disability. Programs
that do not explicitly consider these characteristics are referred to as identity-blind.
Although this study is primarily focused on afﬁrmative action, where appropriate I will
also investigate reactions to other identity-conscious organizational efforts.

mgam beneﬁciaries/non-beneﬁciarﬂ. The term “program beneﬁciary” or
simply “beneﬁciary,” refers to the organization members whom the organizational
programs are designed to assist. Identity-conscious programs are intended to promote the
well being of speciﬁc, protected demographic sub-groups within the organization,
primarily women, and members of ethnic minority (non-Caucasian) groups.

Similarly, “non-beneﬁciaries” of identityconscious programs are those who are

11

not members of the programs’ target groups. However, the term non-beneﬁciary may be
somewhat misleading in that many identity-conscious programs are expected to promote
the beneﬁt of all organization members, as well as that of the organization itself

(U .G.E.S.P., 1978). For example, as part of an afﬁrmative action program, an
organization may require that all job opportunities be advertised through public postings.
Thus, the true beneﬁts of these programs are not limited to their immediate, or direct
beneﬁciaries.

It should be noted again that these rather simplistic descriptions of the
beneﬁciaries and non-beneﬁciaries of afﬁrmative action focus on the intended outcomes
of these programs. However, in later sections of this study, I distinguish between the
intentions of afﬁrmative action’s designers and the perceptions of affected individuals. A
signiﬁcant body of behavioral research has demonstrated that these perceptions and
attributions may be more powerful motivators than an initiative’s “ofﬁcial” goals and
objectives. These distinctions underscore the inappropriateness of assuming that
demographic status is inherently predictive of reactions to, or support of afﬁrmative
action (which unfortunately, is often done). For research purposes, it will be more
important to assess individuals’ p_erceptions of whether or not they stand to beneﬁt from
afﬁrmative action.

Equal employment opp_ortunity. Equal opportunity in employment means not
discriminating against a particular group because of that group’s race, sex, religion,
ethnicity, disability, or veteran status (Holloway, 1989). Equal opportunity is generally a
more reactive, passive approach to preventing discrimination than afﬁrmative action. It

focuses on and provides sanctions for individual acts of discrimination, assuming that in

12

the absence of such, organizations will function fairly. Equal opportunity in effect,
“seeks to make the organization blind to difference, whereas afﬁrmative action
accentuates difference (Tierney, 1997, p. 173).”

Equal opportunity programs prohibit intentional discrimination but they do not
require organizations to assess whether their practices unintentionally harm any groups
(Crosby & Cordova, 1996; Stout & Buffum, 1993). One important distinction between
equal opportunity and afﬁrmative action is that individuals do not have to be
discriminated against, complain about such discrimination, and subsequently prove that it
occurred before afﬁrmative action can be initiated. Afﬁrmative action provides for
mechanisms to monitor discrimination at the aggregate level, which is easier to detect and
rectify (Crosby & Cordova, 1996).

Histogy of afﬁrmative action

Afﬁrmative action is distinct among human resource management initiatives in
the degree to which it is rooted in the American historical, legal, and social context.
Thus, any discussion of afﬁrmative action is incomplete without a review of its history
and origins. This review is not meant to be a comprehensive analysis of afﬁrmative
action and related issues. It focuses on the most pivotal and relevant issues, and in so
doing provides a framework for the subsequent conceptual discussion. Similarly, the
historical discussion of afﬁrmative action must, of necessity, consider matters of
legislation and jurisprudence. In both cases, more detailed analyses of these issues are
available from other sources (e.g. Greene, 1989; Newman, 1989; Taylor, 1991; Turner,
1990).

The historical roots for afﬁrmative action and indeed, for other identity-conscious

l3

programs may be traced to the Freedmen’s Bureau Act of 1866 which provided for
economic opportunities for African-Americans, who were for the most part newly freed
slaves or the immediate descendants of slaves. This Act illustrates the nation’s
recognition of a need to provide economic parity for those who had heretofore been
systematically and legally oppressed (Jones, 1985).

A second legal and historical milestone is the Civil Rights Act of 1875. Again,
this Act was directed toward African-Americans and designed to alleviate the desperate
economic conditions of former slaves and their descendants. The Act provided that all
citizens were “entitled to the full and equal enjoyment of the accommodations,
advantages, facilities, and privileges” of public conveyances, theaters and places of public
amusement (Jones, 1985). However, this Act was declared unconstitutional through a
series of Supreme Court cases. Such political and legal opposition to equality of
opportunity characterized the American experience until 1954 when the Supreme Court
heard Brown v. Board of Education of Topeka, Shawnee County, Kansas (Turner, 1990).
Despite the fact that this case was not employment related, it served as a foundation for
future identity-conscious initiatives.

The ﬁrst usage of the phrase “afﬁrmative action” in the legal context is found in
the 1935 National Labor Relations Act (NLRA). The NLRA gave the National Labor
Relations Board (NLRB) the charge of rectifying unfair labor practices and the authority
to issue cease and desist, reinstatement, and back pay orders, and to order afﬁrmative
a_ct_i9_n_ (NLRA, 1935). Although this authority was limited to the enforcement of the
provisions of the NLRA, this statute is an example of the govemment’s willingness to

intervene in employment situations to rectify past wrongs (Turner, 1990).

14

Executive Order 8802 of 1941 is one of the ﬁrst legal statutes directed speciﬁcally
at the prevention of employment discrimination. The Order prohibited employment
discrimination (based on race, creed, color, or national origin) and mandated the use of
afﬁrmative action in employment. Although the order did not use the phrase afﬁrmative
action, it did “provide for the full equitable participation of all workers in defense
industries without discrimination (Jones, 1985).”

As indicated earlier, a critical turning point in the history of remedial identity-
conscious activity was the Brown v. Board of Education Supreme Court decision of 1954.
This decision, holding that segregation in public schooling was unconstitutional, provided
for legal relief for those affected by past discrimination (Turner, 1990).

The ﬁrst use of the phrase afﬁrmative action in the speciﬁc context of
employment discrimination is found in Executive Order 10925 (1963). This order
required government contractors to “take afﬁrmative action to ensure that
applicants....and employees....are treated....without regard to race, creed, color, or national
origin (Section 301).” This order also established the Equal Employment Opportunity
Commission (EEOC), the organization charged with oversight of employment
discrimination law. The EEOC’S regulatory power was later extended by the Equal
Employment Opportunity Act of 1972. This Act gave the commission the power to
oversee all employers and unions with over ﬁfteen members as well as all State, local, and
federal government employees. The 1972 bill also expanded the scope of employment
discrimination to encompass not only speciﬁc instances of discrimination but also
institutional patterns of exclusion (Mosley & Capaldi, 1996).

One of the most signiﬁcant legislative events in the history of afﬁrmative action

15

was the passage of Title VII of the 1964 Civil Rights Act, which regulates equal
employment opportunity. Title VII explicitly declares that it is “ an unlawful employment
practice for an employer to fail or refuse to hire, or to discharge any individual, or
otherwise to discriminate against any individual with respect to his compensation, terms,
conditions, or privileges of employment, because of such individual’s race, color,
religion, sex, or national origin (Section 703 (a) (1)).” Title VII also provides for the use
of afﬁrmative action, including reinstatement and hiring, as a remedy for unlawful
employment discrimination (Section 706 (g)).

Another legislative milestone is the 1965 Executive Order 11246, which
speciﬁcally regulates the employment practices of government contractors. It requires
them to avoid discrimination against members of protected groups and to “take
afﬁrmative action to ensure that applicants are employed and treated during their
employment without regard to their protected status (Section 202 (1)).”

The Civil Rights Act of 1991 is another pivotal piece of legislative history
affecting equal employment opportunity and afﬁrmative action. This Act, which
amended the 1964 Civil Rights Act, served to codify and clarify issues uncovered in
various Supreme Court decisions, particularly Wards Cove Packing v. Atonio (490 US.
642, 1989). With speciﬁc reference to afﬁrmative action, the 1991 Act prohibited the use
of test score adjustment as a means of implementing afﬁrmative action.

Because of the peculiar history of, and systematic discrimination perpetrated
against African-Americans, early afﬁrmative action initiatives were primarily directed
toward African—Americans. They were later extended to other racial minorities, and to

women. However, a number of afﬁrmative action efforts exist which are directed toward

l6

the advancement of other groups in the workplace. The Age Discrimination in
Employment Act of 1967 forbids discrimination based on age. The Vietnam Veterans
Readjustment Assistance Act of 1974 requires employers to take afﬁrmative action to
employ and advance qualiﬁed Vietnam-era veterans. The Vocational Rehabilitation Act
of 1973 requires government contractors to take afﬁrmative action for qualiﬁed workers
with disabilities, and the Americans with Disabilities Act (ADA) of 1990 forbids
discrimination on the basis of disability. Although afﬁrmative action efforts targeting
individuals with disabilities often involve the development of Speciﬁc afﬁrmative action
programs, these AAPS tend to focus on employment processes and good faith efforts, not
on documentation and utilization analyses. They are less controversial, and have received
less empirical attention, than AAPS directed at women and ethnic minorities (Kravitz et.
al, 1996).
Deﬁnition of afﬁrmativeiaction

As indicated in the preceding section, early afﬁrmative action programs applied to
organizations that had contracts with the federal government. As a condition of working
with the government, contractors were required to “take afﬁrmative action to ensure that
applicants are employed, and that employees are treated during employment, without
regard to their race, color, religion, sex, or national origin (Executive Order 11246).” The
Order further speciﬁed that afﬁrmative action is widely applicable to recruitment,
training, selection, promotion, and compensation decisions (Holloway, 1989).

Although the Order requires federal contractors to engage in afﬁrmative action,
the deﬁnition of afﬁrmative action, provided in the preceding section, is found in the

OFCCP manual. A similar deﬁnition is contained in Title 41 of Chapter 60 of the Code

17

of Federal Regulations. This document speciﬁcally deﬁnes an afﬁrmative action pram
as:

“....a set of speciﬁc and result—oriented procedures to which a contractor

commits itself to apply every good faith effort. The objective of those

procedures plus such efforts is equal employment opportunity (subsection

2.10).”

Holloway (1989) provides a somewhat different perspective on afﬁrmative action.
She indicates that afﬁrmative action promotes equality of opportunity in that it affects the
subjectivity inherent in the judgments of organizational decision-makers. These
judgments are often inﬂuenced by personal experiences and prejudices concerning the
abilities of members of target groups. Afﬁrmative action, according to Holloway, is
designed to alert people to the previously unobserved abilities of members of targeted
groups.

Kravitz and Platania (1993) offer a more intuitive deﬁnition of afﬁrmative action.
They deﬁne afﬁrmative action as actions taken to increase the numbers of an
underrepresented demographic group in an organization. The goal of these programs is
speciﬁcally to redress past discrimination. Although afﬁrmative action is generically
deﬁned in terms of equality of opportunity and compensating for past discrimination, the
majority of employment discrimination in the United States has affected speciﬁc,
identiﬁable groups. It is these groups that are currently considered the “target groups” for
afﬁrmative action efforts.

Crosby and Cordova (1996) contrast the classical deﬁnition of afﬁrmative action

with its contemporary meaning. They trace the classical deﬁnition of afﬁrmative action

to the 1965 Executive Order 11246 and describe it as occurring “whenever people go out

of their way (take positive action) to increase the likelihood of true equality for
individuals of differing categories. Whenever an organization expends energy to make
sure that women and men, people of color and White people, or disabled and fully-abled
workers have the same chances to be hired, retained, or promoted, then the organization
has a policy of afﬁrmative action in employment (pp. 34-35).” This classical approach to
afﬁrmative action is operationalized by federal contractors in the requirement that they
track the rate of utilization of members of target groups in their organizations, and exert
efforts to ensure that the rate of utilization matches the availability of persons in the
relevant labor market. In instances where members of the targeted groups are
underutilized, the contractors must set goals for increasing the proportion of target group
members in their workforces, and develop documentable plans for reaching these goals.
These plans might include special recruitment efforts, closer scrutiny of organizational
practices to exclude those that might result in adverse impact, or the temporary use of set
asides (Crosby & Cordova, 1996).

Classical afﬁrmative action, as introduced by Executive Order 11246 simply
requires organizations to ensure that their utilization of protected groups matches the
availability of such groups in the relevant labor market. In the absence of such a match,
the groups are required to establish goals and timetables for rectifying the discrepancy
(Tomasson et a1, 1996).

In more recent times, afﬁrmative action has taken on a more controversial
meaning. It has come to mean quotas, unjustiﬁed set—asides or preferential treatment,
including the hiring and promotion of unqualiﬁed individuals (Crosby & Cordova, 1996;

Heilman, McCullough, & Gilbert, 1996; Kravitz & Platania, 1993). However, such

19

practices are, for the most part, illegal, and studies reveal that they are not practiced with
any signiﬁcant frequency (Crosby & Cordova, 1996). The pervasiveness of this
contemporary view of afﬁrmative action further undergirds the value of assessing the
attitudinal responses of individuals to the general concept of afﬁrmative action. Again,
this approach allows one to assess individuals’ reactions to their personal conceptions of
afﬁrmative action without attempting to reconcile their differential understandings of
what it entails.

Note that the term “afﬁrmative action” may also be accurately used to refer to set-
aside programs in education and contracting, in which a certain number of positions, or
percentage of the sub-contracting budget respectively is designated for the use of women,
or minorities. Although such programs are conceptually similar to those governing
employment, they are outside of the scope of this study.

Afﬁrmative mm target groups

Early forms of contemporary afﬁrmative action directed redemptive efforts toward
two demographic groups: women and members of ethnic minority groups (Holloway,
1989). Generally, recognized ethnic minority groups are American Indian or Alaskan
Native; Asian or Paciﬁc Islander; African-American, not of Hispanic Origin; and
Hispanic. Later efforts expanded these protections, making them available to the

handicapped and veterans.

Omrationalizing afﬁrmative action -- afﬁrmative action proggams

 

Although the concept of afﬁrmative action tends to arouse controversy, the
practice of afﬁrmative action is even more controversial. AS with other aspects of the

topic, the execution of afﬁrmative action is rooted in a complex network of history, legal

20

statutes, and case law. As such, no attempt is made here to provide a comprehensive
review of afﬁrmative action program components/practices.

A classical means of implementing afﬁrmative action may be traced back to
Executive Order 11246 (1965). This order, enforced by the Ofﬁce of Federal Contract
Compliance (OFCCP) requires federal contractors and subcontractors to monitor the
utilization of members of target groups within their workforces. Covered employers are
required to track the extent to which their workforces mirror the composition of the
relevant pools of potential employees. When the employers’ workforces fall short (for
example, if the number of African-American engineers in the organization is smaller than
the number of qualiﬁed African-American engineers in the labor force) the employer is
required to draw up plans for rectifying the imbalance. The OFCCP requires contractors
to engage in documented good faith efforts to meet their targets (see Crosby & Cordova,
1996 for a more detailed discussion).

One of the most divisive practices associated with afﬁrmative action programs is
the use of quotas. The issue of quotas may be traced to the 1971 Griggs v. Duke Power
Company case, which introduced the use of statistical analyses into the investigation of
employment discrimination‘. Based on the Griggs standard, employers were prohibited
from engaging in neutral appearing business practices, if those practices “operate[d] to
‘freeze’ the status quo of prior discriminatory employment practices (Griggs v. Duke
Power, 1971).” The Griggs standard is quantiﬁed in the Uniform Guidelines on

Employee Selection Procedures (EEOC, 1978). These guidelines introduced the four-

 

’ Although this was not the primary contribution of the Griggs case, the case did focus attention on
statistical analyses in that it shifted focus to evidence of discrimination in the absence of discriminatory
intent.

21

ﬁfths rule, requiring that the selection ratio of protected groups be at least 80% of that of
the demographic group with the largest selection ratio.

Unfortunately, one of the most expedient ways for organizations to satisfy the
Griggs standard is to impose strict numerical quotas, and to use preferential treatment
practices to fulﬁll these quotas (Crosby and Cordova, 1996). Following the Griggs case,
and in anticipation of the proliferation of quotas, the US. Supreme Court developed
standards, requiring plaintiffs in employment discrimination cases to provide more
compelling a priori evidence of discrimination. In effect, the court required the
identiﬁcation of speciﬁc unlawful business practices, rather than just the disparate
statistical evidence which may have resulted from these practices (Watson v. Fort Worth
Bank and Trust, 1988). This standard was later codiﬁed in the Civil Rights Act of 1991,
which implicitly outlawed the use of quotas. Other government documents echo the
unacceptability of quotas. The US. Department of Labor for example, has expressly
stated that “afﬁrmative action is not preferential treatment. It does M mean that
unqualiﬁed persons should be hired or promoted over other people. What afﬁrmative
action does mean is that positive steps must be taken to ensure equal opportunity for
traditionally disadvantaged groups (US. Department of Labor, 1995, p. 1 , underlining in
original).

Except under the most extreme conditions, Supreme Court decisions have
forbidden strong preferential treatment (hiring unqualiﬁed women and minorities rather
than qualiﬁed men and non-minorities) and strict quotas (Bennett-Alexander, 1990). It is
permissible, however, for organizations to assign some unspeciﬁed weight to gender or

race when making employment decisions (Guttman, 1993).

22

In addition to the legal prohibitions against quotas, public opinion suggests that
they are the least appealing identity—conscious employment mechanism. The popular
disapproval is compounded by the fact that quotas, of necessity, require organizations to
engage in preferential treatment of some individuals, based on their group membership.
A 1995 Gallup Poll indicates that these practices are disapproved of by a majority of
Americans (63% oppose quotas; 84% oppose preferential treatment; Roberts, 1995).

Another way in which afﬁrmative action has been operationalized in the
American workplace is through various forms of pre-employment test score adjustment.
Typical among these techniques are race-norming (more broadly referred to as within-
group norming) and banding. Within-group norming is a practice in which employment
test scores are converted from raw or overall percentile scores to percentile scores based
on applicants’ sub-group membership (Sackett & Wilk, 1994). Commonly these scores
are then reported as percentile scores without any mention of this adjustment. Thus, job
candidates with very different raw test scores, may receive Similar reported percentile
scores, if they are members of different sub-groups (usually, different races). A similar
approach to within-group norming is simply to use different cutoff scores for different
groups, or to add bonus points to the scores of members of a group (a technique which is
often used to give preferences to veterans, Sackett & Wilk, 1994). It should be noted that
these forms of score adjustment were explicitly outlawed by the Civil Rights Act of 1991
(Section 106).

Banding is a conceptually distinct form of score adjustment in which individuals
within a speciﬁc score range or band are regarded as having equivalent scores (Sackett &

Wilk, 1994). Band width may be arbitrarily selected, or may be based on the properties

23

of the test (a band may be equal to the standard error of the test, or a multiple thereof,
depending on the desired signiﬁcance level). In the context of afﬁrmative action, banding
may be used to equate individuals’ test scores before factoring group membership, and/or
other factors into the employment decision. A hybrid score adjustment approach is the
sliding band, in which the bandwidth remains constant, but it’s upper boundary is
continually adjusted downward (Cascio, Outtz, Zedeck, & Goldstein, 1991). After
individuals have been selected from the top band, the band is moved downward to allow
for more lower scoring individuals (usually minority group members) to be compared
with higher scoring individuals. This approach has been severely criticized for violating
the assumptions of banding (that is, that scores falling within the same band are
equivalent to one another; see Sackett & Wilk, 1994; Schmidt, 1991). In many cases,
depending on the performance of a speciﬁc sample, banding and sliding band approaches
may produce outcomes equivalent to those produced by the less sophisticated means of
score adjustment described above. However, banding approaches appear to be more
acceptable to the judicial system considering the provisions of the Civil Rights Act of
1991 (Sackett & Wilk, 1994). In general, score adjustments appear to be more acceptable
in the context of more aggressive court ordered afﬁrmative action plans in which group
preferences are legally acceptable.

It is ironic that some of the more expedient and easily developed practices
(quotas, preferential treatment, and score adjustments described above) are both illegal,
and generally unacceptable. However, there are afﬁrmative action efforts which,
although less directly tied to results, are intended to bring about equality of opportunity.

The Ofﬁce of Federal Contract Compliance, for example advises employers to engage in

24

“special efforts in outreach, recruitment, training, and other areas [to] help members of
protected groups compete for jobs and promotions on equal footing with other applicants
and employees (OFCCP, 1979; p. 7).”

Based on recent Supreme Court rulings, the US. government appears to have
shifted back toward a more judicious use of race as a factor in employment-related
decision making. Two cases serve as landmarks: In City of Richmond v. Croson (1989)
the court required that the city government employ a standard of strict scrutiny in
incorporating race into decisions on awarding contracts. Speciﬁcally, it required the City
to demonstrate a direct causal link between the paucity of ethnic minority construction
contractors, and previous policies of racial discrimination. Adarand Constructors v. Pena
(1995) resulted in similar outcomes, applied to the Federal government.

Although both of these cases apply to government contracting, which is distinct
from the focus of the current study, they nevertheless have served as defacto precedents
for the manner in which the use of race will be acceptable in contemporary employment-
related organizational decision-making.

The preceding discussion serves as an overview of the practice of afﬁrmative
action. However, the focus of this Study is on deeper, psychological issues associated
with afﬁrmative action. Subsequent sections reﬂect this emphasis.

Attitudes toward afﬁrmative action

As stated in the introduction section, the current study was primarily motivated by
an interest in alleviating the unintended, negative, behavioral consequences associated
with afﬁrmative action. However, it was necessary to begin with the conceptual

precursor to these behavioral consequences, that is, the attitudinal responses to

25

afﬁrmative action. It is clear from a variety of writings, both popular and scholarly, that
Americans, especially White Americans, tend to have negative reactions to, and attitudes
toward afﬁrmative action (Dovidio, Mann, & Gaertner, 1989; Kluegel & Smith, 1983).
Individuals respond negatively to these programs, largely because they View them as
being “unfair” (Chang, 1996; Opotow, 1996). A number of social science scholars have
asserted that these fairness judgments are rooted in underlying personal predispositions
such as aversive racism (Dovidio, Mann, & Gaertner, 1989); economic self-interest, racial
affect, and stratiﬁcation beliefs (Kluegel & Smith, 1983).

One of the more extensive models examining attitudes toward afﬁrmative action
was developed by David Kravitz (1995). His model focuses on attitudes toward Speciﬁc
afﬁrmative action programs, which are attributed to both structural and individual
difference factors. Structural factors include the judged fairness of the programs, ratings
of threat to personal interest, and collective self interest (the potential harm the programs
may cause to members of one’s group). Each of these factors is hypothesized to be
affected by the weighting assigned to group membership in the afﬁrmative action
program. Kravitz’s individual difference factors included racism, self-interest in general,
and belief in the dominant ideology, a notion that everyone has the opportunity to succeed
and that thus, failure is a result of individual shortcomings (Kluegel & Smith, 1986).
According to the model, attitudes toward speciﬁc afﬁrmative action programs are
moderated by attitudes toward afﬁrmative action in general.

Stout and Buffum (1993) operationalize attitudes toward afﬁrmative action in
terms of support for, or commitment to afﬁrmative action principles. Their study of

Texas social workers found that commitment to afﬁrmative action was predicted by

26

positive and negative experiences with it. Contrary to Kravitz & Platania (1993) they
found no relationship between knowledge of afﬁrmative action and commitment to it.

A few studies have examined the malleability of afﬁrmative action attitudes. Bell
(1996) used positive, negative, and neutral information about afﬁrmative action programs
to attempt to change subjects’ attitudes toward them. The effects were in the
hypothesized direction only for White subjects. Interestingly, negative information either
strengthened or had no effect on the already positive attitudes of Asians, Hispanics, and
African-Americans.

As introduced earlier, Taylor-Carter, Doverspike and Alexander (1995) employed a
similar approach, providing subjects with pro- and anti-afﬁrmative action messages, and
assessing the impact of these messages on the perceived faimess of afﬁrmative action
programs (described Speciﬁcally as preferential treatment). Subjects reported their
attitudes toward weak preferential treatment programs, were exposed to the message,
recorded their thoughts about the message, predicted the importance of afﬁrmative action
for their lives, and rated the fairness of the program. Message polarity inﬂuenced the
number of positive thoughts in the predicted direction, but did not affect the number of
negative thoughts. Fairness ratings were related, as hypothesized, to initial attitude,
message polarity, number of positive thoughts, and number of negative thoughts. This
indicates that the perceived fairness of preferential treatment can be affected by pro- or
anti- afﬁrmative action messages. One might expect this effect to also apply to other
forms of afﬁrmative action, and similar initiatives.

As introduced earlier, afﬁrmative action may also lead to negative

characterizations of those who are perceived to have beneﬁted from them. Minorities

27

may be judged to be less competent if the organization has an afﬁrmative action program
(Garcia, Erskine, Hawn & Casmay, 1981); women leaders are responded to more
negatively, when subordinates believe that leader selection is sex-based rather than merit-
based (Jacobson & Koch, 1977).

A study by Dietz-Uhler & Murrell (1993) suggests that attitudes toward
afﬁrmative action moderate the relationships between other organizational factors, and
individuals’ attitudes toward their female colleagues who may have beneﬁted from
afﬁrmative action. This ﬁnding is also relevant to the thrust of the current study. As
indicated previously, the current study focuses on attitudes toward afﬁrmative action,
with the underlying assumption that these attitudes will predict other outcomes. The rest
of this analysis examines ways of achieving just such attitude change.

Alleviating negative attitudes -- using alternative ethical frameworks

Anecdotal evidence indicates that one of the reasons for the widespread
opposition to afﬁrmative action programs is that they are considered to be unethical or
immoral (Madkins, 1989). One of the aims of this study is to take this assertion beyond
mere opinion, and to analyze afﬁrmative action in the light of established research on
ethics and morality.

For well over a thousand years, philosophers and other scholars have investigated
issues of morality and ethics. Ethics refers to “inquiry into the nature and grounds of
morality where the term morality is taken to mean the moral judgments, standards, and
rules of conduct (Taylor, 1975, p.1).” By extension, ethical behavior refers to “that which
is shown to be objectively morally correct (permissible, obligatory, or desirable) action

and that it is ‘ethical’ precisely because it is the behavior which is required by the theory

28

(Bommer, et al., 1987, p. 267).” Note that although the terms "ethics" and "morality" are
often used interchangeably, there is a technical distinction between the two: Morals and
morality refer to conduct itself, whereas ethics refers to the study of this behavior

(T salikis & Fritzsche, 1989).

The substantial body of research on ethics has resulted in a number of competing
theories or frameworks which individuals and societies are said to use in making
judgments concerning right and wrong. For most acts, there are no universal standards
for judging their inherent rightness or wrongness. Rather, different individuals will arrive
at different ethical conclusions, depending on the analytical perspectives they use.
Additionally, individuals are capable of applying varying ethical frameworks to different
situations (i.e., there is within-person variation in predominant ethical framework). One
of the key propositions of this study is that individuals will react differently to afﬁrmative
action based on their judgments of the morality of the policy. In turn, this moral
judgment will be contingent on the dominant ethical frameworks-in-use that guide their
judgments. The study further investigates the possibility of impacting individuals’
attitudes toward afﬁrmative action by altering the ethical frameworks through which they
examine these programs.

Several of the more inﬂuential ethical frameworks are described below. These
frameworks or theories may be separated into two broad classiﬁcations -- teleological and
deontological ethical frameworks (DeGeorge, 1982; 1990; Hunt & Vitell, 1986;
McDonald & Pak, 1996).

Teleological ethical frameworks

Teleology or consequentialism is an ethical approach that focuses on the outcomes

29

or ends associated with an act or decision. Based on this framework, whether an action is
ethical or unethical depends on the ﬁnal consequences of that action. In the management
context, teleological approaches consider consequences from the perspectives of a variety
of organizational stakeholders (Hunt & Vitell, 1986). Three teleological ethical
frameworks are included in this study -- self-interest, utilitarianism, and light Of day.

Self—interest or agm is an ethical framework which indicates that actions are
moral to the extent that they promote one’s long term self-interest, or personal
satisfaction. Wood, Longenecker, McKinney, & Moore (1988) describe it as a lower
level of moral reasoning which focuses on the furtherance of one’s own interest with little
regard for higher level, more abstract moral principles. Individuals guided by this
framework would only be expected to help others in situations where they derive a net
beneﬁt from such help. Paradoxically, this framework would not advise pure selﬁshness,
as this may prove detrimental to one’s self in the longer term (Sorrell & Hendry, 1994).
Note that this discussion refers to ethical egoism which dictates that individuals M do
what is in their personal best interest. It is distinct from psychological egoism, or simple
self-interest, which suggests that individuals BE motivated by self-interest (Blackburn,
1994).

Placing this ethical framework within the focus of the present study, one would
expect that self-interest would lead non-beneﬁciaries of afﬁrmative action (predominantly
White males) to react negatively to these programs. However, there may be a small
proportion of individuals who view these programs from a broader, societal perspective
and thus see them as being indirectly self-beneﬁcial. These individuals would be

expected to react less negatively.

30

It is even more difﬁcult to predict the reactions of the intended beneﬁciaries of
afﬁrmative action based on the self-interest ethical perspective. Although a simplistic
analysis of afﬁrmative action would assume that they would react positively, many
members of the afﬁrmative action target groups see afﬁrmative action as being,
ultimately, counterproductive to them (Becker, 1995; Chacko, 1992; Gleckman et al,
1991; Heilman, 1994), and as such would react negatively. The self-interest perspective
makes salient a larger issue in predicting reactions to afﬁrmative action (introduced in the
Key Deﬁnitions section of chapter 1). Individuals’ reactions to afﬁrmative action will be
contingent on both their status as intended beneﬁciaries/non-beneﬁciaries of the programs
and their assessments of the long-term consequences of the programs (their parceptions of
whether or not they beneﬁt from the programs).

Hypothesis 1: Individuals’ scores on the self-interest ethical perspective

will relate to positive attitudes toward afﬁrmative action among

individuals who consider themselves to be beneﬁciaries of these programs.

Note should be taken that this hypotheses appears to be quite obvious if one looks
at self interest from a motivational perspective rather than an ethical perspective. The
latter serves as a normative standard, judging the rightness or fairness of an action based
on it’s potential to beneﬁt oneself. Past research has tended to assume that members of
afﬁrmative action target groups are directed by self-interest to be supportive of
afﬁrmative action (Ozawa, Crosby, & Crosby, 1996; Tougas, Crosby, Joly, & Peichat,
1995). However, the results of these studies have led to inconsistent conclusions
regarding self-interest (Veilleux & Tougas, 1989). One concern is that many of these
Studies have operationalized self-interest in terms of immediate individual gain.

However, self interest may be much more broadly conceptualized. Self interest may also

31

be used to evaluate the treatment of various groups with which one identiﬁes. A further
complication is introduced in contexts in which individuals respond altruistically,
showing concern for the members of other groups. In such situations, simple self-interest
serves as a very inappropriate predictor of reactions to afﬁrmative action (Veilleux &
Tougas, 1989).

The current study examines self-interest as an ethical framework rather than a
simple motivational concept. This analytical perspective is supported by studies which
indicate that individuals’ expected personal consequences are poor predictors of their
attitudes toward social policies (Kinder & Rhodebeck, 1982; Sears & Allen, 1984). Self-
interest, particularly in the context of attitudes toward afﬁrmative action, may be a more
complex phenomenon than meets the eye. Dietz-Uhler and Murrell (1993), attempted to
analyze this connection more thoroughly, by Operationalizing self-interest in terms of the
extent to which an afﬁrmative action policy had the potential to disrupt one’s own
institution. Although they did not ﬁnd signiﬁcant results for this evaluation, their work
provides further impetus for the current assessment of self interest (from an ethical
perspective).

Utilitarianism is the premiere theory within the broader set of teleological or
consequentialist ethical approaches. It involves the balancing of good and bad in such a
way that ethical acts are those which maximize net good, or maximize My.
Utilitarianism advises decision makers to act so as to produce the greatest ratio of net
good over bad for everyone affected by the decision (McDonald & Pak, 1996). It also
requires decision makers to assess the impact of their decisions on all concerned parties.

Utilitarianism is often expressed in terms of “the greatest good for the greatest number of

32

people.”

Although utilitarianism dominates the ethical training and reasoning of Western
business persons (Fritzsche, 1984; McDonald & Pak, 1996; Peachment, McNeil, Soutar,
& Molster, 1995), it possesses a number of shortcomings. One that is particularly
relevant to this discussion is that although utilitarianism maximizes net beneﬁt, it does
not consider the more speciﬁc distribution of costs and beneﬁts. As such, an act which is
moral based on utilitarianism may be extremely harmful to a relatively small number of
individuals (Takala & Uusitalo, 1995). Analyzing afﬁrmative action, it is again difﬁcult
to predict the utilitarian judgment because the overall consequences (costs and beneﬁts)
of afﬁrmative action are difﬁcult to assess, and are the subject of considerable dispute.

In general, utilitarianism is expected to produce negative reactions to afﬁrmative
action. In spite of the actual numbers of persons with the potential to be affected by
afﬁrmative action, there is some perception that afﬁrmative action is targeted toward a
relatively small number of individuals. Thus, it is likely that utilitarian judgments will
compare the beneﬁts accruing to this group, to the potential adverse consequences for the
larger, non-beneﬁciary group, and conclude that on the whole, afﬁrmative action is of
more harm than good. Of course, since the relative level of harm versus good attributed
to afﬁrmative action, and the potential number of persons affected are open to
speculation, a variety of individual differences are likely to inﬂuence these results.
Analyses of these ﬁner distinctions are reserved for future study. Overall:

Hyp_othesis 2: Individuals’ scores on the utilitarianism ethical perspective
will relate to negative attitudes toward afﬁrmative action.

The ﬁnal teleological framework is known as the light of day principle. This is a

33

very practical approach to ethics in which the primary consideration of the decision-
maker is the consequences associated with a public revelation of the act. Based on this
approach, acts/decisions are unethical if they would be embarrassing to the decision
maker if reported to the public, including the decision maker’s family, friends, and
associates. Once again, the reactions of individuals who are led by the light of day ethical
framework are hard to predict. Since this framework is primarily concerned with the
reactions of others, the evaluation of the morality of afﬁrmative action for persons guided
by the light of day framework will be contingent on their assessments of the societal
evaluation of afﬁrmative action.

Hymthesis 3: Scores on the light of day ethical perspective will relate to

positive attitudes toward afﬁrmative action among individuals who believe

that society in general, views afﬁrmative action positively.
It might appear that individuals would respond uniformly in their assessments of the
societal View of afﬁrmative action, in that afﬁrmative action is designed to provide help
and support for those in need. In fact, some writers have assumed that our social climate,
which opposes discrimination and racism, will lead to support for afﬁrmative action, at
least at the level of individuals’ public expressions (Veilleux & Tougas, 1989). However,
this assumption may not hold true in contemporary America, in light of the heated,
ongoing public debate on afﬁrmative action.
Deontological Ethical Frameworks

Deontology (or nonconsequentialism) refers to a collection of ethical theories
which dictate that some actions are moral, or right, in and of themselves, independent of

their consequences. These theories result in a collection of rights and obligations which

are based on humanity, general use, and societal consensus. Included in these are

34

obligations to help others, to remedy injustice, and to promote human welfare.

Four deontological ethical frameworks are applied in this study -- categorical
imperative, duty, neutralization, and religious conviction. The categorical imp_erative
based on the work of Immanuel Kant, is comprised of a number of rules or forms, two of
which predominate. The universal rule is based on the question of whether the decision
maker would be willing to have his/her decision reversed or universalized. For a decision
to be ethical, the decision maker would have to be willing to have others reciprocate
his/her decisions, or for the decision rule to become universal law.

The means-end rule is concerned with whether individuals are being treated as
“ends” in themselves (respected and possessing inherent rights) or as “means” to achieve
some other objective (McDonald & Pak, 1996).

The categorical imperative results in absolute pronouncements on what

individuals should do, regardless of the consequences. It is contrasted with the

 

hypothetical imperative which results in if—then statements. In other words, the
pronouncements of the hypothetical imperative only apply to persons who are interested
in certain ends. Those of the categorical imperative apply to all persons (Blackburn,
1994).

Individuals guided by the categorical imperative will be expected to react
positively to afﬁrmative action. Afﬁrmative action is designed to help individuals who
have, in one way or another, been harmed or disadvantaged. In spite of the evidence
suggesting that it has been less than successful, afﬁrmative action’s goals are benevolent
and consistent with the categorical imperative. Thus:

Hypothesis 4: Individuals’ scores on the categorical imperative ethical

35

perspective will relate to positive attitudes toward afﬁrmative action.

The deontological theory of Qty asserts that some actions are inherently right or
moral because of the duty that one possesses. Duty may result from obligations such as
promises made, or from relationships, such as that between parent and child (Honderich,
1995). Duty has even been broadly interpreted by some philosophers as an obligation
owed to all human beings by virtue of their humanness (Blackburn, 1994). It becomes
obvious that these notions are somewhat vague and encompass a broad range of possible
responsibilities. In the more Speciﬁc organizational context, managers’ possess a set of
generic roles which carry with them a number of duties (Mintzberg, 1973). Based on the
duty perspective, fulﬁlling the requirements of these roles constitutes ethical behavior
(McDonald & Pak, 1996).

In the context of this study, duty will be related to positive attitudes toward
afﬁrmative action in organizations that have either an explicit policy, or set of norms
favoring afﬁrmative action and related efforts (such as management of diversity). I
however, do not expect duty to be related to negative attitudes toward afﬁrmative action
in organizations that are not supportive of afﬁrmative action (I do not expect organization
members to develop an obligation to oppose afﬁrmative action).

Hypothesis 5: Individuals’ scores on the duty ethical perspective will relate

to positive attitudes toward afﬁrmative action in organizations that are

supportive of afﬁrmative action and similar policies.

Neutralization, more formally referred to as “the techniques of neutralization” is a
relatively new and under-investigated ethical approach (McDonald & Pak, 1996). Vitell
and Grove (1987) describe this ethical framework as “a learned vocabulary of motives for

misconduct, used to protect one from self-blame (p. 434).” Individuals learn to call upon

36

a set of subconscious, unverbalized values as a means of diminishing their personal
culpability for socially disapproved acts. These values are learned and socially reinforced
counters to more accepted and proclaimed normative moral standards.

As the formal name implies Neutralization is comprised of multiple techniques,
including denial of responsibility; denial of injury; denial of victim; condemning the
condemners, and appeal to higher loyalties (Sykes & Matza, 1957).

According to this theory, unethical activities are legitimated by an abandonment
of previously held ethical convictions. Ethical norms are, after all, seen as “qualiﬁed
guides for action (Sykes & Matza, 1957:666)” rather than imperatives. Individuals in
these cases feel that the norms they are violating do not, or should not apply to particular
times, places, persons, or social circumstances (McDonald & Pak, 1996; Sykes & Matza,
1957; Vitell & Grove, 1987).

Once again, I expect that because afﬁrmative action was designed to consider the
needs of historically oppressed/deprived individuals, most individuals will consider these
initiatives as generally good or right things to do. Thus, I expect that neutralization will
lead individuals to abandon their concerns for others and as such, react negatively to
“helping” programs. However, as with the duty perspective, I do not expect this effect to
be reversed in organizations that are not supportive of afﬁrmative action.

Hypothesis 6: Individuals’ scores on the neutralization ethical perspective

will relate to negative attitudes toward afﬁrmative action in organizations

that are supportive of afﬁrmative action and Similar policies.

Religious or philosophical conviction leads decision makers to refer to their
religious beliefs or the dictates of their faith for direction on ethical questions. Since

most religious/philosophical systems advocate kindness toward others, selﬂessness, and

37

giving, I expect that the religious conviction ethical perspective will be related to positive
attitudes toward afﬁrmative actions.

Hypothesis 7: Individuals’ scores on the religious/philosophical conviction

ethical perspective will relate to positive attitudes toward afﬁrmative

action. «

It will be possible to investigate the preceding hypotheses using correlational
analyses of the pre- and post-test data. However, the focus of the study is on the use Of
theory-based manipulations to change individuals’ attitudinal reactions to afﬁrmative
action. In the ethics manipulation, individuals will be exposed to alternative ethical
frameworks (independently of the afﬁrmative action context) and their reactions to
afﬁrmative action/afﬁrmative action programs will again be assessed (see chapter 3 for a
detailed description of the experimental methods).

Because of the complexity of the ethics theories outlined above, and to keep the
experiment practicable, the ethics manipulation will be simpliﬁed to focus on the two
general forms of ethics frameworks. The manipulations, which will take the form of a
video-taped training program, will attempt to inﬂuence individuals to move from
consequentialist to non-consequentialist perspectives. Based on an overall analysis of
hypotheses 1 through 7, it is expected that:

Hypothesis 8: Individuals’ scores on consequentialist (teleological) ethical
perspectives will relate to negative attitudes toward afﬁrmative action.

Similarly,

Hyp_othesis 9a: Individuals’ scores on non-consequentialist (deontological)

 

5 Although there is vocal opposition to afﬁrmative action from groups associated with religion (the
“Religious Right,” and “Moral Majority”) I do not see these largely political perspectives as equivalent to,
or even correlated with the religious conviction ethical framework. This conservative religious/political
position represents a fraction of those who are led by deeply held religious or philosophical convictions.

38

ethical perspectives will relate to positive attitudes toward afﬁrmative
action.

Additionally,

Hypgthesis 9b: Making non-consequentialist (deontological) ethical

perspectives more salient to individuals will lead them to adopt more

positive attitudes toward afﬁrmative action.

Alleviating negative attitudes -- The role of justice and fajmess norm;

The previous section describes an anticipated direct effect of ethical frameworks
on attitudes toward afﬁrmative action. In addition to this direct effect, it is likely that
justice or fairness norms may play a mediating effect in the relationship between ethics
and attitudes toward afﬁrmative action. Although some authors have described fairness
and ethics as equivalent constructs (Schminke, Ambrose, & Noel, 1997), I see them as
related but distinct. Fairness may be viewed as a particular form of ethics, focusing on
right and wrong in the context of social exchange. Nonetheless, the two constructs, and
by extension the disciplines of organizational justice and business ethics are closely
related, though most often studied in isolationfrom one another. The study by Schminke
and colleagues is one of the ﬁrst to examine the relationship between them (Schminke, et
al, 1997). The current section provides an overview of the well established literature on
organizational justice, focusing on the justice frameworks with the potential to inﬂuence
individuals’ attitudes toward afﬁrmative action.

Research on justice and fairness indicates that there are multiple perspectives with
which individuals appraise Situations, in order to make a decision as to whether those

situations are fair or justé. Issues of fairness and justice arise of necessity within social

 

6 For the purpose of this discussion, “fair” and “just” are considered synonymous.

39

exchange processes and the social interactions within organizations are no exceptions.
More speciﬁcally, it has been demonstrated that fairness is one of the considerations used
by individuals to evaluate afﬁrmative action (Heilman & Herlihy, 1984; Tougas, et al.,
1995).

In the organizational context, issues of fairness and justice, and their resultant
behavioral and performance consequences make up the organizational justice domain (a
term coined by Greenberg, 1987). For the purposes of this study, I will not attempt a
comprehensive review of this literature but will discuss seminal works, especially those
that pertain to the issue of afﬁrmative action (for comprehensive reviews of the
organizational justice literature, see Cropanzano & Greenberg, 1997; Greenberg, 1990a).
Distributive justice

The literature on organizational justice has its genesis in the study of equity theory
(Adams, 1965; Walster, Berscheid, & Walster, 1973). According to equity theory,
individuals intuitively arrive at a ratio of the outputs, or beneﬁts they derive from the
organization (such as pay, beneﬁts, social status, positive affect) to the inputs they put
into the organization (effort, ability, qualiﬁcations, experience). They then compare this
ratio to their perceptions of what this ratio is for other (referent) organization members.
If the ratios are unequal, the person with the higher ratio is said to be overpaid and is
expected to feel guilty, whereas the person with the lower ratio is considered underpaid
and expected to feel angry.

These affective reactions are consistent with the predictions of cognitive
dissonance theory. Cognitive dissonance theory asserts that persons in such situations

whose experiences do not match their expectations, experience dissonance. This is

40

accompanied by feelings of tension, proportional to the magnitude of the perceived gap or

 

inequity. The tension motivates individuals to act to eliminate or reduce the inequity.
Again, the motivation to act is proportional to the degree of perceived inequity. Drawing
on cognitive dissonance theory, Adams (1965) advanced a number of speciﬁc means for
reducing this inequity.In general, individuals may act to alter their perceptions of the
relative value of the inputs or outputs of themselves or their comparison others (Adams,
1965; Greenberg, 1984). This may be accomplished using either psychological
adjustments such as devaluing the outcomes received by themselves or another
individual, or by behavioral changes such as leaving work early, or stealing from the
organization (Greenberg, 1990a; Greenberg, 1990b). In situations where individuals’
perceived input-output ratios are equal to those of their comparison others, a certain
equilibrium exists and the parties consider the situation to be fair or equitable.

The postulations of equity theory have been quite thoroughly tested empirically,
and a number of revisions have been made to the early conceptualizations of the theory.
However, the theory has held up quite well under this scrutiny and is today “a well
accepted midrange theory of organizational behavior (Weick, 1966:439).”

Equity theory focuses in part, on fairness or justice judgments based on the
outcomes individuals receive as a part of their organizational experience. This focus on
outcomes is the basis for research in distributive justice (Homans, 1961; Leventhal,
1976). Although the original concept of distributive justice (Homans, 1961) predates
equity theory, equity theory is one of the ﬁrst theories to apply these ideas in the
organizational context (which is the focus for the remainder of this discussion).

An important facet of the issue of distributive justice is that individuals do not

41

evaluate the fairness of their work outcomes in isolation, but compare their outcomes to
those of others in the organization. Thus, one’s absolute organizational outcomes may
not contain much information in terms of fairness or justice until they are compared to the
outcomes obtained by referent others. Indeed, empirical studies demonstrate that
considerable variance in individuals’ satisfaction with their work outcomes is inﬂuenced
by these comparisons, over and above the variance explained by the actual magnitude of
the outcomes (Sweeney, McFarlin, & Inderrieden, 1990). Research also indicates that the
referents that individuals use in making distributive justice comparisons are not stable,
but may change under certain conditions (Stepina & Perrewe, 1991). Similarly,
distributive justice assessments are inﬂuenced by national culture and changes in that
culture (Chen, 1995).

The importance of distributive justice has been supported by several empirical
studies. For example, some researchers have shown that it can have a robust effect on a
variety of important behaviors in organizations including job performance, and
withdrawal behavior such as turnover and absenteeism. For example, Greenberg (1990b)
found a connection between distributive justice and employee theft rates. He found that
employees in two plants where across-the-board pay cuts were introduced, stole more
than demographically similar employees in another plant where there were no pay cuts.

Although equity is a very prevalent distributive justice norm, it is not the only one.
Equity theory and the discussion above, assume that individuals are basing their justice
considerations on a concern for allocating rewards to individuals in proportion to their
inputs. This assumes an equity or merit-based norm of distributive justice, also referred

to as a contributions rule (Leventhal, 1980). Decision makers use different allocation

42

norms based on the instrumental value of these allocation rules. In other words, different
distributive justice norms are expected to produce different responses from affected
individuals (Leventhal, 1976).

The equity norm described above is used to foster high levels of task performance
(Leventhal, 1976). By reinforcing beneﬁcial inputs, the decision maker is expecting
individuals to contribute increasing amounts of such inputs to the organization. An
alternative distributive justice norm is eguality, in which all participants receive the same
outcomes, regardless of their individual performance levels. This norm is used to
minimize dissatisfaction and reduce the incidence of disruptive, antagonistic behavior on
the part of those who might be dissatisﬁed under an equity allocation rule (Leventhal,
1976).

A third, relatively common distributive justice norm (or allocation rule) is that of
peg. This is a norm in which outcomes or rewards are distributed to those who have the
greatest need for them. This distributive justice norm emphasizes a concern for social
responsibility and mutual responsiveness. It is particularly prevalent in situations where
the allocator feels a sense of responsibility for the welfare of the individuals affected by
his or her decisions. Like the equality norm, use of the need norm may also have the
effect of reducing antagonism and conﬂict, and thus maintaining harmony. The need
norm may also have the practical beneﬁt of reducing waste by allocating resources only to
those who are most likely to use them (Leventhal, 1976). The difference between these
norms of distributive justice takes on special signiﬁcance in the analysis of individuals’
responses to organizational initiatives such as afﬁrmative action. Although all three

norms are legitimate and lead to “fair” allocation decisions (Leventhal, 1976), one would

43

expect individuals to react quite differently to such an initiative depending on which
norm they employ to analyze the policy. This concern is discussed in greater detail
below.
Procedural justice

In addition to the evaluation of the fairness of outcome distribution, individuals
also assess the degree to which the processes through which social exchanges occur (the
processes used to determine the outcomes) are fair. This assessment is described as a
concern for mcedural justice (Deutsch, 1975; Leventhal, 1976; Thibaut & Walker,
1975). Procedural justice issues include such speciﬁc concerns as the degree Of
participation or y_o_i_cp exercised in the decision process (Deutsch, 1975; Hirschman,
1970), and the perceived fairness of the rules, procedures, and values that guided the
process (Folger, 1977). In turn, the perceived fairness of procedures is determined by
resource distributions that are consistent across persons and over time, free from bias,
based on accurate information, correctable, representative of all recipients’ concerns, and
based on prevailing moral and ethical standards (Leventhal, 1976, 1980; Leventhal,
Karuza & Fry, 1980). Subsequently, a number of scholars (Folger & Greenberg, 1985;
Greenberg, 1987) have speciﬁcally applied this concept to the organizational context (for
a review, see Greenberg, 1990a).

The effects of procedural justice in affecting fairness judgments are postulated to
stem from two possible models. The self-interest, or instrumental model (Tyler, 1987)
suggests that individuals seek control over processes with the assumption that they can
inﬂuence the outcomes. Alternatively, the group value, or relational model suggests that

people attach value to the opportunity to express their beliefs, even when those beliefs do

44

not necessarily inﬂuence the decision outcomes (Lind, 1995; Lind & Tyler, 1988; Tyler,
1989; Tyler & Lind, 1992; Tyler, Rasinski, & Spodick, 1985). According to this model,
people value long-term relationships with groups and often see group membership as a
source of self-esteem and social status. This leads them to value procedures that promote
group solidarity.

A number of empirical studies have supported these standards for assessing
procedural justice. Speciﬁcally testing the group value model, Tyler (1989) found that a
large proportion of variance in individuals’ procedural justice assessments (speciﬁcally,
assessments of the Chicago legal authorities) was determined by non-outcome related
factors. He found that perceptions of neutrality of treatment, trust in the authorities, and
indicators of social standing (such as respectful treatment) explained more variance than
the perception Of perceived control over the process. In other words, “pure” procedural
considerations outweighed those procedural concerns that were tied to outcomes. Similar
support for such an emphasis on procedures has been obtained by other researchers both
in the lab (Cropanzano & Randall, 1995) and in the ﬁeld (Conlon, 1993).

Other empirical studies have demonstrated that justice concerns have robust
effects on a variety of organizational outcomes. For example, procedurally fair treatment
has been shown to make individuals more accepting of smoking bans (Greenberg, 1994),
and less likely to steal from their employers in the event of a pay cut (Greenberg, 1990b).
Impressions of procedural fairness have also been shown to make employees more
accepting of strategic organizational changes (Novelli, Kirkman, & Shapiro, 1995), and
behavioral reactions to a workplace drug screening program (Konovsky & Cropanzano,

1991). Similarly, procedurally just explanations of workforce diversity programs have

45

been shown to improve the assumptions program beneﬁciaries make about the attitudes
of their co-workers toward them (Richard, Fubara & Castillo, 2000).

Procedural fairness may also inﬂuence the way in which individuals develop
overall attributions of their organizations. Such considerations may inﬂuence their
loyalty to the organization, and their willingness to exert effort on its behalf (Tyler &
Lind, 1992). However, this is a very complex relationship -- two other studies by
Brockner and colleagues (Brockner, Tyler, & Cooper-Schneider, 1992, Studies 1 and 2)
suggest that relational concerns are contingent on previously held commitments to the
organization. Individuals who are more committed to the organization are more likely to
become upset over procedural violations. Thus, the effect of procedural justice on other
organizational outcomes is moderated by individuals’ commitment to the organization.
This notion has been supported empirically (Brockner, et al., 1992).

A distinct attribute of procedural justice research is research on the social aspects
of procedural justice. This domain, also referred to as interactional justice (Bies & Moag,
1986) examines individuals’ reactions to the interpersonal treatment they receive from
decision makers (Cropanzano & Greenberg, 1997). This form of procedural justice is
further subdivided into concerns of social sensitivity -- the extent to which people believe
they have been treated with dignity and respect; and informational justiﬁcation -- the
extent to which people believe they have adequate information about the procedures
affecting them (Cropanzano & Greenberg, 1997; Greenberg, 1993; Tyler & Bies, 1990).
Since interactional justice considerations are not directly relevant to this work, they will
not be discussed in any further detail (for a more detailed discussion, see Bies & Moag,

1986; Cropanzano & Greenberg, 1997; Greenberg, 1993; Tyler & Bies, 1990).

46

One of the contemporary concerns of organizational justice researchers is the
relationship between distributive and procedural justice, particularly in terms of their
distinct effects on various organizational outcomes. One approach (the two factor model)
postulates that procedural justice concerns are expected to primarily affect satisfaction
with organizational systems, whereas distributive concerns affect organizational outputs
(Sweeney & McFarlin, 1993; Tyler & Lind, 1992).

An alternative model (known as the interaction model) proposes that there is an
interactive relationship between the two, in which procedural justice has a stronger effect
on work related attitudes when outcomes are low than when they are high. Conversely,
this same perspective suggests that outcomes are more strongly related to work attitudes
when procedures are unfair (Brockner & Wiesenfeld, 1996; Cropanzano & Greenberg,
1997).

Organizational justice and afﬁrrmtiveaction

AS introduced earlier, justice is an important concern in analyzing various issues
surrounding afﬁrmative action, particularly, individual reactions to afﬁrmative action and
afﬁrmative action programs. According to Ozawa and colleagues “some people’s
reaction to afﬁrmative action may be based on their understanding of how the programs
support or violate aspects of distributive and procedural justice” (Ozawa, Crosby, &
Crosby, 1996, p. 1140). Similarly, Kravitz (1995) discovered that the effect of an
afﬁrmative action program manipulation was completely mediated by judgments of
fairness. In other words, although some individuals are opposed to afﬁrmative action
based on political, racist, or simple self-interest motives, there may also be an inﬂuence

of genuine justice concerns.

47

Apparently, the primary justice norm used in analyses of afﬁrmative action is
equity (Nacoste, 1990) which reﬂects an emphasis on outcomes (distributive justice).
Equity theory analyses lead to negative attitudes toward afﬁrmative action, in that sex,
ethnicity, and other non job-related variables are used as inputs in the equity formula, in
afﬁrmative action situations. Both beneﬁciaries and observers of afﬁrmative action
devalue these inputs in their equity comparisons. As a result, they develop feelings of
inequity (Nacoste, 1990; Stout & Buffum, 1993). On the contrary, one might expect that
equality or need norms of distributive justice would lead to more accepting attitudes.

Nacoste (1990) identiﬁes three reasons why equity theory may be inappropriate
for analyzing afﬁrmative action. First, non-performance criteria are often used in equity
comparisons, but the theory does not provide any a priori speciﬁcation as to when they
are acceptable. Secondly, equity theory does not account for the weight assigned to non-
perforrnance related factors which varies quite signiﬁcantly from one policy to another.
Third, equity theory, being a static model of motivation, focuses only on the ﬁnal
outcome distributions, with no consideration of the dynamic processes used to arrive at
these outcomes.

The conﬂict between equity and non-equity norms is discussed in the context of
afﬁrmative action by Weining Chang (1996). She argues that although equality is a
central concern of the American collective consciousness, public efforts to promote
equality have encountered opposition. The nation’s need for equality has come into
conﬂict with its equally dominant need for equity or meritocracy, which Chang (1996)
describes as “the favored principle for the distribution of wealth and resources (p.94).” In

the context of afﬁrmative action, individuals resolve this conﬂict by evaluating the

48

overall perceived fairness of the policy. This evaluation is inﬂuenced by subjective
individual differences in the weights assigned to equality and equity. Again, in the
speciﬁc context of afﬁrmative action, the overall fairness judgment is inﬂuenced by the
extent to which afﬁrmative action serves to highlight group differences, and individuals’
perceptions of the scarcity of organizational resources. Individuals who view
organizational resources as ﬁxed in capacity will tend to see afﬁrmative action as a zero-
sum game, leading them to oppose the policy.

Veilleux and Tougas (1989) interpret the equity-equality conﬂict in terms of
support for two different kinds of organizational programs. Equity concerns lead to
opposition to programs which involve any form of preferential treatment. On the other
hand, equality concerns lead to support for equal employment opportunity and equal
employment opportunity programs.

Susan Opotow (1996) also frames reactions to afﬁrmative action in terms of
justice concerns. She discusses both distributive and procedural justice issues as pivotal
questions in the afﬁrmative action debate. One distributive justice issue involves which
groups of individuals fall into one’s justice framework. There are established boundaries
for determining whom one considers in one’s justice comparisons, and thus, for
determining which individuals deserve to be treated justly. For example, groups that are
considered beneﬁcial; groups whose conﬂict with the focal individuals’ groups are seen
as not being severe; and groups of persons considered to be similar to the members of the
focal group (under conditions of low conﬂict) are generally included in one’s scope of
justice. One of the ironic effects of afﬁrmative action is to bring different groups into

closer proximity. Under these conditions, the conﬂict between groups is likely to be seen

49

as more severe and the afﬁrmative action beneﬁciaries are then excluded from the focal
individuals’ sc0pe of justice.

In addition to the concern for which individuals are allowed into one’s scope of
justice, there is an additional question of whether the means for accommodating those
who have been included are fair. In Opotow’s model of procedural justice, there are thus,
two hurdles in judging the fairness of afﬁrmative action.

At least theoretically, the one thing that all afﬁrmative action programs and
policies have in common is that they give some positive weight to membership in the
target group so as to increase opportunities for members of these groups. This positive
weight may be applied at various points in the employment process, from recruitment to
selection and promotion. The issue of greatest contention is how much weight is, or
should be, ascribed to group membership (Nacoste, 1990). In procedural justice terms,
this is a question of voice. As introduced in the general discussion of procedural justice
(above), voice refers to the degree of participation individuals are able to exercise in the
decision process (Deutsch, 1975; Hirschman, 1970).

Another analysis of the role of procedural justice in afﬁrmative action decision
making was conducted by Nacoste (1994). He expands upon his earlier work (Nacoste,
1990) which established that procedures used to enact afﬁrmative action have signiﬁcant
effects on the reactions of individuals to the policy. Over the course of a number of
studies, he has investigated in some detail, the mechanisms, through which procedural
concerns impact individuals’ reactions to afﬁrmative action. Nacoste’s analyses are
discussed brieﬂy below.

In one of these studies, N acoste (1993) developed the theory of procedural

50

interdemndence. This theory focuses on the effect of afﬁrmative action procedures on
the patterns of triadic interdependence between organizational interest groups. The three
focal groups are generally the agency enacting the policy, the target group members, and
the non-target group members. Individuals tend to prefer procedures which serve to
preserve existing intergroup relationships and encourage continued interaction among the
groups (Thibaut & Walker, 1975). High voice procedures, which, as discussed earlier,
are seen as more procedurally just, are considered to be more effective in preserving the
pattern of procedural interdependence (Tyler, 1990).

Nacoste (1994) describes afﬁrmative action programs as providing individuals
with varying levels of voice, depending on the degree to which group membership is
considered in the decision process. Technical credentials are seen as high-voice inputs
into the decision making process because individuals have a relatively signiﬁcant degree
of control over them. On the other hand, group membership is generally beyond the
control of the individual. As such, organizational decisions which take group
membership into consideration reduce voice and thus, reduce conceptions of procedural
justice.

Nacoste also asserts that individuals’ reactions to afﬁrmative action stem from a
more complex analysis of three different factors: The ﬁrst involves a comparison of the
(real or perceived) policy to one which gives too much weight to group membership. The
second is a comparison between the policy and the set of better altem‘atives that the
observer considers to be realistically available. The third analysis assesses the extent to
which the overall effect of the policy places target group members at an advantage, on

par, or at a disadvantage, compared to non-target group members (N acoste, 1994).

51

Bobocel and Farrell (1996) conducted two experiments assessing White men’s
reactions to the explanations (accounts) given for promoting a woman over an equally
qualiﬁed man. The experiments speciﬁcally assessed the role of the adequacy and
fairness of the accounts on subjects’ ratings of interactional fairness. The ﬁndings of
Experiment 1 suggest that providing a clear, ideologically based account (a social and
moral argument for parity) for the preferential treatment was seen as considerably fairer
and more adequate than one which abdicated responsibility for the decision. This account
was also seen as fairer than, but equally adequate as no justiﬁcation at all.

Experiment 2 was designed to assess one of the weaknesses of the ﬁrst study, the
lack of distinction in the ideological account between the responsibility for the decision,
and the justiﬁcation or reason for the decision. In the responsibility manipulation,
subjects were provided with scenarios in which the decision maker either assumed or
denied responsibility for the decision. The justiﬁcation manipulation provided either a
redressing-discrimination justiﬁcation, an increasing-diversity justiﬁcation, or no
justiﬁcation at all. The ﬁndings revealed that justiﬁcation rather than responsibility
inﬂuenced subjects’ judgments of interactional fairness, regardless of the particular
justiﬁcation used. Additionally, the effect of justiﬁcation was completely mediated by
the perceived adequacy of the justiﬁcations.

Rosen & Mericle (1979) found that weak and strong afﬁrmative action statements
were equally effective in counteracting sex bias in selection decisions. They conducted
an experiment in which subjects played the role of hiring managers in an organization
with an afﬁrmative action policy. Although the weak and strong policies were equally

effective in counteracting sex bias, comparatively lower starting salaries were

52

recommended for females in the strong preferential treatment condition. This outcome
was explained in the context of reactance theory, that is, the subjects (hiring managers)
were exercising their authority over these candidates as a reaction to the lack of .authority
imposed by the strong preferential treatment policy.

Matheson, Echenberg, Taylor, Rivers and Chow (1994) took a unique approach in
examining (Canadian) women’s reactions to sex-based afﬁrmative action. They
examined the effect of the level of perceived discrimination in the organization on
women’s attitudes toward a range of programs. The levels of discrimination varied in
terms of the extent to which women were disadvantaged (mistreated) by the organization.
The afﬁrmative action programs varied in the extent to which women were advantaged
(preferentially treated) by the organization. Contrary to the hypotheses, women’s support
for afﬁrmative action programs was solely contingent on the nature of the program. They
expressed no support for any form of preferential treatment, regardless of the
organization’s prior history of discrimination (Study 1). This ﬁnding was replicated in a
second study (Matheson, et al., 1994, Study 2) even when the subjects were limited to
women who expressed a personal value for social equality. A third study (Matheson, et
al., 1994, Study 3) indicated that women who were more personally threatened by
discrimination were more likely to endorse preferential treatment. Interestingly, their
support for various forms of action did not differ with the strength of the remedy.

In an experimental study designed to assess the reactions of non-beneﬁciaries of
afﬁrmative action programs, Heilman, McCullough, and Gilbert (1996) studied the
reactions of a group of males to a leadership task in which females were selected to serve

as leaders. Their study examined reactions to the task itself, reactions to the female

53

leaders, and the participants’ willingness to engage in additional pro-social behavior. The
study’s results indicated that the procedural aspects of an afﬁrmative action policy were
irrelevant to non-target group members (in terms of their task-related reactions) as long as
they received their deserved outcomes. Non-target group members who believed they
had been treated in accord with their qualiﬁcations reacted similarly whether a merit-
based or preferential treatment selection process were used. However, in cases where the
non-target group members believed that they deserved the positions under consideration,
procedural manipulations had a signiﬁcant impact on their task-related reactions.
Interestingly, a similar reaction was observed in cases where the bypassed non-target
group members were unaware of their relative qualiﬁcations. Their responses were
equivalent to those of persons with superior qualiﬁcations.

Additionally, it was found that the preferential treatment manipulation uniformly
reduced participants’ willingness to engage in pro-social behavior. This result persisted,
regardless of how the participants perceived their task-related ability. This ﬁnding
suggests that although individuals overall justice perceptions are affected by both
distributive and procedural concerns, that these dimensions may have differential impacts
on the core and peripheral work domains.

The presence of mitigating justiﬁcations for preferential selection had a very
limited impact on participants’ reactions. The presence or absence of an explanation only
affected participants work-related reactions’ in situations where they understood that they
were equally qualiﬁed with the preferentially selected target-group members. The ﬁnal
category of reactions investigated in this study was reactions to the selected target group

member herself. Interestingly, neither distributive nor procedural justice manipulations

54

affected the likeability ratings of the female leader. In each case, she was seen as highly
likable.

Both observers and members of the target groups may question their qualiﬁcations
and performance (Nacoste, 1989). Nacoste refers to this devaluing of the performance of
individuals perceived to be afﬁrmative action beneﬁciaries as atjgpia (Nacoste, 1990).
Individuals who View afﬁrmative action (operationalized as preferential treatment) as an
unfair policy, and who see themselves as p_ossible afﬁrmative action beneﬁciaries also
tend to reduce their assessments of their own performance. Those who View afﬁrmative
action as fair also reduce their assessment of their own performance if they feel that they
may have been selected on the basis of afﬁrmative action. However, this effect is much
less severe (and not statistically signiﬁcant) for those who consider afﬁrmative action to
be fair (Nacoste, 1989). Interestingly, the individuals in this study who evaluated
themselves the most negatively were also the highest performers on the experimental
task. Nacoste’s work (1989; 1990) places an emphasis on the fact that afﬁrmative action
is not inherently stigmatizing but that important procedural concerns inﬂuence the extent
to which beneﬁciaries are stigmatized. From an applied perspective, it is thus vitally
important for managers to structure programs in such a way that they appear to be
procedurally fair and that they make the nature of these initiatives widely known
throughout the organization.

The common theme of the aforementioned studies is that justice concerns are of
pivotal importance in inﬂuencing individuals’ reactions to afﬁrmative action. Both
distributive and procedural justice concerns have been shown to have a robust effect.

Most of these studies though, have focused on speciﬁc afﬁrmative action programs and

55

the differential reactions resulting from hard and soft preferential treatment. Unlike these
studies, the current study focuses on the different reactions to afﬁrmative action
associated with different individual justice frameworks as opposed to different forms of
afﬁrmative action program. In fact, rather than present differing afﬁrmative action
programs to subjects, they will be presented simply with the concept of afﬁrmative action
and allowed to enact their own schema for what the policy entails.

Three justice frameworks are examined in terms of their relationship to attitudes
toward afﬁrmative action. The ﬁrst two, equity and non-equity are both distributive
justice concerns. Note that rather than contrast equity with equality, the comparison is
broadened: the non-equity norm would include norms of need (described above) and
other less common distributive justice norms (such as reciprocity). The third justice
concern is procedural justice. In the model of attitudes toward afﬁrmative action
presented below, these justice norms serve as possible intervening variables between
individuals’ ethical frameworks and their attitudes toward afﬁrmative action.

The relatively gross comparison of equity versus non-equity for examining
distributive justice is primarily adopted for the sake of simplicity and practicality. In
addition to the practical research advantages of this approach, it is also reasonable in that
equity is such a dominant perspective in American society. Contrasting the effects of
equity and non-equity will thus provide useful information. Formally:

Hymthesis 10: Individuals’ scores on equity (versus non-equity) norms of

distributive justice will relate to negative attitudes toward afﬁrmative

action.

Some scholars have also suggested that afﬁrmative action would be more

palatable if organization members (including afﬁrmative action beneﬁciaries) focused on

56

procedural rather than distributive justice concerns (Nacoste, 1990). The pivotal
procedural justice issue involved in individuals’ reactions to afﬁrmative action appears to
be the extent to which group membership is factored into organizational decision making.
For example, it has been shown that afﬁrmative action programs that include an emphasis
on merit (a procedural concern) are perceived as more acceptable (Kravitz & Platania,
1994; Tougas, et al., 1995). Similarly, Veilleux & Tougas (1989) found that afﬁrmative
action programs involving preferential treatment, elicited more negative responses than
those involving helping strategies (a more passive, equal opportunity approach), even
when the programs had the same goals.

Other procedural justice related manipulations have been Shown to positively
impact individuals’ reactions to afﬁrmative action and similar programs (Richard, Fubara
& Castillo, 2000).

Opposition to afﬁrmative action may be reduced by minimizing threats to the self
interest of organization members, and by ensuring that they do not violate norms of
procedural fairness. Kravitz (1995) emphasizes structuring programs in such a way that
they satisfy these conditions. I am approaching the same issue from a slightly different
direction: in many cases, managers have little leeway in restructuring organizational
programs (because of legal restrictions, cost, or organizational inertia). Similarly,
organization members may respond to the very existence of afﬁrmative action, rather than
Speciﬁc initiatives. In such cases organizational decision makers may be able to alleviate
negative responses by addressing the fairness norms used by organization members in
responding to afﬁrmative action.

Procedural justice research has demonstrated that an emphasis on procedural

57

justice can ease the negative feelings associated with personal hardship (Folger & Bies,
1989; Greenberg, 1990c). In other words, even though individuals may see themselves as
having been disadvantaged by an organizational initiative, their negative reactions may be
alleviated if they focus on the procedural, rather than distributive aspects of the initiative.
In the case of afﬁrmative action:

Hypothesis 11: Individuals’ scores on procedural (versus distributive)
justice will relate to positive attitudes toward afﬁrmative action.

As introduced earlier, organizational justice concerns serve as possible mediating
variables in my model of attitudes toward afﬁrmative action (below). The work of
Schminke and colleagues (Schminke et al., 1997) is foundational in establishing this
relationship.

Schminke at a] (1997) use simpliﬁed taxonomies for examining both justice
frameworks and normative ethics frameworks and then postulate an integrated model.
Their model of justice contrasts distributive and procedural justice, while their model of
ethics compares formalism to utilitarianism. These terms are used synonymously with
deontology (nonconsequentialism) and teleology (consequentialism) respectively, as
described earlier in this chapter (see also Brady, 1990). Although these two theoretical
traditions are usually studied in isolation from one another, they both exhibit similar
emphases on process versus outcome. Individuals' ethical predispositions focus them on
process or outcome aspects of Situations, making them more attentive to one or the other
in organizational interactions. Experimental analysis supported the proposed
relationship: ethical frameworks served as lenses through which individuals evaluated

organizational decisions and inﬂuenced how they reacted to those decisions. Individuals

58

who were higher on utilitarianism (consequentialism) were more sensitive to distributive
justice concerns (outcomes). Those who were higher on formalism
(nonconsequentialism) were more sensitive to procedural justice concerns. The authors
encourage further investigation of the effect of individual ethical predispositions on
reactions to various organizational events.
Summgy and Conceptual Model

This chapter has provided a review of the more signiﬁcant research on afﬁrmative
action, and afﬁrmative action programs, primarily their psychological aspects. It also
included a brief overview of relevant historical and legal issues associated with the
practice. Based on additional research ﬁndings from the ethics and organizational justice
disciplines, a number of avenues have been suggested for analyzing and altering
individuals’ attitudinal reactions to afﬁrmative action, and afﬁrmative action programs.
These have been summarized into a number of formal, testable hypotheses (see Table l).
The associated experimental manipulations are not only of academic interest, but also
hold practical, real-world implications for alleviating the many adverse reactions
to/consequences of afﬁrmative action. Below, a model is presented, linking the various
theoretical constructs outlined above. The next chapter describes means of testing these

hypotheses using both within- and between- person approaches.

59

Table 1 - Summary of Hypotheses

 

 

9a

9b

10

11

Individuals’ scores on the self-interest ethical perspective will relate to
positive attitudes toward afﬁrmative action among individuals who consider
themselves to be beneﬁciaries of these programs.

Individuals’ scores on the utilitarianism ethical perspective will relate to
negative attitudes toward afﬁrmative action.

Scores on the light of day ethical perspective will relate to positive attitudes
toward afﬁrmative action among individuals who believe that society in
general, views afﬁrmative action positively.

Individuals’ scores on the categorical imperative ethical perspective will
relate to positive attitudes toward afﬁrmative action

Individuals’ scores on the duty ethical perspective will relate to positive
attitudes toward afﬁrmative action in organizations that are supportive of
afﬁrmative action and similar policies

Individuals’ scores on the neutralization ethical perspective will relate to
negative attitudes toward afﬁrmative action in organizations that are
supportive of afﬁrmative action and similar policies

Individuals’ scores on the religious/philosophical conviction ethical
perspective will relate to positive attitudes toward afﬁrmative action
Individuals’ scores on consequentialist (teleological) ethical perspectives will
relate to negative attitudes toward afﬁrmative action

Individuals’ scores on non-consequentialist (deontological) ethical
perspectives will relate to positive attitudes toward afﬁrmative action
Making non-consequentialist (deontological) ethical perspectives more salient
to individuals will lead them to adopt more positive attitudes toward
afﬁrmative action

Individuals’ scores on equity (versus non-equity) norms of distributive justice
will relate to negative attitudes toward afﬁrmative action

Individuals’ scores on procedural (versus distributive) justice will relate to
positive attitudes toward afﬁrmative action

 

60

 

nouo< gang: e339 Beagle. me ﬁne: A charm

            
    

833. 35300.5

 

:33. 225.8...
.539 snag.

 

+ 0.68095 ﬁatomBnU

+

an E a»:

 

 

Esau.
Seaman—0:338:02

 

 

61

Chapter 2
METHOD
M

Study participants were upper level undergraduate students from a large
midwestem university. The students were enrolled in a business policy course and
participated in the experiment in order to obtain course credit.

Of approximately 600 subjects who began participation in the study, 523
completed the requirements. The sample was approximately 53% male. It was 63%
White, 4% Black, 4% Asian, 1.6% Asian-American, 0.7% Hispanic, and 2.8% of the
subjects classiﬁed themselves as “Other.” The remaining 23.9% of the sample did not
indicate their race/ethnicity.

Power Analysis

As stated earlier, the variables under study in this analysis represent deep-seated
values that are not easily manipulable. Thus I expected that the effect sizes would be
limited. Cohen and colleagues (Cohen, 1977; Cohen & Cohen, 1983), provide guidance
for estimating statistical power. Their conventions for small and medium effect sizes are
r = .10 and r = .30 respectively. Conventional alpha levels and power levels are .05 and
.80 respectively. Together, these conventions suggest a required sample size (n*) of 200
to 783. My sample size fell into this range.

Study design

The basic study design was a 2 x 2 pretest-posttest experimental design with

control (Solomon 4-group, Kerlinger, 1986; Solomon, 1949) design. As indicated earlier,

both within- and between-person effects were investigated. Subjects were randomly

62

assigned to one of four conditions. Depending on condition they were presented with a

pretest, “trained” in ethical nonconsequentialism, and tested again (see table 2 and ﬁgures

2 & 3). The tests and the training manipulation are described in greater detail below.

They are included in their entirety in Appendixes B and D respectively.

The Solomon 4-group design is unique in that it accounts for possible priming

effects by the pretest, and for other extraneous inﬂuences that might impact the results.

This is accomplished by the addition of two additional “control” groups to the traditional

two-group design. One of these groups received the training without the pretest; the other

received only the posttest (see ﬁgure 2 and table 2).

 

 

 

 

 

 

 

 

TRAINING NO TRAINING
PRETEST CONDITION 1 CONDITION 2
N0 PRETEST CONDITION 3 CONDITION 4
Figure 2 - Study Design
Table 2 - Study Conditions
CONDITION 1 Pretest Nonconsequentialism training A Post Test
CONDITION 2 Pretest Post Test
CONDITION 3 Nonconsequentialism training Post Test
CONDITION 4 Post Test

 

63

Bret—est.

The pretest was a questionnaire assessing subjects’: attitudes toward afﬁrmative
action as a concept, and toward Speciﬁc afﬁrmative action programs; perceptions of
whether afﬁrmative action is a net beneﬁt or detriment to them personally; the attitudinal
and behavioral responses that they might exhibit, or have exhibited, as a result of being a
beneﬁciary of an afﬁrmative action program; and attitudinal and behavioral responses
directed toward colleagues who (at least in the respondents’ perception) may have
beneﬁted from an afﬁrmative action program. The instrument also assessed subjects’
perspectives on general societal perceptions of afﬁrmative action, and of their employing
organizations’ support for such efforts; subjects’ ethical frameworks-in-use (independent
of the afﬁrmative action context); and their organizational justice models-in-use (also
independent of the afﬁrmative action context).

Finally, relevant subject demographic (sex, race, ethnicity, and age) and
personality data (collectivism) were collected. The pretest measures are described in
more detail below. Complete pre-test measures are included in Appendix B. Additional
details about the reliability and validity of these scales are also located in Appendix C.
Attitudewward afﬁrmative action

The subjects’ attitudes toward afﬁrmative action were assessed using a modiﬁed
version of the paper and pencil measure developed by Kravitz (1995). This is a 6-item
test with a 5-point, Likert type response scale (1=strongly disagree, 5=strongly agree). In
previous studies, this scale has demonstrated an internal consistency reliability

(Cronbach’s alpha) of .80 - .86 (Kravitz, 1995; Kravitz & Platania, 1993).

Attitudinal and behavioral outcomes associated with afﬁrmative action

This portion of the pretest asked subjects to think about afﬁrmative action in
terms of actual persons with whom they have worked, who may have been beneﬁciaries
of afﬁrmative action. This instrument is modeled after Heilman, Block, & Lucas, 1992,
Study 2. The subjects were asked to respond to items assessing their attitudes
(assessments of their competence, the extent to which they deserve their position,
interpersonal attraction), and behavior (collaborating, helping, general social interaction)
toward such person(s). Subjects who did not have any relevant work experience were
requested to complete this portion of the study in reference to a group member from a
class project, or a hypothetical co-worker. Note that subjects should not have difﬁculty
identifying such reference persons: when an organization has an afﬁrmative action policy,
both members and non-members of the target groups tend to assume that target group
members beneﬁted, at least to some extent from the afﬁrmative action policy (Garcia et
al, 1981; Heilman & Herlihy, 1984; Nacoste, 1989; 1990).

An additional section of this questionnaire was addressed to those who felt that
they may have been beneﬁciaries of afﬁrmative action programs. They were asked about
attitudes and behaviors that may be associated with being a part of these programs
(feelings of incompetence, withdrawal from difﬁcult tasks, effects on social interaction).
These responses were chosen because they reﬂect the unintended consequences of
afﬁrmative action described and demonstrated in previous research (Heilman, 1994;
Kravitz, et al., 1996).

Ethical frameworks

The ethical frameworks pretest was modeled after the two-part questionnaire

65

developed by McDonald and Pak (1996). Subjects were presented with 15 scenarios
derived from organizational/business situations in which the subjects were placed in a
position of having made decisions involving ethical considerations. For each scenario,
subjects were required to rate the degree to which they agreed with the decision using a 5-
point, Likert type scale. The second portion of the questionnaire asked respondents to
indicate the extent to which different considerations (representing different ethical
frameworks or theories) guided them in their decision making on the scenarios. The
questions in this second portion of the ethics questionnaire represent a direct query
approach to assessing subjects’ ethical frameworks. As discussed in chapter 2, the study
focused on 3 teleological (self-interest, utilitarianism, light of day) and 4 deontological
(categorical imperative, duty, neutralization, religious conviction) ethical frameworks.

McDonald and Pak (1996) also incorporated a policy-capturing component to
their analysis of subjects’ ethical frameworks. They conducted a regression analysis
comparing the subjects’ espoused ethical frameworks, to a subset of responses to the
scenarios presented earlier.7 The regression analysis revealed differences between the
subjects’ espoused ethical frameworks, and their actual frameworks-in-use. A similar
approach was used in the current study.

Additional data on subjects’ ethical frameworks were collected using the character
trait version of the Measure of Ethical Viewpoints (MEV) developed by Brady and
Wheeler (1996). This instrument assesses respondents’ predispositions toward

utilitarianism (consequentialism) and formalism (nonconsequentialism). Respondents

 

7The subset of scenario responses was identiﬁed using an exploratory factor analysis of the subjects’
responses to the ﬁrst portion of the questionnaire. One of the more robust factors from the scenario
questions was a factor labeled “deceptive practices” comprising items such as deceptive gaining of

66

rate the importance of various character traits (l = not important to me, 7 = very
important to me). The scale has demonstrated reliability (Brady & Wheeler, 1996;
Schminke et al., 1997) of .85 - .86 (utilitarianism), and .75 (formalism).

Jpstice frameworks

A multifaceted approach was taken in assessing subjects’ justice norms. In
traditional empirical justice research, subjects are queried about various aspects of actual
organizational situations. Subjects’ perceptions of procedural and distributive fairness
are then compared with other organizational variables, such as organizational
commitment (Fol ger & Konovsky, 1989 is one example). These approaches are
inappropriate for the current study in that they confound individual differences in
preferences for, or sensitivity to justice norms with the components, or perceived
components of the actual Situation.

The ﬁrst approach for assessing individuals’ justice norms is an individual
differences approach, a measure of equity sensitivity. The equity sensitivity construct
suggests that individuals react differently to perceived equity or inequity because they
have differing preferences for (are differentially sensitive to) equity (Huseman, Hatﬁeld
& Miles, 1985; 1987). Based on this construct, individuals’ preferences range on a
continuum from Benevolents to Entitleds. Benevolents are those who prefer for their
outcome/input ratios to be less than those of their comparison others. Entitleds prefer that
their outcome/input ratios be greater than those of their comparison others. Equity .
Sensitives are in the middle, opting for the traditional norm of an outcome/input ratio that

is equal to those of their comparison others.

 

competitor information, insider trading, and international bribery.

67

Given that equity sensitivity is an assessment of individuals’ preferences for
equity, it follows that subjects’ equity sensitivity score may serve as a measure of the
degree to which equity serves as their dominant justice norm. Thus one of the measures
used in this study to assess subjects’ justice norms was the Equity Sensitivity Instrument
(ESI; Huseman, Hatﬁeld & Miles, 1985). This instrument has been shown to reliably
measure equity sensitivity (Cronbach’s alpha = .79 - .88), and to be empirically distinct
from other, related constructs (King & Miles, 1994).

In addition to this individual differences approach to assessing subjects’ justice
norms, they were also assessed using direct queries and policy capturing. In the policy
capturing approach, subjects were asked to review, and respond to scenarios in which
they have the option of focusing on norms of equity, non-equity distributive justice,
and/or procedural justice concerns.

Training Manipulations

The experimental manipulation took the form of a training program, in which
subjects were trained in the use of the nonconsequentialist ethical framework. The
training was conducted within the context of various organizational problem scenarios,
unrelated to afﬁrmative action. The training was conducted using a videotaped
presentation of alternative ethical frameworks augmented by printed handouts. It
included opportunities for subjects to apply the various ethical frameworks to ﬁnd
solutions to realistic (identity-blind) organizational problems.

The training was administered to subjects either individually or in groups, based
on scheduling convenience. Subjects worked alone regardless of the training setting. The

experimenter gave all subjects time to complete the practice and application requirements

68

of the training. An outline and complete script of the training is included in Appendix D.

Kavathatzopoulos (1993) demonstrated that this kind of scenario-based training
within a realistic organizational context can effectively induce a shift in subjects’ ethical
problem solving approaches. In addition, the scenario-based approach was demonstrated
to be more effective than more traditional approaches employing lectures on ethical
principles. In the Kavathatzopoulos (1993) study subjects’ were induced to adopt an
autonomous moral function, a pattern of moral reasoning very similar to consequentialist
ethics.

This approach is also supported by studies based on the seminal work of Kohlberg
and his colleagues (such as Kohlberg, 1969; 1977). These studies (Goldman &
Arbuthnot, 1979 is one example) have Shown that it is possible to use scenario-based
training to move individuals to higher stages of moral reasoning. Kohlberg's work is
based on the notion that individuals progress through stages of moral development that
parallel their intellectual development. Individuals move from simple to more complex
patterns of moral reasoning as they progress from one stage, to the next. At Simpler and
more child-like levels, individuals use simple reasoning patterns focusing for example, on
self-interest or the consequences of their actions as indicators of morality. At more
advanced stages, moral reasoning becomes more complex and moral decisions are
informed by more abstract concepts such as individual rights and universal moral
principles. Note that these stages bear some resemblance to consequentialist and
nonconsequentialist ethical frameworks respectively. It follows then that a similar
scenario-based approach is appropriate for the training manipulation used in the present

study.

69

«map.

cwmmoﬂ RecoEtcmxm 23m .. m can—mi

 

 

Emamzcoscogooéoz
E wEEEH

 

 

 

 

mcoEmommEBm
853. can 335m

sauce. gaunt—be.
2.39 833.

 

~m~é

 

 

 

 

 

mcoummommmeocm
353. 28 325m

eouo< gums—cg
e338 3832

 

 

70

Data Analysis

All of the data analyses for this study were conducted using ordinary least squares
(OLS) regression (Cohen & Cohen, 1983). Hypotheses 1 through 9a were investigated
using analyses of both pre- and post-test data. Between-person effects were assessed
using pre- and post—test data as separate samples. For both data sets, attitudes toward
afﬁrmative action served as the primary dependent variable. A variety of regressions
were conducted assessing the main effects of subjects’ (espoused and demonstrated)
ethical frameworks (see Table 5). For hypothesis 1 subjects’ assessments of whether or
not they expect to beneﬁt from afﬁrmative action served as moderator. For hypotheses 3
subjects’ assessments of society’s perspective on afﬁrmative action served as moderator.
For hypotheses 5 and 6 subjects’ assessments of their organizations’ support of
afﬁrmative action served as moderator.

Hypothesis 9b is the primary experimental hypothesis. The effect of the
experiment was assessed using regression. Subjects’ attitudes toward afﬁrmative action
again, served as dependent variable. This within-person analysis will address attitudes
before and after the manipulation.

The analyses of hypotheses 10 and 11 took on a similar form as those for

hypotheses 2 and 4.

71

Table 3 - Variable List

 

 

\OWQGUI1hUJN—t

dpscale
doscale
sescale
siscale
nscale
dtyscale
jscale
rcscale
ldscale
uscale
ciscale
conscale
nonscale
attscale
famscale
orgscale

impscale
benscale
hrrnscale
dstscale
proscale
eqlscale
neescale
eqscale
colscale
rolscale
pcscale
r12scale
pc2scale
blkscale

Deceptive practices

Disadvantaging others

Sexual equity

Self interest

Neutralization

Duty

Justice

Religious conviction

Light of day

Utilitarianism

Categorical imperative

Consequentialism

Nonconsequentialism

Attitude toward afﬁrmative action

Familiarity with afﬁrmative action

Employer’s support of afﬁrmative action/ Importance of afﬁrmative
action to my organization

Perceived potential impact of afﬁrmative action on my job
Likelihood that I will beneﬁt from afﬁrmative action
Likelihood that I will be banned by afﬁrmative action
Concern for distributive justice

Concern for procedural justice

Concern for equality

Concern for need

Concern for equity

Collectivism

Role of afﬁrmative action in selection of coworker
Perceived competence of afﬁrmative action coworker
Role of afﬁrmative action in my career

Perceived competence — myself

Attitude toward afﬁrmative action for African-Americans

 

72

Chapter 3
RESULTS
Descriptive Statistics

The means, standard deviations, range of scores and number of respondents are
Shown in Table 4. A total of 30 variables were measured. Table 43 provides descriptive
statistics for the pretest, and Table 4b for the posttest. Neither set of data presents any
unusual ﬁndings. The total sample size for the pretest was 272, and for the posttest it was
523. In addition, a complete correlation matrix is shown in Table 5. Table 5a Shows zero-
order intercorrelations for the pretest data, Table 5b for the posttest data. The effect of the
pretest was assessed using a dummy coded variable added to the posttest data. However,
due to the nature of the hypotheses and the lack of an observable priming effect from the
pretest, no further analyses were conducted on the pretest data. The analyses and results
discussed below were all conducted on the posttest data.

Internal consistency reliabilities (coefﬁcient alphas) are shown in Table 6. Of the
thirty variables that were measured 11 displayed internal consistency reliabilities of .70 or
greater (alpha = .70 - .96). Another eight had reliabilities ranging from .50 - .69. Six of
the variables were one-item scales so no reliabilities could be computed. The remaining
ﬁve had internal consistency estimates ranging from .28 to .49.

Factor Analysis of Dependent Variables

In order to further assess the internal consistency of the primary dependent
variables (those assessing subjects’ attitude toward afﬁrmative action) factor analyses
were conducted (see Table 7). Using orthogonal rotation and a minimum eigenvalue of

1.0, four factors emerged among the items that were designed to assess these variables.

73

Items were assigned to the factors where they displayed the highest loading, with a
minimum factor loading of .30. Seven of the eight items designed to measure Attitude
toward Afﬁrmative Action loaded on the same factor. All ﬁve of the items designed to
measure Perceived Competence loaded on the same factor. Five of the six items designed
to measure Attitude toward Afﬁrmative Action for African-Americans loaded on the
same factor. See Table 7 for further details.

Based on the factor analysis, two new Attitude scales were developed. The
Attitude items that did not load as expected were removed from the Attitude toward
Afﬁrmative Action and Attitude toward Afﬁrmative Action for African Americans scales.
Their internal consistency reliabilities and correlations with other variables were assessed
and compared with those of the original variables. The results were not meaningfully
different, so no further analyses were conducted using these variables. See Table 6 and

Table 9b for additional details.

74

Table 4a - Pretest Descriptives

 

 

M SD Minimum Maximum N .
(1) Deceptive practices 3.21 .61 1.50 4.83 272
(2) Disadvantaging others 2 . 3 1 . . 66 1 . 00 4 . 67 272
(3) Sexual equality 1.90 .98 1.00 5.00 272
(4) Self interest ethical 3 - 25 - 57 1 . 00 5 - 50 270
framework
(5) Neutralization 2 . 60 . 73 1 . 00 4 . 75 270
(6) Duty ethical framework 3 . 55 . 76 1 . 00 5 . 00 270
(7) Justice ethical framework 3 - 85 - 69 1 . 00 S . 0 0 270
(8) Religious convictioo 2 . 65 1 . 05 1 . 00 5 . 00 270
(9) Lightofday 2.76 .90 1.00 5.00 270
(10) Utilitarianism 3.37 .66 1.00 5.00 270
(11) Categorical imperative 3.89 .78 1.00 5.00 270
(12) Consequentialism 5.55 .91 1 .50 7.00 272
(13) Nonconsequentialism 5 . 68 . 92 1 . 7O 7 . 10 272
(14) Attitude toward 4.13 1.27 1.13 6.75 271
afﬁrmative action
(15) Familiaritywith 5.25 1.30 1.00 7.00 269
afﬁrmative action
(16) Subject’s organization’s 3 - 32 . 96 1 . 00 S . 00 265
support for afﬁrmative
action
(17) Potential impactof 4.80 1.57 1.00 8.00 271
afﬁrmative action on
subject’s career
(18) Likelihood that subject 2.89 1.82 1.00 7.00 271
will beneﬁt from
afﬁrmative action
(19) Likelihood that subject 4.07 1.63 1.00 7.00 271
will be harmed by
afﬁrmative action
(20) Concern for distributive 2 - 36 - 85 1 . 00 4 . 50 2'71
justice
(21) Concern for procedural 3 - 01 .82 1 .00 5 .50 272
justice
(22) Equality-based 2 . 73 . 90 1 . 00 5 . 00 272
distributive justice
standard
(23) Need-based distributive 2 . 62 . 75 1 . 00 4 . 67 272
justice standard
(24) Equity-based distributive 4 - 08 . 58 1 . 2 5 5 . 2 5 272
justice standard
(25) Collectivism 3 . 27 . 44 1 . 73 4 . 73 272
(26) Role of afﬁrmative action 3 - 32 1 . 84 1 . 00 7 . 00 195
in co-worker’s career
(27) Perceived competence of 5 - 13 1 . 27 1 . 00 7 . 0 0 193
co-worker

 

75

Table 4a - Pretest Descriptives (cont’d).

 

(28) Role of afﬁrmative action 2 . 04 1 . 59 1 . 00 7 . 00 271
in my career

(29) Perceived competence of 5 - 98 - 94 2 . 2 0 7 . 0 0 272
self

(30) Attitude toward 3 . 44 . 88 . O O 5 . O O 271

afﬁrmative action for
African-Americans

76

Table 4b - Posttest Descriptives

 

 

M SD Minimum Maximum N
(l) Deceptive practices 3.10 .67 1.00 4.83 523
(2) Disadvantaging others 2 . 32 . 65 1 . 00 4 . 67 523
(3) Sexual equality 1.93 .96 1.00 5.00 522
(4) Self interest ethical 3 - 2 1 - 67 1 - 0 0 4 . 75 52 1
framework
(5) Neutralization 2 . 65 . 75 1 . 00 4 . 50 520
(6) Duty ethical framework 3 . 50 - 72 1 . 2 5 S . 00 52 l
(7) Justice ethical framework 3 - 83 . 63 1 . 50 5 . 00 522
(8) Religious conviction 2 - 69 - 99 1 . 00 5 . 00 521
(9) Lightofday 2.86 .81 1.00 5.00 522
(10) Utilitarianism 3.31 .66 1.00 5.00 522
(11) Categorical imperative 3 - 89 - 72 1 . 50 5 . 25 521
(12) Consequentialism 5.36 1 .11 1 .00 7.00 522
(13) Nonconsequentialism 5 . 51 1 . 18 1 . 00 7 . 00 522
(14) Attitude toward 4.13 1.23 1.00 6.88 522
afﬁrmative action
(15) Familiarity with 5.04 1.36 1.00 7.00 521
afﬁrmative action
(16) Subject’s organization’s 3 ~ 27 - 91 1 . 00 5 . 00 512
support for afﬁrmative
action
(17) Potential impact of 4.62 1.69 1.00 7.00 520
afﬁrmative action on
subject’s career
(18) Likelihood that subject 3.08 1.82 1.00 7.00 519
will beneﬁt from
afﬁrmative action
(19) Likelihood that subject 4.00 1.53 1.00 7.00 511
will be harmed by
afﬁrmative action
(20) Concern for distributive 3 - 0 0 . 8 5 1 . 0 0 5 . O 0 5 19
justice
(21) Concern for procedural 3-00 .82 1.00 5.00 522
justice
(22) Equality-based 2.77 .81 1.00 4.75 522
distributive justice
standard
(23) Need-based distributive 2 . 68 . 74 1 . 00 5 . 00 522
justice standard
(24) Equity-based distributive 3 - 98 . 67 1 . 50 5 . 50 522
justice standard
(25) Collectivism 3 . 23 . 42 1 . 73 4 . 55 522
(26) Role of afﬁrmative action 3 - 53 1 . 79 1 . 00 7 . 00 373
in co-worker’s career
(27) Perceived competence of 5 . 02 1 - 2 S 1 . 00 7 . 00 371
co-worker

 

77

Table 4b - Posttest Descriptives (cont’d).

 

(28)
(29)

(30)

Role of afﬁrmative action 2 . 19
in my career

Perceived competence of 5 - 8 9
self

Attitude toward 3 . 4 5
afﬁrmative action for
African-Americans

1

.66

.93

.85

78

1.

1

1.

00

.60

00

.00

.00

.00

517

517

516

Table 5a - Pretest Correlations

 

 

(1) (2) (3) (4) (5) (6) (7) (8) (9)

(1) dpscale

(2) doscale .30‘

(3) sescale .27‘ .40*

(4) siscale .30* .14* . 19*

(5) nscale .36* .44* .46* .38*

(6) dtyscale . 19* -.03 -.01 .42* . 16*

(7) jscale -.17* -.18* -.28* .31* -.12t .35*

(8) rcscale -.21* .02 .1 1t .00 .03 -. 13* .15*

(9) ldscale .01 .30* . 13* .25* .25* -.00 .06 .28*
(10) uscale .19 .07 .14* .47* .33* .45* .45* .03 . 14*
(11) ciscale -.26* -.24* -.19‘ .15* -.17* .27* .66* .29* .13*
(12) conscale —.09 -.09 -.06 .14* -.08 .18* .18* .10t -.02
(13) nonscale -.25* -.20* -.23* .03 -.22* .1 1t .32* . 15* -.01
(14) attscale -. 16* -.02 -.25* -.09 -. 10 -.02 .13* .17* .02
(15) famscale -.05 -.1 1t -.01 .00 -.07 -.02 .06 -.00 .02
(16) orgscale -. 15* -.07 -. 18* -.04 -.19* -.03 .16* .02 .00
(17) impscale .00 -.03 .06 .01 .05 .04 -.02 -.03 .08
(18) benscale -. 15* -.01 -.1 1t -.01 -.06 -.05 .04 .17* .09
(19) hrmscale .10t .09 .25* .10t .21 * .07 -.09 -.12t .02
(20) dstscale .05 .28 .26* .20* .30* . 1 1t .04 .09 .08
(21 ) proscale -.1 1t -.04 .03 .09 .01 .04 .14* .12t .09
(22) eqlscale -. 16* .14* -.O3 .06 .12* .07 .09 .10t . 15*
(23) neescale .09 .29* .26* .17* .28* . 13* -.01 .23* . 15*
(24) eqscale .07 -.06 -.01 . 15* .01 .21* .27 * -.06 -.02
(25) colscale -.04 -. 15* -. 16* -.03 -. 12* .1 1t . 16* .09 -.00
(26) rolscale -.10 .09 .10 -.05 .13t .01 .01 -. 10 .02
(27) pcscale .12 -.15* 316* -.01 -.14t -.05 .04 .1 1 -.01
(28) rlZscale -.05 .21* .18* .04 .20* -.04 -. 10 . 19* .08
(29) pc25cale -. 12* -.29* -.25* .05 -.22* .09 . 19* .01 -.01
(30) blkscale -.20* -. 13* -.39* -.20* -.25* -.06 . 17* . 16* -.09

 

N = 272.‘p <.10; * p < .05.

79

Table 5a, (cont’d). — Pretest Correlations

 

 

(10) (11) (12) (13) (14) (15) (16) (17) (18)
(10) uscale
(l 1) ciscale .29*
(12) conscale .17* . 16*
(13) nonscale .15* .35* .81*
(14) attscale -.08 .06 . 16* .21 *
(15) famscale .02 .06 .1 1t .10t -. 10
(16) orgscale -.03 .10 . 10 .17* .25* .20*
(17) impscale .06 -.06 .07 -.00 -.01 .05 .09
(18) benscale -.06 -.01 .12* .12* .44* -.14* .11t .28*
(19) hrmscale . 16* -.08 -.01 -.07 -.37* .14* -.08 .34* -.35*
(20) dstscale .20* .06 .06 -.06 -.08 -.00 -.08 -.04 —.05
(21) proscale .08 .24* .07 .09 .00 -.04 .00 .07 .03
(22) eqlscale . 10 .14* -.07 -.02 .20* -.03 -.03 -.03 .14*
(23) neescale .12* .03 -.08 -.12t -.01 -.O3 - 16* -.03 .04
(24) eqscale .24* .18* . 16* .12t .02 .05 . 16* -.02 -.14*
(25) colscale .05 .20* .06 . 13* .12t .02 .1 1t .02 .01
(26) rolscale . 1 l -.00 -.02 -.05 -.09 -.00 .10 .08 .02
(27) pcscale -.10 .07 .06 .06 . 12t .01 .03 -.13t .06
(28) r12scale .02 -.07 -.07 -. 12* .12* -. 13* .02 .12t .33*
(29) pc2scale -.00 .24* . 18* .18* -.04 .20* . 13* .02 .00
(30) blkscale -.12* . 13* . 19* .27* .74* -.01 .27* -.02 .38*

 

N = 272.‘p <.10; * p < .05.

80

Table 5a, (cont’d). - Pretest Correlations

 

 

(19) (20) (21) (22) (23) (24) (25) (26) (27)
(19) hrmscale
(20) dstscale .1 2t
(21) proscale .08 .08
(22) eqlscale -.1 1t .10 .19*
(23) neescale .03 .25* . 16* .29*
(24) eqscale .04 .09 .05 -.09 -.O6
(25) colscale -.02 -.05 .01 -.02 -.07 . 10
(26) rolscale .12t .14t .06 -.04 -.05 -.12t -.15*
(27) pcscale -.22* -.10 -.03 .05 .09 .01 .13t -.54*
(28) rl2scale -.11t .13* .02 .08 .17* -.11t -.07 .24* .01
(29) pc2scale -.03 -.07 -.00 -.09 -.17* .16* .09 -.13t .19*
(30) blkscale -.37* —.08 .01 .14* -.03 .08 .15* -.14t .19*

 

N = 272.‘p <.10; * p < .05.

Table 5a, (cont’d). - Pretest Correlations

 

 

(28) (29) (30)
(28) r125cale
(29) pc2scale -.25*
(30) blkscale .12t .03

 

N = 272.‘p <.10; * p < .05.

81

Table 5b - Posttest Correlations

 

 

(1) (2) (3) (4) (5) (6) (7) (8) (9)

(1) dpscale

(2) doscale .37*

(3) sescale .27* .49*

(4) siscale .42* .31* .25*

(5) nscale .43* .47* .41 * .47*

(6) dtyscale .23*.01.03 .44* . 13*

(7) jscale -.15* - .40* - .37* .02 -.25* .26*

(8) rcscale - .25* -.09* -.02 -.14* -.13* -.08t .14*

(9) ldscale .01 - .07t .07t .24* .13* .06 .03 .21*
(10) uscale .29* .09* .10* .40* .29* .39* .22* -.05 .15*
(11) ciscale -.13* - .37* - .27* -.01 -.25* .18* .60* .30* .15*
(12) conscale .08t - .06 - .04 .17* .04 .31 * .24* .04 -.01
(13) nonscale -.05 - .22* - 2.0* .02 -.15* .23* .40* .14* -.06
(14) attscale -.17* -07t -.22* -.06 -.09* .02 .12* .19* .06
(15) famscale .03 -.15* -.03 .00 -.05 -.04 .10* .02 -.06
(16) orgscale -.14* -.18* —.13* -.11* -.19* -.04 .15* -.00 -.04
(17) impscale .02 -.03 —.02 .10* .10* .1 1* .05 .07t .13*
(18) benscale -.10* .08t -.05 -.01.01 - .06 -.06 .11* .10*
(19) hrmscale .12* .02 .12* .09* .12* .14* -.02 -.01 .09t
(20) dstscale 2.0* .22* .20* 2.4* .29* .06 -.16* - .06 .03
(21) proscale - .00 .04 .12* .04 -.05 -.03 -.03 .13* .05
(22) eqlscale - 0.8t .1 1* -.02 —.01.07 -.02 - .04 .14* .04
(23) neescale .11* .21* .19* .12* .27* - .04 - .16* .09* .07t
(24) eqscale .06 -.24* -.15* .1 1* -.10* .18* .33* -.01 -.09*
(25) colscale .06 -.17* -.21* -.09* -.14* .11* .18* .01 -.11*
(26) rolscale .01 .14* .14* .06 .17* -.06 -.12* .05 .06
(27) pcscale -.08 -.20* -.26* -.10t -.17* .03 . 16* -.01 -.1 1*
(28) rl2scale -.07 .19* .17* .02 .15* -. 13* -.20* .13* .08t
(29) pc28cale .07t -.20* -.18* .06 -.14* .17* .26* -.07 -. 10*
(30) blkscale -.13* -. 16* -.27* -.12* -. 15* -.02 . 16* -15* -.04

 

N = 523.'p<.10; * p < .05.

82

Table 5b, (cont’d). - Posttest Correlations

 

 

(10) (11) (12) (13) (14) (151 LIGL (17) (18)
(10) nscale
(1]) ciscale .17*
(12) conscale .21* .24*
(13) nonscale .13* .40* .85*
(14) attscale .01 .13* .17* .21*
( 15) famscale .07t .06 .03 .06 -.O8t
(16) orgscale —.01 .11* .04 .08t .18*
(17) impscale .1 1* .10* .1 1* .08t .12* .08t
(18) benscale -.04 -.01 .01 -.01 .40* .05 .32*
(19) hrmscale .09t .04 .07 .03 -.35* -.10* .23* -.29*
(20) dstscale .10* -.1 1 * -.02 -.09* -.06 -.12* -.02 .00
(21) proscale .05 .10* .01 .02 .05 -.01 -.03 .00
(22) eqlscale -.05 .06 .01 .02 .14* -.05 .07t .16*
(23) neescale .02 -.06 .05 -.02 -.02 -.17* .04 .19*
(24) eqscale .19* .31* .27* .3 1 * .08t .10* .02 -.08t
(25) colscale .08t .15* .1 1* .18* .04 .07 .05 -.10*
(26) rolscale -.02 -. 14* -.08 -.1 1* -.13* -.13* .08 .03
(27) pcscale -.04 .14* .06 .09t .19* .20* -.08 .02
(28) rl2scale -.05 -.14* -.15* -.15* .20* -.l3* .12* .43*
(29) pc2scale .1 1* .19* .17* .20* -.01 .27* .02 -. 19*
(30) blkscale -.01 .17* .14* .22* .73* .26* .08t .29*

 

N = 523.‘p<.10;* p< .05.

83

Table 5b, (cont’d). - Posttest Correlations

 

 

(19) (20) (21) (22) (23) (24) (25) (26) (27)
(19) hrmscale
(20) dstscale .09*
(21) proscale -.03 -.05
(22) eqlscale -.04 .08t .20*
(23) neescale .05 .20* .10* .3 1 *
(24) eqscale .02 .07t .03 -. 1 3* -.08t
(25) colscale .06 -.12* .02 -.13* -.15* . 16*
(26) rolscale .07 .09t -.07 -.02 .08 -.10* —.l2*
(27) pcscale -.11* -.04 -.03 .00 -.08 .10t .14* -.37*
(28) r12scale -.19* .08t .02 .13* .21* -.20* -.15* .32* -.07
(29) pc2scale -.00 -.13* -.02 -.17* -.24* .25* .24* -.18* .34*
(30) blkscale -.29* -.07t .04 .03 -.03 .17* .14* -.16* .24*

 

N = 523.‘p<.10;* p < .05.

Table 5b, (cont’d). - Posttest Correlations

 

 

(28) (29) (30)
(28) r12scale
(29) pc2scale -.30*
(30) blkscale .07 .08t

 

tp<.10; * p< .05.

84

Table 6 - Scale Reliabilities (Internal Consistency Estimates)

 

 

(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(1 1)
(12)
(13)
(14)

(15)
(16)

(17)

(18)
(19)

(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)

(28)
(29)
(30)

Note. Reliabilities were assessed based on internal consistency (coefﬁcient alpha). Scales

Deceptive practices

Disadvantaging others

Sexual equity

Self interest

Neutralization

Duty

Justice

Religious conviction

Light of day

Utilitarianism

Categorical imperative

Consequentialism

Nonconsequentialism

Attitude toward afﬁrmative action

Attitude toward afﬁrmative action (modiﬁed)
Familiarity with afﬁrmative action

Employer’s support of afﬁrmative action/
Importance of afﬁrmative action to my organization
Perceived potential impact of afﬁrmative action on
my job

Likelihood that I will beneﬁt from afﬁrmative action
Likelihood that I will be harmed by afﬁrmative
action

Concern for distributive justice

Concern for procedural justice

Concern for equality

Concern for need

Concern for equity

Collectivism

Role of afﬁrmative action in selection of coworker
Perceived competence of afﬁrmative action
coworker

Role of afﬁrmative action in my career

Perceived competence — myself

Attitude toward afﬁrmative action for African-
Americans

Attitude toward afﬁrmative action for African-
Americans (modiﬁed)

where reliabilities are not recorded are one-item scales.

85

dpscale
doscale
sescale
siscale
nscale
dtyscale
jscale
rcscale
ldscale
uscale
ciscale
conscale
nonscale
attscale
attscal2
famscale
orgscale

impscale

benscale
hrmscale

dstscale
proscale
eqlscale
neescale
eqscale

colscale
rolscale
pcscale

rlZscale
pc2scale
blkscale

blkscal2

.64
.61
.59
.46
.60
.71
.53
.86
.70
.56
.68
.93
.95
.92
.96

.72

.28
.30
.67
.42
.71
.49

.88

.73
.85

.83

Table 7 - Rotated Factor Matrix — Dependent Variables

 

Item Factor] Factor 2 Factor 3 Factor4

 

71 .87 .10 .20 .05
69 .87 .08 .30 .01
73 .86 .09 .29 m00
70 .86 .06 .32 .02
72 .86 .06 .25 .04
68 .85 .03 .34 .04
125 .63 .07 .51 .14
74 .61 .02 .08 .44
114r .03 .88 .17 .03
115r .06 .88 .15 .07
117r .11 .85 .04 a07
118r .16 .84 .07 .03
116r a03 .68 n00 .01
126r .23 .15 .72 .05
129 .25 .09 .72 a07
127 .29 .06 .72 208
130r .25 .ll .65 .28
128 .28 .02 .62 .09
75r .07 .03 .08 .93

 

' Reverse coded items

Nate, Actual items are shown in Appendix B.

Factor 1 corresponds to the attitude toward afﬁrmative action scale.

Factor 2 corresponds to the perceived competence scale.

Factor 3 corresponds to the attitude toward afﬁrmative action for African-Americans
scale.

Factor 4 is a single item assessing the subjects’ View of the societal perception of
afﬁrmative action. It was designed to serve as an indirect measure of the subjects’
attitudes but did not load on the attitude scale.

86

Hyp_otheses Tests

All of the hypotheses in this dissertation were directed toward assessing
individuals’ attitudes toward afﬁrmative action. Three dependent variables were used to
assess this construct. Two of these were direct measures of subjects’ attitudes (the
Attitude toward Afﬁrmative Action and Attitude toward Afﬁrmative Action for African
Americans variables). The Perceived Competence variable served as an indirect measure
of attitude. It was expected that subjects’ affective reactions to afﬁrmative action would
be reﬂected in their assessments of the competence level of individuals they perceive to
have beneﬁted from these initiatives (see Heilman, 1994; Nacoste, 1990).

Seven of the eleven hypotheses were conceived as simple linear relationships
between variables. As such they were assessed based on the zero-order correlations
between them. The other four hypotheses were conceived as conditional relationships and
therefore assessed using moderated regression. The results of the tests for the linear

relationships are shown in Table 8. The moderated regressions are shown in Tables 9-11.

87

Table 8a - Hypotheses Tests (Zero-Order Intercorrelations)8

 

 

Attitude toward Perceived Attitude toward

Afﬁrmative Action Competence Afﬁrmative Action

for African

Americans

Hyp.
1 Self Interest -.06 -.10t -.12*
2 Utilitarianism (-) .01 -.04 -.01
3 Light of Day9 .06 -.11* -.04
4 Categorical .13* .14* .17*
Imperative

5 Duty .02 .03 -.02
6 Neutralization (-) -.09* -.17* -.15*
7 Religious Conviction .19* -.01 .15*
8 Consequentialism (-) .17* .06 .14*
9a Non-Consequentialism .21 * .09t .22*
9b Video -.05 .06 -.03
10 Equity (-) .08t .10t .17*
ll Procedural Justice .05 -.03 .04

 

N=523.‘p<.10;*p<.05.

8 The results for hypotheses 1, 3, 5 and 6 are not substantively meaningful here because the hypotheses were
gonceived with moderators.

Hypothesis 3 was developed with the societal view of afﬁrmative action as a moderator. However the
hYPOthesis cannot be assessed because the societal view of afﬁrmative action was not measured.

88

Table 8b - Hypotheses Tests (Zero-Order Intercorrelations) with Revised Dependent

 

 

Variables")
Attitude toward Perceived Attitude toward

Afﬁrmative Action Competence Afﬁrmative Action

for African

Americans

Hyp.
1 Self Interest -.08t -.10t -.l 1*
2 Utilitarianism (-) -.01 -.04 -.02
3 Light of Day” .06 -.11* -.05
4 Categorical .14* .14* .17*
Imperative

5 Duty .01 .03 -.03
6 Neutralization (-) -.10* -.17* -.14*
7 Religious Conviction .21* -.01 .16*
8 Consequentialism (-) .15* .06 .13*
9a Non-Consequentialism .20* .09t .20*
9b Video -.04 .06 -.04
10 Equity (-) .08t .10t .15*
11 Procedural Justice .04 -.03 .05

 

N=523.‘p< .10; * p< .05.

Hypothesis 1 assessed the relationship between the self-interest ethical framework
and attitude toward afﬁrmative action. It predicted that subjects’ scores on the self-
interest variable would be positively related to their attitudes toward afﬁrmative action if
they considered themselves to be potential beneﬁciaries of afﬁrmative action efforts. This
relationship was assessed using multiple regression, with the Likelihood that I will
Beneﬁt from Afﬁrmative Action variable as the moderator. Both the interaction (Beta =

.42; p < .05) and the overall model (Adj. R2 = .16; p < .05) were signiﬁcant for the

 

'0 The Attitude toward Afﬁrmative Action and Attitude toward Afﬁrmative Action for African-Americans
variables were modiﬁed based on the factor analyses. Items that did not load on the scales were removed.
The change had a negligible effect on the variables in terms of their reliabilities and their relationships with
the predictors. No further analyses were conducted with these revised dependent variables.

" Hypothesis 3 was developed with the societal View of afﬁrmative action as a moderator. However the
hypothesis cannot be assessed because the societal view of afﬁrmative action was not measured.

89

Attitude toward Afﬁrmative Action measure (see Table 8). Although signiﬁcant results

were not obtained for Perceived Competence or Attitude toward Afﬁrmative Action for

African-Americans because Of the robustness Of the general Attitude variable, these

results provide support for Hypothesis 1.

Table 9 - Moderated Hypotheses Tests - Hypothesis 1

 

 

Step Attitude Perceived Attitude toward
toward AA Competence AA for African-
Americans
] Beneﬁt
Beta .39 .02 .28
T 9.72 .46 6.70
Sig T .00 .65 .00
R2 .16 .01 .09
Adj. R2 .16 .01 .09
2 Self Interest
Beta -.06 -.10 -.12
T -1.52 -2.99 -2.76
Sig T .13 .48 .01
R2 .16 .01 .09
Adj. R2 .16 .01 .09
3 Self Interest x Beneﬁt
Beta .42 .28 .02
T 2.20 1.20 .13
Sig T .03 .23 .90
R2 .17 .02 .09
Adj. R2 .16 .01 .09
Overall F 34.13" 1.86 17.56’

 

Note. +p< .10; 'p < .05

Hypothesis 2 predicted that the Utilitarianism ethical perspective would be

negatively related to attitudes toward afﬁrmative action. This relationship was assessed

using the zero order correlation between the utilitarianism variable and the three

dependent variables. The correlations ranged from r = -.04 tO r = .01 and were not

90

signiﬁcant (see Table 8a). Hypothesis 2 was not supported.

Hypothesis 3 predicted that the Light Of Day ethical perspective would be
positively related to attitudes toward afﬁrmative action for subjects who perceive that
society has a positive perception of afﬁrmative action. Unfortunately, the societal
perception variable was not properly measured. This hypothesis could not be tested.

Hypothesis 4 predicted that the Categorical Imperative ethical perspective would
be positively related tO attitudes toward afﬁrmative action. The zero-order correlations for
this relationship ranged from r = .13 to r = .17 and were all Signiﬁcant (p < .05; see table
8a). Hypothesis 4 was supported.

Hypothesis 5 predicted that the Duty ethical perspective would be positively
related to attitudes toward afﬁrmative action for subjects who perceive that their
organizations are supportive of afﬁrmative action. This relationship was assessed using
multiple regression, with the Employer’s Support of Afﬁrmative Action variable as the
moderator. The adjusted R2 for this model ranged from .03 to .06 for the three dependent
variables (see Table 10). The Betas for the interaction term were not signiﬁcant for either
Of the direct attitude measures. The relationship with Perceived Competence was
signiﬁcant (Beta = -.50; p < .10) but in the wrong direction. Hypothesis 5 was not

supported.

91

Table 10 - Moderated Hypotheses Tests — Hypothesis 5

 

 

Step Attitude Perceived Attitude toward AA
toward AA Competence for African-
Americans
1 Organizational Support
Beta .18 .20 .25
T 4.07 3.82 5.87
Sig T .00 .00 .00
R2 .03 .04 .06
Adj. R2 .03 .03 .06
2 Duty
Beta .03 .04 -.01
T .60 .72 -.32
Sig T .55 .48 .75
R2 .03 .04 .06
Adj. R2 .03 .03 .06
3 Duty x Org. Support
Beta .03 -.50 .24
T . 15 -1.84 1.03
Sig T .88 .07 .30
R2 .03 .05 .07
Adj. R2 .03 .04 .06
Overall F 5.58" 6.16‘ 11.95’

 

Note.+p<.10;'p<.05

Hypothesis 6 predicted that the Neutralization ethical perspective would be
negatively related to attitudes toward afﬁrmative action for subjects who perceive that
their organizations are supportive Of afﬁrmative action. Like hypothesis 5, this
relationship was also assessed using multiple regression, with the Employer’s Support Of
Afﬁrmative Action variable as the moderator. The adjusted R2 for this model ranged from
.03 to .07 for the three dependent variables (see Table 11). The Betas for the interaction
term were not signiﬁcant for any of the dependent variables. Hypothesis 6 was not

supported.

92

Table 11 - Moderated Hypotheses Tests — Hypothesis 6

 

 

Step Attitude Perceived Attitude toward
toward AA Competence AA for African-
Americans
1 Organizational Support
Beta .17 .17 .24
T 3.79 3.21 5.45
Sig T .00 .00 .00
R2 .04 .06 .07
Adj. R2 .03 .05 .07
2 Neutralization
Beta -.06 -. 1 3 -.09
T -1.30 -2.54 -2.04
Sig T .19 .01 .04
R2 .04 .06 .07
Adj. R2 .03 .05 .07
3 Neutralization x Org.
Support .05 .18 .30
Beta .27 .79 1.54
T .79 .43 .13
Sig T .04 .06 .08
R2 .03 .05 .07
Adj. R2
Overall F 6.19‘ 7.12’ 13.98’

 

Note. ‘p< .10; ‘p< .05

93

Hypothesis 7 predicted that the Religious Conviction ethical perspective would be
positively related to attitudes toward afﬁrmative action. This hypothesis was conceived as
another simple linear relationship with the three dependent variables. Correlation
coefﬁcients for both Of the direct attitude measures were signiﬁcant (p < .05) and in the
expected direction (I = .15 and .19). The correlation with the Perceived Competence
measure was not signiﬁcant (see Table 8a). These results provide support for Hypothesis
7.

Hypothesis 8 predicted that the Consequentialism ethical perspective would be
negatively related to attitudes toward afﬁrmative action. This relationship was assessed
using the zero-order correlations with the dependent variables. The relationship with
perceived competence (r = .06) was not signiﬁcant. Interestingly, the relationship between
Consequentialism and both Of the direct attitude measures was signiﬁcant but positive (I
= .17 and .18; see Table 8a). This unexpected ﬁnding will be explored in more detail in

the next chapter.

94

Hypothesis 9 was the focus Of the dissertation experiment. It was predicted that
the Nonconsequentialism ethical perspective would be positively related to attitudes
toward afﬁrmative action (Hypothesis 9a). In addition it was expected that the
experimental manipulation would move subjects to adopt more Of a nonconsequentialist
ethical perspective and as such, become more positively disposed toward afﬁrmative
action (Hypothesis 9b). The zero order correlation between nonconsequentialism and the
two attitude measures was signiﬁcant and relatively robust (r = .21 and .22; p < .05). In
addition there was a less robust relationship with perceived competence (r = .09; p < .10).
These results Show strong support for Hypothesis 9a (see Table 8a).

In spite of the strong relationship between nonconsequentialism and attitudes
toward afﬁrmative action, the experimental manipulation did not have the expected
effect. The experiment was designed to induce higher scores on nonconsequentialism.
Instead, it had a marginally signiﬁcant negative relationship with nonconsequentialism (r
= -.08, p < .10; see Table 12). Interestingly the video manipulation also had a signiﬁcant
negative relationship with consequentialism (r = -.09, p < .05) and utilitarianism (r = -.09,
p < .05). Although not predicted, this relationship is consistent with the theoretical
framework. This will be discussed in more detail in the next chapter. The video
manipulation variable did not have a signiﬁcant relationship with attitudes toward

afﬁrmative action. As such, Hypothesis 9b was not supported

95

Table 12 - Effect of Video Manipulation (Zero-Order Intercorrelations)

 

Correlation with Video

Manipulation

 

(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(1 1)
(12)
(13)
(14)
(15)
(16)

(17)
(18)
(19)

(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)

Deceptive practices
Disadvantaging Others

Sexual equality

Self interest ethical framework
Neutralization

Duty ethical framework

Justice ethical framework
Religious conviction

Light Of day

Utilitarianism

Categorical imperative
Consequentialism
Nonconsequentialism

Attitude toward afﬁrmative action
Familiarity with afﬁrmative action

Subject’s organization’s support for afﬁrmative
action

Potential impact Of afﬁrmative action on
subject’s career

Likelihood that subject will beneﬁt from
afﬁrmative action

Likelihood that subject will be harmed by
afﬁrmative action

Concern for distributive justice

Concern for procedural justice

Equality-based distributive justice standard
Need-based distributive justice standard
Equity-based distributive justice standard
Collectivism

Role of afﬁrmative action in co-worker’s career
Perceived competence of co-worker

Role Of afﬁrmative action in my career
Perceived competence of self

Attitude toward afﬁrmative action for African-
Americans

-. 12*
-.04
-.04
-.06

.01
-.07
.01
.05
.03

-.09*
-.00

~.09*

-.08t
-.05
-.01
.05

.03
-.01
.08t

-.04

.03
.08t

.03
-.05
-.01
-.02

.06
-.02
-.06
-.03

 

Note.'p<.10;‘p<.05

Video manipulation was designed to have a positive effect on nonconsequentialism, and

in turn, on attitudes toward afﬁrmative action.

96

Hypothesis 10 predicted that the Equity distributive justice norm would be
negatively related tO attitudes toward afﬁrmative action. Interestingly, the results were
signiﬁcant but in the opposite direction from that hypothesized (see Table 8a). The
correlation with the Attitudes toward Afﬁrmative Action and Perceived Competence
variables (r = .08 and r = .10 respectively) were both signiﬁcant at the p < .10 level. The
correlation with the Attitudes toward Afﬁrmative Action for African Americans variable
(I = .17) was signiﬁcant at p < .05. These results do not support Hypothesis 10.

The ﬁnal hypothesis (H. 11) predicted that subjects’ scores on the Procedural
Justice variable would be positively related to their attitudes toward afﬁrmative action.
The results were not signiﬁcant for any Of the dependent variables. Correlations ranged
from r = -.03 to r = .05. Hypothesis 1 l was not supported.

Supplemental Analyses

The ability to effectively and accurately test the hypotheses outlined above was tO
some extent limited by the measures that were used. The initial statistics suggested that
some of the scales may not have been accurate measures of the intended constructs. For
example a number Of the scales displayed relatively poor internal consistency (see Table
6); others related to each other in unexpected ways. As such, supplemental analyses and
hypotheses tests were conducted in order to better assess the ﬁndings.

The greatest initial concern came from the consequentialism and
nonconsequentialism scales. These two scales were expected to be distinct from, and
possibly orthogonal tO each other, as they represent two alternative ethical frameworks.
The video manipulation was designed to move individuals toward a nonconsequentialist

ethic with the underlying assumption that they would move away from a consequentialist

97

ethic. Instead, the video manipulation was signiﬁcantly negatively related to m the
nonconsequentialism and consequentialism scales. Contrary to expectation, the two scales
related in the same direction to the video manipulation and showed a relatively high
positive correlation between each other (r = .85, p < .05).

In order to further investigate this concern, factor analyses were conducted of all
the items designed to measure ethical frameworks. Using orthogonal rotation and a
minimum eigenvalue Of 1.0, ﬁve factors emerged among the items that were designed to
assess these variables. Items were assigned to the factors where they displayed the highest
loading, with a minimum factor loading Of .30. This analysis yielded very interesting
results. The items designed to measure consequentialism and nonconsequentialism
(which were taken from the Brady & Wheeler, 1996) scale all loaded on the same factor.
There is some indication that the subjects responded tO the form Of the items over and
above the content. The other items designed to measure the individual ethics frameworks
(self interest, neutralization, duty, justice, religious conviction, light Of day, utilitarianism

and categorical imperative) showed greater discrimination (see Table 13).

98

Table 13 - Factor Matrix - Ethics Framework Items

 

Item Factor] Factor 2 Factor 3 Factor 4 Factor 5

 

62 .86 .1] a16

53 .86 a18

51 .85 n12

63 .84 n10

55 .84 a18

54 .83 .16 al4

61 .81

52 .80 .15 a12

49 .80 216

64 .79 n13

59 .78 .12

66 .77 a12

56 .75 .24 .15

58 .75 n17 .12

48 .74 .18

57 .72 .14 .10
60 .71 .20

65 .68 .24 a15 .12

50 .58 .11

67 .56 .30 .17

31 .43 :41 .31 a32

42 ~17 .64 .26 .15
16 .61 .15 .22

28 -20 .54 .37

38 -11 .51 .23 .17
18 .ll .47 .27

37 .31 ﬂ41 .34 all
39 .40 .27 .10 a15
40 .12 .39 .35 all .11
23 .26 a39 a26 all
15 .13 .38 .28 n30

44 .29 a36 .22 a34

41 .32 .29 .24

 

99

Table 13 - Factor Matrix - Ethics Framework Items (cont’d).

 

Item Factor] Factor 2 Factor 3 Factor4 Factor 5

25 .1] a14 .57 .22 a36
27 .17 .14 .56 n33
26 .26 .14 .49 a29

29 .34 .25 .42 238 .11
30 .3] a38 .39 =17
33 .33 .14 .36 «33 .15
19 .28 «28 .32 ~14

35 a44 .30 .59 .32
20 n38 .28 .53 .35
45 .14 a43 .39 .48 .36
24 .15 a47 .32 .47 .35
46 .17 .32 .46 n34

 

Note. Actual items are shown in Appendix B.

Factor 1 corresponds to the Brady & Wheeler (1996) consequentialism and
nonconsequentialism scales.

Factor 2 is comprised of individual ethics framework items, most of which are
consequentialist.

Factor 3 is comprised of individual ethics framework items, most of which are
nonconsequentialist.

Factors 4 & 5 are both predominantly comprised of religious conviction items.

Based on the factor loading patterns of these items and a review Of their content,
the ethical framework measurement was re-conceptualized. The initial consequentialism
and nonconsequentialism measures (Brady & Wheeler, 1996) were discarded and the
emphasis placed instead on the other ethics items (designed to measure the individual
sub-frameworks). The loadings and content of these items suggested that they could
appropriately be placed along a continuum from nonconsequentialism to

consequentialism. In other words, the data suggest that consequentialism and

nonconsequentialism ought to be viewed as Opposite ends of the same scale rather than as

100

distinct entities.

In order to develop this single scale, the items originally designed to measure
nonconsequentialism were reverse-coded and combined with the consequentialism items.
The internal consistency reliability Of the resulting l4-item consequentialism scale
(on2scale) proved to be within the acceptable range (alpha = .76). The items were again
subjected to factor analysis. Once again, I used orthogonal rotation and a minimum
eigenvalue of 1.0. Based on this criterion three factors emerged among the items that
were designed to assess these variables. Based on the previously set rule-of-thumb (items
were assigned to the factors where they displayed the highest loading, with a minimum
factor loading of .30) most of these items loaded as expected (see Tables 14 & 15). This

new consequentialism scale was used for further hypothesis testing (below).

Table 14 - Factor Matrix — New Consequentialism Scale Items

 

Item Factor] Factor2 Factor 3

 

42 .75 .05 a03
16 .69 .20 a16
28 .67 «00 n30
38 .62 .02 n07
18 .42 .47 .09
39 .40 .37 n39
40 .35 .52 .03
15 .24 .44 .59
41 .36 .36 .31
37r .44 n5] .3]
23r .52 a25 .04
44r .49 a40 n19
30r .44 a46 .42
l9r .37 a5] a06

 

' Reverse coded items

101

Table 15 - New Consequentialism Scale

 

Item
Number

 

42

16
28

38
18
39
40

15
41

37r
23Ir
44r
30r
19r

That it is the outcome that is important not the means by which it
is achieved.

That as long as nO-one gets hurt an action is OK.

That an unethical action is OK if it is directed at someone, or an
organization, that also acts unethically.

It is wasted energy worrying about the effect that an action may
have, one should just get on with what one has to do.

In today's business world one must look after oneself and one's
interests.

Whether any inherent harm in an action is outweighed by the
good.

That as long as the consequences of the decision affect the
majority in a positive way.

What was best either for myself or for my company.

That one can't be expected to be responsible for everyone and
everything.

DO unto others as you would have them do unto you.

One should not treat people as merely a means to an end.

It is important that discriminatory practices be avoided.

How I would feel if someone did that to me.

That it is important that justice is seen to be done.

 

Note: The items were preceded with the phrase “When deciding on the cases I generally

considered.”

Responses were based on a Lickert scale where 1 = Strongly Disagree, 5 = Strongly

Agree

These questions were follow-up questions to ethics dilemma scenarios. See Appendix B.

Hypotheses 8 and 9 were re-tested with the new unidimensional consequentialism

scale. Since the hypotheses were conceived as simple linear relationships, they were

tested using bivariate correlation. The new consequentialism scale was expected to be

negatively related to attitude toward afﬁrmative action. The video manipulation was

expected to predict lower scores on this scale. The results are Shown in Table 16.

102

Table 16 — Revised Hypothesis Test (Zero-Order Intercorrelations) with Unidimensional
Consequentialism Scale

 

 

Attitude Attitude Perceived
toward A. A. toward A. A. Competence
for African-
Americans
Hyp.
8 Consequentialism(-) -.15* -.23* -.22*

 

Note.‘p<.10;*p<.05

As Table 16 shows, the revised consequentialism variable is signiﬁcantly
negatively correlated with all three measures Of attitude toward afﬁrmative action. This
provides support for Hypothesis 8. It also provides indirect support for Hypothesis 9a.
Given that consequentialism and nonconsequentialism have been conceived as opposite
ends of a continuum, the negative relationship between consequentialism and attitudes
toward afﬁrmative action suggests a positive relationship between attitudes and
nonconsequentialism.

In addition to the revised hypothesis test, the relationship between the video
manipulation and the new consequentialism measure was also assessed (Hypothesis 9b).
It was expected to be negative. It was negative but not signiﬁcant (r = -.05). Further
analysis showed that this effect was more pronounced for subjects who scored in the mid-
range On the new consequentialism variable. The correlation for subjects in the middle
third (ranked by their scores on the new consequentialism variable) between the video

manipulation and the new consequentialism measure was r = -.12 (N = 79). However, it

103

was not signiﬁcant. This ﬁnding though not statistically signiﬁcant is quite encouraging
given the small sample size. It can be expected that with a larger sample this
manipulation may be effective in moving people to adopt more of a nonconsequentialist

ethics framework.

104

, - . Chapter 4
DISCUSSION

In spite Of the vast changes that have taken place on the political and
organizational landscape over the last few years, afﬁrmative action continues to be a
controversial and important issue for the American work place. Research that adds to the
understanding of these and similar efforts holds the potential to alleviate much of the
conﬂict and difﬁculty surrounding various identity conscious organizational efforts.
Although a number of studies have examined the role of various factors in predicting
individuals’ affective and behavioral responses to afﬁrmative action, this dissertation is
unique in attempting to alter individuals’ reactions. Thus, in spite Of the mixed results, it
makes a signiﬁcant contribution to the literature on identity conscious organizational
initiatives.
Discussion of the Effects of Ethics Variables on Attitudes toward Afﬁrmative Action

The primary thrust of this dissertation was that affective reactions to afﬁrmative
action can be predicted by individuals’ ethics and justice frameworks, and that by
manipulating individuals’ justice frameworks their attitudes toward afﬁrmative action
could also be altered. More speciﬁcally, it was expected that nonconsequentialist ethical
perspectives would predict positive attitudes toward afﬁrmative action, and that
consequentialist ethical perspectives would predict negative attitudes. Additionally it was
expected that individuals who focus on equity norms of distributive justice would have
more negative attitudes toward afﬁrmative action and that persons who focus on
procedural (versus distributive) justice would exhibit more positive attitudes toward

afﬁrmative action.

105

The initial ﬁndings suggested that Opposition to afﬁrmative action is my rooted in
individuals’ predominant ethical perspectives. It appeared that pot_h consequentialist and
nonconsequentialist ethics were positively related to attitudes toward afﬁrmative action.
This suggested, though not conclusively, that negative reactions to afﬁrmative action are
predicted by other individual differences. Scholars have suggested a variety of predictors
including aversive racism (Dovidio, Mann, & Gaertner, 1989); economic self-interest,
racial affect, and stratiﬁcation beliefs (Kluegel & Smith, 1983).

These initial ﬁndings are however, questionable. As explained in the previous
chapter, the pattern of results suggested that the original consequentialism and
nonconsequentialism measures did not effectively represent the underlying constructs.
One of the unexpected outcomes of this study is thus, a reanalysis of the ethics constructs
and their measurement. This will be addressed in additional detail in a later section of this
chapter.

Given the reconceptualization of the ethics framework measures and reanalysis of
the relevant hypotheses, the most robust ﬁnding of this dissertation was that
consequentialist ethics norms are predictive of negative attitudes toward afﬁrmative
action. In other words persons who determine the difference between right and wrong
based on the potential outcomes or consequences of a decision or action are less likely to
support these efforts. This ﬁnding supports the popular anecdotal assertion that
Opposition to afﬁrmative action is rooted in judgments that the policy is unethical or
immoral (Madkins, 1989). However, these judgments are limited by the ethical
frameworks that these persons are applying. Based on our ﬁndings, we would expect that

persons who apply nonconsequentialist ethical approaches to their analysis of afﬁrmative

106

action would be positively disposed toward it.

The reanalysis of the ethics data also lent support to the continued use of the
Religious Conviction variable. The items designed to measure Religious Conviction all
hung together in the factor analysis. This variable Showed a robust positive relationship
with the Attitude toward Afﬁrmative Action (I = .19, p < .05) and the Attitude toward
Afﬁrmative Action for African-Americans (r = .15, p < .05) measures. This ﬁnding
provides added support for the overall thrust of this dissertation since the Religious
Conviction variable is conceptually, very closely related to the nonconsequentialist ethical
perspective.

Discussion of the Effects of Justice Variables on Attitudes toward Afﬁrmative Action

As stated in the previous section, it was expected that individuals who focus on
equity norms of distributive justice would have more negative attitudes toward
afﬁrmative action and that persons who focus more on procedural (versus distributive)
justice would exhibit more positive attitudes toward afﬁrmative action. Neither one of
these propositions was supported by the data. Unfortunately, it is somewhat difﬁcult to
substantively interpret these ﬁndings. The justice measures were all new scales that were
developed speciﬁcally for this study. They were not extensively piloted prior to the study.
With the exception of the equity variable these variables all exhibited unacceptably low
reliabilities (internal consistency estimates ranged from alpha = .28 to .67) and cannot
appropriately be used for any analyses.

Interestingly, the only signiﬁcant ﬁnding for the justice-based independent
variables was for the equity variable and it was in the opposite direction from that

expected. Scores on the equity variable predicted positive scores on the Attitude toward

107

Afﬁrmative Action (r = .08, p < . 10) and Attitude toward Afﬁrmative Action for African-
Americans (r = .17, p < .05) measures. These ﬁndings are contrary to the hypothesized
relationship and contradictory to the ﬁndings described above for the ethics variables.
Equity, like consequentialism, is concerned with the outcomes or consequences of
decisions and actions over and above the procedures or principles used to obtain them.
This outcome must be subjected to additional research.
Reexamining Ethics Measurement

One of the interesting though unexpected outcomes of this study was the
reanalysis and reconceptualization of the ethics measures. Previous studies have treated
consequentialism and nonconsequentialism as distinct ethical frameworks but have not
clariﬁed the relationship between these constructs. It has been asserted that individuals
work out of multiple ethical frameworks at the same time (Hunt & Vitell, 1986). Others
have posited that a multidimensional approach to ethical challenges is more appropriate
for dealing with the complex Situations managers face (Peachment et al, 1995). In general
the empirical studies Of ethics frameworks have been descriptive, examining the
propensity of managers in various ﬁelds/geographical areas to apply different theories to
their ethical decision making (e.g. Vitell, Singhapakdi & Thomas, 2001). Again, these
studies do not investigate the relationship between consequentialism and
nonconsequentialism.

A potential source of additional understanding of the relationship between these
ethics frameworks may be derived from the domain of moral maturity. The research of
Lawrence Kohlberg (1977) describes stages in moral reasoning corresponding to the

stages of human chronological development. Kohlberg’s stages 5 and 6 bear considerable

108

similarity to consequentialism and nonconsequentialism respectively. The substantial
body of research on moral maturity holds promise for additional understanding of the
nature of ethical frameworks.

Limitations and Future Research

The ﬁrst challenge for additional research iS to replicate the study of the
relationship between consequentialism and afﬁrmative action. Although it is in the
expected direction, it was only found after supplemental analyses of the data. The
ﬁndings and conclusions can only be tentatively accepted until they have been replicated.

The experimental manipulation did not have the desired effect on attitudes toward
afﬁrmative action or on the focal ethics variables. It was not signiﬁcantly related to any of
these measures. The apparent challenge is to develop a more robust manipulation. Given
the relationship between consequentialism and attitudes toward afﬁrmative action, it is
likely that an effective nonconsequentialism manipulation will have an effect on attitudes.
However, this manipulation was ineffective in doing so. The reason for this lack of effect
is difﬁcult to assess. It is interesting to note that the manipulation had a signiﬁcant effect
on the Deceptive Practices ethical variable (r = -.12, p < .05). Additionally, as outlined in
the previous chapter, the video manipulation could be expected to exert a signiﬁcant
effect on nonconsequentialism, at least for some subjects.

As with all laboratory experiments, especially those using undergraduate subjects,
there is a concern for the generalizability of the study ﬁndings. There is also a question of
how meaningful the questions and scenarios used in the study were to the subjects. For
example, subjects with limited work experience may have had difﬁculty accurately

assessing the perceived competence of their colleagues or accurately answering questions

109

about their organizations’ policies on afﬁrmative action. In hindsight, it may have been
inappropriate to test hypotheses 5 and 6 on this sample. Future efforts should include
attempts to conduct Similar studies in ﬁeld settings and/or studies with subjects who have
more work experience. Subjects’ work experience Should also be measured, so that its
effects can be directly assessed.

On the other hand, the limitations of the lab setting are in many ways offset by the
advantages of this design. This study was primarily designed to investigate the possibility
of obtaining an alleviating effect, that is, to demonstrate that an effect cg happen. Such a
study would be much more difﬁcult to Observe in a ﬁeld setting. In spite of the
limitations, the study serves as a valuable starting point for future controlled studies in
this vein. Future studies might be conducted in quasi-experimental and/or ﬁeld settings.
They might also be conducted using more mature subjects (those with more work
experience and/or more exposure to afﬁrmative action).

AS mentioned earlier in this chapter, a number of measurement problems arose in
the course of the study. The Societal Perception of Afﬁrmative Action was not properly
assessed, which precluded the testing of Hypothesis 3. The subjects may have had limited
work experience. As such, it may have been inappropriate to test Hypotheses 5 and 6
using this sample. Additionally, the scoring system (optically scanned “bubble sheets”)
did not support the scoring scheme for the Equity Sensitivity scale (King & Miles, 1994;
Huseman, Hatﬁeld, & Miles, 1985; see Appendix B). As such, this scale could not be
used in additional analyses. Fortunately, some redundance of measurement was built into
the study plan. FOr example subjects’ propensity for the equity distributive justice norm

was measured with an original scale in addition to the Equity Sensitivity scale. There

“0

were also several hypotheses assessing the relationship between subjects’ ethical
frameworks and their attitudes toward afﬁrmative action. As such, the measurement
problems did not prevent the discovery of meaningful ﬁndings.

As indicated above, there is a need to further clarify the relationship between
consequentialism and nonconsequentialism. This will require improvements in the
measurement of these constructs and further empirical analysis of their relationships in
multiple contexts. Such analysis has the potential ultimately to contribute to several ﬁelds
of study in which ethical decisions are critical including education, public administration,
medicine and criminal justice.

In addition to the immediate future studies suggested by this dissertation a number
of other directions for future study are evident. The initial driver behind this study was the
unanticipated behavioral consequences of afﬁrmative action. However the study focused
on attitudes and attitude change. A natural next step will be investigation of behavior and
behavior change. It will also be interesting to replicate this study in the context of speciﬁc
afﬁrmative action beneﬁciary groups (e.g. Hispanic, Asian or Arab Americans).

Finally, all research on identity-conscious organizational initiatives has the
potential for international application. There are nations outside of the United States that
are much earlier in the process of dealing with the challenges of equal employment
opportunity. Countries like Afghanistan and South Africa are currently grappling with
issues of multiculturalism and racial reconciliation on the political level. It follows that
these issues will eventually become prominent in the organizational context. One of the

directions for future research is thus investigation in the international context.

1]]

APPENDIX A

Consent Form

112

INFORMED CONSENT AGREEMENT

This study is designed to investigate your opinions and attitudes about a variety of organizational and public
policy issues. If you choose to participate in this study, you will be asked to complete a number of
questionnaires, and possibly to participate in a brief training program Your participation should take a
maximum of two hours. In exchange for this, you will receive credit for having participated in the study. In
addition, you will be entered into a lottery to receive cash prizes ($100, $50, or $20). Other credit alternatives
may be available from your instructor if you decide not to participate in this study.

later in the semester (after all data has been collected) your instructor will provide you with a thorough
debrieﬁng (including a more detailed description of the purpose and preliminary results of the study). In
addition, further information, including group level data reports may be obtained from the experimenters upon

request.

Your participation in this research is entirely voluntary; you are free to decline any questionnaire items, or to
terminate your participation at any time without penalty. All of your responses to the surveys and data
collected will be kept conﬁdential. Your name will only be used to ensure that you receive credit for
participating. All analyses will be conducted using subject numbers. Data will only be reported in summary
form, but will never be broken down in any way that will allow for the identiﬁcation of individual persons or
responses.

If you have any questions regarding the research, you may contact Edward Fubara in the Management
Department at 353-6435. If you have any questions about your rights as a research subject, you may contact
Dr. David E. Wright, MSU Institutional Research Board chair, at 355-2180.

Statement of Agreement

I agree to participate in this study on Organizational and Public Policy as described to me. I understand that I
will be asked to complete a questionnaire, and possibly to participate in a training program. I also understand
that my participation is voluntary, that I may discontinue at any time without penalty, and that all my responses
will be kept conﬁdential so that no reports released will associate the data with my particular responses.

 

Signature:

Name (please print):

 

Date:

 

MSU PID Number:

 

Address:

 

 

Phone Number:

 

Social Security #
(for University purposes only)

 

 

113

DEMOGRAPHIC INFORMATION:

 

Age: Gender: __ GPA:
Ethnicity (please circle one): White/Caucasian African-American Hispanic
Native American Asian Asian American Multiracial Other

114

APPENDIX B

Measurement Instruments

115

Name Section TA

 

ORGANIZATIONAL BEHAVIOR QUESTIONNAIRE II

On the pages that follow there are questions about you and about your
thoughts/attitudes on a variety of organizational and public policy issues.
The questions are designed to give us a picture of how people in general
respond to various policies and practices used in contemporary American
society, especially in organizations.

If we eventually publish this research it is possible that the results will
inﬂuence organizational/public policy decision making. Thus it is very
important that you respond as honestly as you can. The results of this
research will only be reported at a group level, such that individual
responses cannot be identiﬁed. All information collected will be handled
with Strict conﬁdentiality.

You may have completed a very similar questionnaire in the last few weeks.
Please ignore any redundancy and honestly answer these questions. Do not
try to recall earlier answers, but respond based on your true, current
reactions and opinions.

There are no right or wrong answers; the only "right" answer is an
accurate statement of your beliefs. Please express your personal
opinions, not those that you think are popular or expected.

PLEASE BE SURE TO PUT YOUR NAME, SECTION NUMBER AND
TA’S NAME ON THE TOP OF YOUR BUBBLE SHEET. DO NOT FILL
IN THE BUBBLES FOR YOUR NAME, PID, OR SECTION
NUMBER.

116

Section 1: Please use the following scale to respond to questions 1-14. In each case, your
answer refers to the extent to which you agree with a proposal, decision, or action. Please
shade the number corresponding to your answer on the bubble sheet.

1 = Strongly disagree

2:
3:
4:
5:

1.

Moderately Disagree
Neutral

Moderately Agree
Strongly agree

You have heard that a competitor has a new product feature that will make a big
difference in sales. The competitor plans to have a hospitality suite at the annual
trade show and unveil this feature at a party thrown for dealers. You propose sending
someone to this meeting hiding the fact they work for you and pretending to be a
potential client in order to gain valuable information.

At your personal recommendation, your company/client invests considerable
resources in a creative new production process. later you discover that the new
process was by no means as effective or cost efﬁcient as you originally anticipated.
The error is easily recognized as yours, but if brought to the attention of others,
further losses could be avoided. You propose to say nothing and hope that the error is
not recognized.

The research and development department has modernized one of your products. It is
not really new and improved but you know that putting this statement on the
packaging and in current advertising will increase sales. To this proposal do you:

Your Organization employs a large number of women and there is a sympathetic
understanding that the needs of the home or family may, on occasions, take
precedence over their jobs. To compensate for temporary absences for personal
reasons the women are, on average, paid 15% lower than their male counterparts in
the same job. Do you:

You work for a mineral exploration company and you are currently in negotiation
with a property owner for the purchase of a sizable piece of land. One-third of the
land is in mountainous terrain and is of little value and the owner has merely included
it in the total package in order to increase the price. Preliminary research results have
discovered valuable deposits which would put the real value of the property at twice
the asking price. You are now considering what offer you should make to the owner
and you are proposing to pay the asking price with no consideration for the added
value of the property.

You are interested in working for a multi-national and you have sent out a number of

resumes. The recruitment process has now proceeded to the point where you have
been invited to New York for an interview by two different companies, both of which

117

 

offer to pay your expenses. You propose to take one trip but to bill both companies
for the full expense of the trip which you justify to yourself on the grounds that both
companies were willing to pay if you did, in fact, make two trips.

There is a suggestion, as yet unconﬁrmed, that one batch of your consumer products
may be contaminated. The source of the contamination is not known but you do
know that the result is probable gastric illness. You propose waiting until further
cases are reported.

You are a senior executive with an American company that produces bicycles. All
efforts to enter the very lucrative market of a large Asian country have been
frustrated. The company could expect to make substantial proﬁts from sales if it
could penetrate the market. Last week an inﬂuential individual from this country
contacted you and stated that he could smooth the way for the company to sell in his
country for a price. The price is equivalent to 5% of expected proﬁt for the ﬁrst year
of sales. Your company has no policy to guide you in this situation, however, you
plan to go ahead with the arrangement.

You are a production supervisor in a company which provides household electrical
appliances such as ovens and washing machines. You have recently become aware
that one of the products produced by your ﬁrm is defective and unsafe. You have
already brought this to the attention of management but they have done nothing to
remove the defect. You were considering reporting the matter to external authorities
but have decided against this as in all likelihood it would cost you your job.

10. You are the owner and managing director Of a trading company that is involved with

11.

12.

import/export primarily with European markets. AS the third generation manager of
this company you are very proud of what your family has built up over the years
although as you are approaching retirement next year, you are anxious about the
question of a successor. Your eldest son is presently in the United States completing
an MBA degree but is an unknown factor when it comes to practical experience.
Alternatively, you have one other individual in mind who is in his forties, has ﬁfteen
years' experience with the company, and is presently at management level. For
reasons of tradition, you propose to appoint your son as your successor.

You are the ﬁnancial controller for a listed company. Your company has a long
established policy of buying up its own shares for its treasury Stock, pension portfolio,
and for executive stock options. The directors know that the value of the stock will
rise dramatically when certain discoveries and innovations are announced to the
public. You propose buying up as much stock as possible prior to the announcement.

A member of your sales team has recently revealed that he is a practicing homosexual.
He is involved in considerable client contact in an industry which has a particularly
masculine culture. You are afraid that his homosexuality might affect your business.
You propose to release this individual from your company some time over the next

118

 

13.

14.

few months.

You work in a branch ofﬁce of a multi-national company. Your head ofﬁce has quite
speciﬁc rules about the giving of money or gifts to potential customers. However,
you perceive that it will be difﬁcult to conclude business in your area without giving
something to those with whom you negotiate. Subsequently, you are considering
appointing a sub-agent in order to assist in market development and also to let the
agent handle any of the "under the table" activities.

You work for a large construction company. In keeping with the industry, your
company experiences a high staff turnover. Recently, you have discovered that one of
your most experienced employees has been cutting down pipes for resale. Your initial
reaction was to report this employee but you know that he will probably be ﬁred.
This employee is, however, extremely valuable to you at work and the proﬁt he makes
from the pipes is not signiﬁcant. It is unlikely that you will take action in this case.

Section 2: Questions 15-47 refer to the process that you went through in answering the
ﬁrst 14 questions. Please use the following scale and shade the number corresponding to
your answer on the bubble sheet.

1 = Strongly disagree

2:
3:
4:
5:

Moderately Disagree
Neutral

Moderately Agree
Strongly agree

When deciding on the cases I generally considered:

15. What was best either for myself or for my company.

16. That as long as no-one gets hurt an action is OK.

17. That as a manager my ﬁrst responsibility and ultimate duty is to my company and its

shareholders.

18. In today's business world one must look after oneself and one's interests.

19. That it is important that justice is seen to be done.

20. What my spiritual advisor might do or recommend.

21.

1 would feel embarrassed if people found out what I had decided to do.

22. That sacriﬁces are often needed in order to ensure the greatest good for the greatest

ll9

number.
23. One should not treat people as merely a means to an end.
24. What is the right thing to do in the light of my religious beliefs.

25. What might be the reaction I would get from my family and friends if the details
relating to this action were revealed.

26. Whether the outcome of my decision produces the greatest net value to all parties
involved.

27. What effect the action might have on my personal reputation and career.

28. That an unethical action is OK if it is directed at someone, or an organization, that
also acts unethically.

29. What would be the most efﬁcient and effective outcome that as a manager I can
achieve for the company.

30. How I would feel if someone did that to me.
31. That people must be treated fairly.
32. What would be the most equitable decision.

33. That ultimately one should ask whether actions are consistent with organizational
goals and do what is good for the organization.

34. That many actions which are described as unethical are in reality common business
practices.

35. What advice is available from a religious or philosophical source.
36. Would I lose face if my involvement in this decision was publicized.

37. Do unto others as you would have them do unto you.

 

38. It is wasted energy worrying about the effect that an action may have, one should just

get on with what one has to do.

39. Whether any inherent harm in an action is outweighed by the good.

40. That as long as the consequences of the decision affect the majority in a positive way.

120

41. That one can't be expected to be responsible for everyone and everything.

42. That it is the outcome that is important not the means by which it is achieved.
43. My ﬁrst priority and duty should be in fulﬁlling my company Obligations.

44. It is important that discriminatory practices be avoided.

45. My religious faith would not permit such an action.

46. I would not want knowledge of my actions to be known by others.

47. Some things in life are deﬁnitely right or wrong regardless of the consequences of the
decision.

12]

Section 3: Questions 48-67 ask about personal characteristics that may be important for
you to have. Remember there are no right or wrong answers. What matters are your
individual choices. Please use the following scale and shade the number corresponding
to your answer on the bubble sheet.

1 = Extremely unimportant

2 = Very unimportant

3 = Somewhat unimportant

4 = Neither important nor unimportant
5 = Somewhat important

6 = Very important

7 = Extremely important

How important is it to you to be:

48. Innovative

49. Principled

50. Benevolent

51. Dependable

52. Resourceful

53. Trustworthy

54. Effective

55. Honest

56. Inﬂuential

57. Dutiful

58. Independent

59. Dedicated to cause
60. Results-oriented
61. Good-intentioned
62. Productive

63. Noted for integrity
64. Compassionate
65. Financially secure
66. Law-abiding

67. A winner

 

122

Section 4: Questions 68-78 assess your opinions about various organizational and social
policies. Once again, we are interested in your personal opinion, not what is popular
or politically correct. Each question has its own response options. Please Shade the
number corresponding to your answer on the bubble sheet.

68. Afﬁrmative action is....

1 = Extremely bad

2 = Quite bad

3 = Slightly bad
4 = Neutral

5 = Slightly good
6 = Quite good

7 = Extremely good
69. Afﬁrmative action is....

1 = Extremely worthless
2 = Quite worthless

3 = Slightly worthless

4 = Neutral

5 = Slightly valuable

6 = Quite valuable

7 = Extremely valuable

70. Overall afﬁrmative action is .....
1 = Extremely foolish

2 = Quite foolish
3 = Slightly foolish

4 = Neutral
5 = Slightly wise
6 = Quite wise

7 = Extremely wise
71. In my Opinion, afﬁrmative action is .....
1 = Extremely unfair

2 = Quite unfair
3 = Slightly unfair

4 = Neutral
5 = Slightly fair
6 = Quite fair

7 = Extremely fair

123

72. In my opinion, afﬁrmative action is .....

1 = Extremely unjust
2 = Quite unjust
3 = Slightly unjust

4 = Neutral
5 = Slightly just
6 = Quite just

7 = Extremely just

73. In my opinion, afﬁrmative action is to the United States.
1 = Extremely harmful

2 = Quite harmful

3 = Slightly harmful

4 = Neutral

5 = Slightly beneﬁcial

6 = Quite beneﬁcial

7 = Extremely beneﬁcial

74. In my opinion, afﬁrmative action is to most Americans.
1 = Extremely unacceptable

2 = Quite unacceptable

3 = Slightly unacceptable

4 = Neutral

5 = Slightly acceptable

6 = Quite acceptable

7 = Extremely acceptable

75. In general, people in the United States view afﬁrmative action:

1 = Extremely positively
2 = Quite positively

3 = Slightly positively

4 = Neutral

5 = Slightly negatively

6 = Quite negatively

7 = Extremely negatively

124

 

76. Overall, how familiar are you with afﬁrmative action?

1 = Extremely familiar

2 = Quite familiar

3 = Slightly familiar

4 = Neutral

5 = Slightly unfamiliar

6 = Quite unfamiliar

7 = Extremely unfamiliar

77. How supportive is your current employer of afﬁrmative action?

1 = Extremely supportive

2 = Slightly supportive

3 = Neutral

4 = Slightly unsupportive

5 = Extremely unsupportive

78. How important is afﬁrmative action to your current organization?

1 = Extremely important

2 = Slightly important

3 = Neutral

4 = Slightly unimportant

5 = Extremely unimportant

125

 

Section 5: Please use the following scale for questions 79-81. Please shade the number
corresponding to your answer on the bubble sheet.

1 = Extremely unlikely
2 = Quite unlikely

3 = Slightly unlikely

4 = Neutral

5 = Slightly likely

6 = Quite likely

7 = Extremely likely

79. How likely is it that afﬁrmative action will have an impact on your employment
opportunities at some point in your life?

80. How likely is it that you will beneﬁt from afﬁrmative action?

81. How likely is it that you will be harmed by afﬁrmative action?

126

Section 6: The questions below (82-86) ask what you’d like for your relationship to be
with any organization for which you might work. On each question, divide 10 points
between the two choices (choice A and choice B) by giving the most points to the choice
that is most like you and fewest points to the choice that is least like you. You can, if
you’d like, give the same number of points to both choices (that is, 5 points to choice A,
and 5 points to choice B). You can use zeros if you’d like.

Just be sure to allocate all 10 points per question between each pair of possible responses.
Please write your responses in on this sheet.

82. It would be more important for me to:

A. Get from the organization
B. Give to the organization

83. It would be more important for me to:

A. Help others
B. Watch out for my own good

84. I would be more concerned about:

A. What I received from the organization
B. What I contributed to the organization

85. The hard work I would do should:

A. Beneﬁt the organization
B. Beneﬁt me

86. My personal philosophy in dealing with the organization would be:

A. If I don’t look out for myself, nobody else will
B. It’s better for me to give than to receive

127

Section 7: Please use the following scale to respond to questions 87-112. Please shade
the number corresponding to your answer on the bubble sheet.

1 = Strongly disagree
2 = Moderately Disagree
3 = Neutral

4 = Moderately Agree
5 = Strongly agree

87. The outcomes of the decisions that my supervisors at work/professors in school use to
make decisions are more important to me than the processes that led to those decisions.

88. The actual compensation that employees receive on their jobs is not very important as
long as their employers use fair methods for determining it.

89. The processes that my supervisors at work/professors in school use to make decisions
are more important to me than the actual outcomes of their decisions.

90. In the organization where I work, I would like it if everyone on the same level
received the same pay.

91. When people at work are given projects in groups or teams, everyone in the group
should receive the same pay.

92. Schools and employers should do more to reduce competition and help people feel
equaL

93. It is important to me for people to receive equal outcomes in life.

94. In giving pay raises supervisors should consider how employees’ personal
circumstances may have influenced their work and adjust their raises accordingly.

95. In giving grades, instructors should make adjustments to beneﬁt students who need to
meet certain GPA requirements (e.g. to graduate, or get a scholarship).

96. As long as students get grades that accurately reflect their ability, it doesn’t matter
how instructors determine the grades.

97. It is important to me for people to receive what they need.

98. Employees’ compensation (salary and beneﬁts) should be directly proportional to the
amount of effort they put into their jobs.

99. Better workers in an organization should get higher salaries than worse ones.

128

 

100. It is OK for some students to pass a class and others to fail, as long as their scores
reﬂect their effort and ability.

101. It is important to me for people to receive outcomes in life in proportion to their
contributions.

102. It is inappropriate for a supervisor to ask subordinates about their personal life
(such as, where they plan to go on vacation).

103. When I am among my colleagues/classmates, I do my own thing without really
thinking about them.

104. One needs to return a favor if a colleague lends a helping hand.
105. I would not loan a camera or coat to a colleague or classmate.

106. We ought to develop more independence among students so that they do not have
to rely on others’ help in their schoolwork.

107. A group of people at their workplace were discussing where to eat. A popular
choice was a restaurant which had recently opened. However, someone in the group had
discovered that the food there was unpalatable. Yet the group disregarded the person’s
objection and insisted on trying it out. There were only two alternatives for the person
who objected: either to go, or not to go with the others. In this situation, not going with
the others is a better choice.

108. There is everything to gain and nothing to lose for employees to work together in
groups.

109. Classmates’ assistance is very valuable for getting good grades in school.

110. I would help if a colleague at work told me that he/she needed money to pay a
utility bill.

111. In most cases, to cooperate with someone whose ability is lower than one’s own is
not as desirable as doing something alone.

112. Do you agree with the proverb “too many cooks spoil the broth”?

129

Section 8: In many kinds of organizations and occupations, women, and members of
various ethnic minority groups have traditionally not been employed in signiﬁcant
numbers. In the last 20 years or so, that situation has changed, and more women/people
of color have been hired into such jobs.

Please take a few seconds to think of a speciﬁc co-worker, one who has joined your unit
in the past few years and is a member of a group who in the past did not typically have
your type of job. For example, if you work in a male dominated environment, you might
consider a female colleague. This person should be someone who is either at or close to
your organizational level.

Please answer the following questions (113-118) about him or her. Please shade the
number corresponding to your answer on the bubble sheet. If you are not currently
employed, please skip this section and proceed to question 119.

113. To what extent do you think this individual was given this position because of
afﬁrmative action?

1 = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

114. All in all, how competently does this individual perform the job?

1 = Very competently

2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all competently

115. In your estimation, how effectively does he or she get the work done?

I = Very effectively

2 =

3 =

4 = Neutral

5 =

6 =

7 = Very ineffectively

130

116. How likely is it that this individual will move up in the organization?

1 = Very likely

2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all likely

117. To what extent was this person hired because of his or her qualiﬁcations to do the
job well?

1 = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

118. To what extent did you think this person was qualiﬁed to do the job well, when he or
she was hired?

1 = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

131

 

Section 9: Questions 119—124 assess your personal reactions to beneﬁting from
afﬁrmative action. Please shade the number corresponding to your answer on the bubble
sheet.

119. To what extent do you think afﬁrmative action played a role in helping you obtain
your present or most recent position?

1 = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

120. How competent do you think you are/were in performing your current/most recent
job?

1 = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

121. How effectively do you get the work done?

1 = Very effectively

2 =

3 =

4 = Neutral

5 =

6 =

7 = Very ineffectively

122. In your opinion, how likely is it that you will move up in the organization?

1 = Very likely

2 =

3 =

4 = Neutral

5 =

6 =
7 = Not at all likely

132

123. To what extent do you think you were hired because of your qualiﬁcations to do the
job well?

1 = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

124. In hindsight, to what extent did you think you were qualiﬁed to do the job well,
when you were hired?

l = Completely
2 =

3 =

4 = Neutral

5 =

6 =

7 = Not at all

133

Section 10:. Please tell us about your attitude toward race-based afﬁrmative action plans
in the work place. We are specifically interested in affirmative action plans designed
to improve the situation of Black or African-American applicants and employees.

Again, there are no right or wrong answers, all that matters is your opinion. Please
shade the number corresponding to your answer on the bubble sheet.

1 = Strongly disagree

2 = Moderately Disagree
3 = Neutral

4 = Moderately Agree

5 = Strongly agree

125. Afﬁrmative action is a good policy.

126. I would _rlo_t_ like to work at an organization with an afﬁrmative action plan.
127. The goals of afﬁrmative action are good.

128. Employees should be actively involved in attempts to improve the afﬁrmative
action conditions at their place of employment.

129. I would be willing to work at an organization with an afﬁrmative action plan.
130. All in all, I oppose afﬁrmative action plans in industry for Blacks.

134

APPENDIX C

Scale Sources, Reliability And Validity

135

Scale Sources, Reliability And Validity12

Sections 1 & 2: Ethical Frameworks
McDonald & Pak, 1996
.73

Scale Validity: This is a new scale original to these authors. As a part of the pretesting of
the scale, the authors had 23 subjects sort the questions back into the scales that they
came from. This task supports the content or face validity of the items. Factor analysis
also placed all but one of the items into its correct scale.

Section 3: Ethical Predisposition - Character Traits
Brady & Wheeler, 1996

.73 (utilitarianism)

.74 (formalism)

Scale Validity: Again, the validation of these items focused on their content, and whether
or not the statements reﬂected the perspectives that they were designed to. This “content
validation” (a sorting task actually) was conducted by 9 graduate students with training in
ethics. 94% of their judgments were correct.

Sections 4 - 5: Attitude toward Afﬁrmative Action13
Taylor-Carter, Doverspike, & Alexander, 1995

Initial attitude toward preferential treatment .92
Perceived justice of preferential treatment .92
Familiarity with preferential treatment .72

Scale Validity: As with the two previous scales, the scales in this section are new and
have not been formally validated. They were however, developed based on prior research
in a variety of disciplines. The attitude measures were simple semantic differential scales
(good-bad, favorable-unfavorable, valuable-worthless, wise-foolish). The authors assert
that these speciﬁc attitudes were chosen to capture the evaluative dimension of attitudes
[they cite Osgood & Suci, 1969].

The fairness scale takes a similar approach, asking respondents whether the afﬁrmative
action initiatives were fair-unfair, just-unjust, beneﬁcial-harrnful, and acceptable-
unacceptable. The authors provide no additional justiﬁcation for using these items.
However, they appear to be simple queries of individual responses/attitudes rather than
measures of a unique construct. The same may be said for the measures of ego pre-

 

'2 The secondary citations in this section are included only to provide general direction to the reader.
Complete citations and further discussion may be found in the articles that my scales were adapted from. In
each case, the validity of the scales is supported by the fact that they scales appear in published reports of
studies that produced signiﬁcant results. For the purposes of this study I did not deem it necessary to
review the studies that informed the development of these scales.

'3 The authors primarily used the term “preferential treatment” in the questionnaire items, tending to equate
it with afﬁrmative action. I have substituted the term “afﬁrmative action” throughout.

136

occupation, familiarity with preferential treatment, and past/present behavioral
involvement.

Two survey questions assess respondents’ perceptions of their organizations’ support of
afﬁrmative action and their perceptions of its importance to the organization respectively.
These items are direct queries of subjects’ perceptions and serve as moderators for
hypotheses 5 and 6.

Questions 78-80 are similar direct queries, assessing subjects’ perceptions of the potential
effects of afﬁrmative action on their careers. They serve as moderator for hypothesis 1.

Section 6: Equity Sensitivity

King & Miles, 1994
Huseman, Hatﬁeld, & Miles, 198514
.79 - .88

Scale Validity: The King and Miles (1994) study is a validation study of the construct of
equity sensitivity. Evidence from ﬁve samples (N = 2399) is assembled, comparing
equity sensitivity to 16 other variables. Equity sensitivity is shown to be distinct from
these other variables, demonstrating an acceptable pattern of discriminant and convergent
validity.

Section 7: Justice Norms & Collectivism
These are my original items. They have not yet been pretested.

DISTRIBUTIVE JUSTICE
PROCEDURAL JUSTICE
EQUALITY

NEED

EQUITY

COLLECT IVISM
Hui, 1988 (co-worker subscale)
.52 - .71

Section 8: Characterizations of Persons Who May Have Beneﬁted from Affirmative
Action

Heilman, Block, & Lucas, 1992

The authors assess individuals’ ratings of the competence, potency, activity, and various
interpersonal characteristics of “paper people,” some of whom were described as
afﬁrmative action hirees (Study 1). In Study 2, employed persons completed the same

 

'4 Huseman and colleagues converted the scores on this instrument into a nominal scale, labeling individuals
as benevolents, sensitives, or entitleds. I will use a deviation score (away from the midpoint) as a measure
of preference for non-equity versus equity.

137

assessments of real co-workers. In this study, the subjects were also asked to assess the
extent to which afﬁrmative action played a role in the hiring decisions of these
colleagues.” This relationship between afﬁrmative action status and various
performance-related variables may serve as a proxy for subjects’ attitudes toward
afﬁrmative action. Subjects who have relatively positive attitudes toward afﬁrmative
action are likely to have more positive assessments of their colleagues, who may have
beneﬁted from the policy.

Perceived competence .95
Activity (hardworking, persistent, energetic) .85
Potency (strong, forceful, tough) .78
Interpersonal characteristics (responsible, helpful,

cooperative, trustworthy, good co-worker) .91

Scale Validity: Once again, the items in this scale do not appear to be measures of unique
constructs but rather are direct queries, asking the subjects’ opinions about their co-
workers. They are designed to assess subjects’ perceptions and do so in a direct manner.

Section 9: Personal reactions to beneﬁting from afﬁrmative action.

These are original items based on Heilman, Block, & Lucas, 1992 (Study 2). They have
yet been pretested.

Section 10: Attitude toward AA in general

Kravitz, 1995
.80

Scale Validity: These items are simple attitudinal reactions to afﬁrmative action. Kravitz
mentions that they are drawn from multiple sources but provides no additional validation
information. Again, the items appear to be straightforward assessments of respondents’
attitudes toward afﬁrmative action for African-Americans.

 

‘5 In the actual administration of the questionnaire, the afﬁrmative action questions and the competency
questions were separated from each other by other, inconsequential questions about the individual.

138

 

APPENDIX D

Training Script

139

 

ETHICS PRESENTATION
INTRODUCTION

Hello, my name is Scot Yoder. I’m a philosopher, specializing in applied ethics. I‘d like
to talk to you for a few minutes about ethics. Ethics focuses on the age-old question of
“What is right?” How can I tell the difference between right and wrong? How do other
people make decisions about right and wrong? How do their decision-making patterns
affect mine? Are there any standards or guidelines that I can refer to for telling the
difference between right and wrong? Does the difference depend on time and place?
These are questions that have puzzled philosophers, decision-makers, managers, and
leaders of all sorts for hundreds of years. The following presentation seeks to help answer
these difﬁcult questions. It provides you with a set of standards or principles for
determining the difference between right and wrong.

Many people determine what is right or ethical by assessing the potential consequences of
a decision or action. Pe0ple ask themselves a series of "what it” question. However, in
this mini-course we want to present you with an ethical decision making model that
transcends the consequences of decisions, and is rooted in stable, universally applicable
principles. Here are a few examples. Please refer to page 1 of your booklet.

HEINZ AND THE DRUG

In Europe a woman was near death from a special kind of cancer. There
was one drug that doctors thought might save her. It was a form of radium
that a druggist in the same town had recently discovered. The drug was
expensive to make, but the druggist was charging ten times what the drug
cost to make. He paid $200 for the radium and charged $2,000 for a small
dose of the drug. The sick woman's husband, Heinz, went to everyone he
knew to borrow the money, but he could only get together about $1,000,
which is half of what it cost. He told thedruggist that his wife was dying,
and asked him to sell it cheaper, or let him pay later. But the druggist said
"No, I discovered the drug and I'm going to make money from it." So
Heinz got desperate and began to think about breaking into the man's store
to steal the drug for his wife.

In this story, one could easily say that Heinz should steal the drug in order to save his
wife. Although most of us would not steal, in this case we might assert that the end
justiﬁes the means. In other words, stealing is not wrong in this particular case because
of the circumstances. However, some things (such as stealing) are inherently wrong. In
other words, they are wrong in and of themselves. These standards of right and wrong
are almost universally acceptable among reasonable individuals. Based on this
reasoning, stealing is wrong, regardless of the circumstances. Thus, we might advise
Heinz to continue to ﬁnd ways to get the money to pay for the drug (for example, by
selling his personal property).

140

ESCAPED PRISONER

A man has been sentenced to prison for 10 years. After one year, however,
he escaped from prison, moved to a new area of the country, and took on
the name of Thompson. For 8 years he worked hard and gradually he
saved enough money to buy his own business. He was fair to his
customers, gave his employees top wages, and gave most of his own
proﬁts to charity. Then one day, Mrs. Jones, an old neighbor, recognized
him as the man who had escaped from prison 8 years before, and whom
the police had been looking for.

In this example, the basic ethical question is whether or not Mr. Thompson should have
to go back to jail. Once again, if we take a means-ends approach to the question, we
would say that putting Mr. Thompson back in jail will not serve any societal good, and
that he is obviously not a danger to anyone. Here again though, the bigger principle is
that crimes require punishment. Disconnecting crime and punishment is inherently
wrong. Thus, based on our principled approach to ethics, the answer to the ethical
question is that Mr. Thompson should have to go back to jail.

ROLLFAST BICYCLES

The collusive efforts of local manufacturers have barred the Rollfast
Bicycle Company from entering a large third world market. Rollfast
management expects to net 5 million dollars annually from bicycle sales if
it could penetrate the market. Last week a businessman from the country
contacted Rollfast management and assured them that he could smooth the
way for the company to sell in his country for a “grease” fee of $500,000.

If you were the manager at Rollfast with the ﬁnal decision-making authority on this issue,
would you pay the bribe or so-called grease fee? If you said “no” then you are getting the
hang of the principled ethical reasoning approach. Once again, the bigger principle at
stake here is that bribery is a bad thing, even in cases where “everyone is doing it.”

OTHER PRINCIPLES

From the scenarios above, we see a number of principles emerging. Some actions (like
stealing and bribery) are inherently wrong, others (such as abiding by the law) are
inherently right. There are a number of other principles that stem from this same ethical
approach. Can you pick them out of the list below?

[READ ALOUD THROUGH THESE]

141

1. Persons who have been harmed should receive restitution or compensation for their
harm.

2. People should always be treated as ends in themselves, never as means to an end.

3. The focus of ethical inquiry should be on decisions or behavior itself, not on the
outcomes of the decisions, or consequences of the behavior.

4. Sometimes the rights of one person have to be sacriﬁced for the good of the many.

5. Ethical decision making principles must be reversible -- a person making a decision

about someone else would want others to use the same standard for decisions that
affect him or her.

6. Ethical decision making principles must be universalizable -- any single individual
would not mind if the decision making principle were used by anyone or everyone
else in the world.

7. In order to avoid offending others, people should adapt their ethics to match the
customs of the cultures in which they live or work.

[BRIEF PAUSE FOR STUDENTS TO COME UP WITH ANSWERS]

Answer: The consistent ethical principles are 1, 2, 3, 5 & 6. Statement number 1 is
similar to the Escaped Prisoner situation described above. Generally speaking, crimes
should be punished and persons who have been victimized should restored. Statement 2
may be considered to be the foundation of the others. It reminds us that people are to be
treated with respect and dignity, and not to be counted as mere components in economic
comparisons.

Statement 3 is also foundational. Although wrong actions can often produce
positive outcomes, the focus is on the actions themselves. Statements 5 and 6 serve as
ethics tests or measures. Principled ethics result in standards that are acceptable to all
persons regardless of their social standing or ideological position.

Contrast these statements with the other 2. Statement number 5 denies
individuals of their rights and holds the potential to treat a person as a means to an end.
Statement number 7 reﬂects a situational ethics which, again, might allow for the
trammeling of the rights of an individual and/or the violation of a larger ethical principle
given the right environment.

APPLICATIONS OF THIS PRINCIPLE-CENTERED APPROACH TO SITUATIONS

Here’s one more chance to apply our ethics approach. Consider how you would respond
to the following. Pay special attention to the reasoning behind your decisions, not just the
decisions themselves. Write down your decisions and the rationale for them in the space
on the worksheet.

SELL THE CAR
You are trying to sell your car because you need money, but at the annual

car inspection a few days ago, you were informed that the car’s motor is
going to break down soon. You were also informed that it would cost you

142

$3,500 to repair it. The car looks all right though, and a man has offered
you $5,000 for it after he took a short test drive.

FAITHLESSNESS

You are the general manager of a big company, which cooperates, with a
small specialist company to develop a component that is basic to one of
your products. The small company works with great enthusiasm on the
project, since they believe that they will get the order to supply you with
the component though no agreement was made. However, when the
project is ﬁnished, it is shown that a company in Southeast Asia is able to
produce the component at a considerably lower price than the company
you had previously cooperated with.

[FADE - PAUSE THE TAPE]

By now you should be able to apply the principled ethical approach quite easily. In the
“Sell the Car” scenario you would inform the man of the problems with the car and give
him the choice of purchasing it or not. This scenario is a particularly good illustration of
Principle number 6 above. If you were in the buyer’s position it would be pretty clear to
you which decision is right.

The ﬁnal scenario illustrates a situation in which an implied commitment has been
made. The right thing to do is to honor it, even though it was neither legal nor ofﬁcial.
Again put yourself in the place of the managers/employees of the smaller ﬁrm. From a
practical standpoint, it may be possible to work cooperatively with the smaller ﬁrm to
reduce costs, rather than to abandon it.

CONCLUSION

This presentation provides you with a framework for principled, consistent, and generally
acceptable model for determining the difference between right and wrong. It has practical
relevance for both organizational and personal decision making and behavior. It is
particularly valuable because, in the long run, it makes for a safer and more stable society.
An added beneﬁt is that in spite of its long term and principled focus, it often leads to
shorter-tenn beneﬁts. For example organizational research has demonstrated tangible
connections between ﬁrm’s ethical conduct and their ﬁnancial performance.

143

 

Notes:

 

l. Heinz, and Escaped Prisoner scenarios adapted from Rest, J. (1986). Manual for the
Deﬁning Issues Test 3rd edition. Minneapolis: Center for the study of Ethical
Development, University of Minnesota.

2. Rollfast Bicycles scenario adapted from Premeaux, S. R., & Mondy, R. W. (1993).
Linking management behavior to ethical philosophy. Journal of Business Ethics, _1_2_,
349-357.

3. Sell the Car and Faithlessness scenarios adapted from Kavathatzopoulos, I. (1993).
Development of a cognitive skill in solving business ethics problems: the effect of
instruction. Journal of Business Ethics, 12, 379-386.

Scenarios were selected so as to be general and relevant/applicable to all subjects.
Scenarios that included content on topics such as racism, abortion, or homosexuality were
avoided to prevent possible confounds.

144

REFERENCES

145

 

REFERENCES

Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (ed.), Advances in
experimental social psychology (Vol. 2). New York: Academic Press.

Becker, G. S. (1995, August 21). End afﬁrmative action as we know it. Business Week,
16.

Bell, M. P. (1996). Attitudes toward afﬁrmative action: Tests of a comprehensive
framework. Unpublished doctoral dissertation. University of Texas at Arlington.

Bell, M. P., McLaughlin, M. E. & Harrison, D. A. (1996). The belief structure of attitudes
toward afﬁrmative action: sex, race, and population differences. Paper presented
in the Women in Management Division of the 1996 Academy of Management
meetings, Cincinnati, OH.

 

Bennett-Alexander, D. D. (1990). The state of afﬁrmative action in employment: a post-
Stotts retrospective. American Businesgw Journal, 2_7, 565-597.

Bies, R. J ., & Moag, J. S. (1986). Interactional justice: Communication criteria of
fairness. In R. J. Lewicki, B. H. Sheppard, & M. Bazerrnan (Eds.) Research on
negotiation in organizations (Vol. 1, pp. 43-55). Greenwich, CT: JAI Press.

Blackburn, S. (1994). The Oxford dictionary of philosopmr. New York: Oxford
University Press.

Bobocel, D. R., & Farrell, A. C. (1996). Sex-based promotion decisions and interactional
fairness: investigating the inﬂuence of managerial accounts. Journal of Applied

Psychology, _8_1, 22-35.

Brady, F. N. (1990). Ethical Managing: Rules and Results. New York: Macmillan.

Brady, F. N ., & Wheeler, G. E. (1996). An empirical study of ethical predispositions.
Journal of Business Ethics. 15,, 927-940.

Brockner, J. & Wiesenfeld, BM. (1996). An integrative framework for explaining
reactions to decisions: Interactive effects of outcomes and procedures.
Psychological Bulletin. 120. 189-208.

Brockner, J ., Tyler, T. R., & Cooper-Schneider, R. (1992). The inﬂuence of prior
commitment to an institution on reactions to perceived unfairness: the higher they

are, the harder they fall. Administrative Science Quarterly, 3_7, 241-261.

146

Brown v. Board of Education of Topeka, Shawnee County, Kansas 347 US 483 (1954).

Cascio, W. F., Outtz, J ., Zedeck, S., & Goldstein, I. L. (1995). Statistical implications of
six methods of test score use in personnel selection. Human Performance, 8, 133-
164. ‘

Chacko, T. I.(1982). Women and equal opportunity: some unintended effects. Journal of
Applied Psychology, 6_7, 119-123.

Chang, W. C. (1996). Toward equal opportunities: fairness, values, and afﬁrmative action
programs in the US. Journal of Social Issues, _52, 93-97.

 

Chen, C. C. (1995). New trends in rewards allocation preferences: A Sino-U.S.
comparison. Academy of Management Journal, 3, 408-428.

Civil Rights Act of 1991. Public Law 102-166.
Code of Federal Regulations, 41, 60-2.] 1.

Collison, M. N. (1992). Colleges have done a bad job of explaining afﬁrmative action to
students, critics say. The Chronicle of Higher Education. 38(22), A37-A38

Conlon, D. E. (1993). Some tests of the self-interest and group-value models of
procedural justice: Evidence from an organizational appeal procedure. Academy
of Management JourryaJ, 36, 1109-1124.

Cropanzano, R. & Greenberg, J. (1997). Progress in organizational justice: tunneling
through the maze. In C. L. Cooper & I. T. Robertson (Eds.), International Review
of Industrial and Organizational Psychology. New York: John Wiley & Sons.

Cropanzano, R., & Randall, M. L. (1995). Advance notice as a means of reducing
relative deprivation. Smiﬂustiee Resezlch, 8, 137-151.

Crosby, F. J ., & Cordova, D. I. (1996). Words worth of wisdom: toward and
understanding of afﬁrmative action. Journal of Social Issues. ﬂ, 33-49.

De George, R. T. (1982) Business Ethics. New York: Macmillan
De George, R. T. (1990) Business Ethics. New York: Macmillan

Deutsch, M. (1975). Equity, equality and need: what determines which values will be
used as the basis of distributive justice? Journal of Social Issues, 3_1, 137-150.

 

Dietz-Uhler, B., & Murrell, A. J. (1993). Resistance to afﬁrmative action: a test of four
explanatory models. Journal of College Student Development, 34, 352-357.

147

 

Dovidio, J. F., Mann, J ., & Gaertner, S. L. (1989). Resistance to afﬁrmative action: the
implications of aversive racism. In Blanchard, F. A., & Crosby, F. J ., Afﬁrmative
action in mrsmctive. New York: Springer-Verlag.

Executive Order 11246

Fletcher, J. F., & Chalmers, M. (1991). Attitudes of Canadians toward afﬁrmative action:
opposition, value pluralism, and nonattitudes. Political Behavior. _1_3, 67-95.

Folger, R. & Bies, R. J. (1989). Managerial responsibilities and procedural justice.
Employee Responsibilities and Rights Journal, _2_, 79-90.

Folger, R. (1977). Distributive and procedural justice: combined impact of “voice” and
improvement on experienced inequity. Journal of Personality and Social
Psychology, _3_5, 108-119.

 

Folger, R., & Greenberg, J. (1985). Procedural justice: an interpretive analysis of
personnel systems. Research in Personnel and Human Resources Management, 3,
141-183.

Folger, R., & Konovsky, M. A. (1989). Effects of procedural and distributive justice on
reactions to pay raise decisions. Academy of Management Journal, 32, 115-130.

Fritzsche, D. J. & Becker, H. (1984). Linking management behavior to ethical philosophy
— an empirical investigation. Academy of Management Journal, ﬂ, 166-175.

Garcia, L. T., Erskine, N., Hawn, & Casmay, S. R. (1981). The effect of afﬁrmative
action on attributions about minority group members. Journal of Personality, ﬂ,
427-437.

Gleckman, H., Smart, T., Dwyer, P., Sega], T., & Weber, J. (1991, July 8). Race in the
workplace: is afﬁrmative action working? Business Week. pp. 50-62.

Goldman, S. A., & Arbuthnot, J. (1979). Teaching medical ethics: the cognitive-
developmental approach. Journal of Medical Ethics, _5_, 170-181.

 

Greenberg, J. (1984). On the apocryphal nature of inequity distress. In R. Folger (Ed.),
The sense of injustice: 167-188. New York: Plenum.

Greenberg, J. (1987). Reactions to procedural injustice in payment distributions: do the
means justify the ends? Journal of Applied Psychology, 3, 55-61.

Greenberg, J. (1990a). Organizational justice: yesterday, today and tomorrow. Journal
of Management, 16, 399-432.

148

 

Greenberg, J. (1990b). Employee theft as a reaction to underpayment inequity: the
hidden cost of pay cuts. Journal of Applied Psychology, E, 561-568

Greenberg, J. (1990c). Looking fair vs. being fair: Managing impressions of
organizational justice. In B. M. Staw & L. L. Cummings (Eds.), Research in
Organizational Behavior (Vol. 12, pp. 111-157). Greenwich, CT: JAI Press.

Greenberg, J. (1993). The social side of fairness: Interpersonal and informational classes
of organizational justice. In R. Cropanzano (Ed.), Justice in the Workplace:
Approaching Fairness in Human Resource Managment (pp. 79-103). Hillsdale,
NJ: Lawrence Erlbaum Associates.

Greenberg, J. (1994). Using socially fair treatment to promote acceptance of a work site
smoking ban. Journal of Applied Psychology, 72, 288-297.

Greene, K. W. (1989). Afﬁrmative action and principles of justice. Westport, CT:
Greenwood Press.

Griggs v. Duke Power Company, 401 U. S. 424 (I971).
Guttman, A. (1993). EEO law and personnel practices. Newbury Park, CA: Sage.

Heilman, M. E. (1994). Afﬁrmative action: some unintended consequences for working
women. Research in Organizatiopal Behavior. 1_6_, 125-169.

Heilman, M. E., Battle, W. 8., Keller, C. E., Lee, R. A. (1998). Type of afﬁrmative
action policy: a determinant of reactions to sex-based preferential selection?
Journal of Applied Psychology, 88, 190-205.

Heilman, M. E., & Herlihy, J. M. (1984). Afﬁrmative action, negative reaction? Some
moderating conditions. Organizational Behavior and Human Performance, a,
204-21 3.

Heilman, M. E., Block, C. J ., & Lucas, J. A. (1992). Presumed incompetent? .
stigmatization and afﬁrmative action efforts. Journal of Applied Psychology, _7_7,
536-544.

Heilman, M. E., McCullough, W. F., & Gilbert, D. (1996). The other side of afﬁrmative
action: reactions of nonbeneﬁciaries to sex-based preferential selection. Journal

of Applied Psychology, 8_1, 346-357.

Holloway, F. A. (1989). What is afﬁrmative action? In Blanchard, F. A., & Crosby, F. J .,
Afﬁrmative action in perspective. New York: Springer-Verlag.

149

 

Homans, G. C. (1961). Social behavior: its elementaryforms. New York: Harcourt,
Brace & World.

Honderich, T.(Ed) (1995). The Oxford companion to philosophy. New York: Oxford
University Press.

Hui, C. H. (1988). Measurement of individualism-collectivism. Journal of Research in
Personality, 2, 17-36.

Hunt, S. D., & Vitell, S. (1986). A general theory of marketing ethics. Journal of
Macromarketing, 6, 5-16.

Huseman, R. C., Hatﬁeld, J. D., & Miles, E. W. (1985). Test for individual perceptions of
job equity: some preliminary ﬁndings. Perceptual and Motor Skills. 61, 1055-
1064.

Huseman, R. C., Hatﬁeld, J. D., & Miles, E. W. (1987). A new perspective on equity
theory: the equity sensitivity construct. Academy of Management Review. E,
222-234.

Jacobson, M. B., & Koch, W. (1977). Women as leaders: performance evaluation as a
function of leader selection. Organizational Behavior and Human Perforrnﬂe.

29, 149-157.

Johnston, W. B., & Packer, A. E. (1987). Workforce 2000: Workand Workers for the
21 st Century. Indianapolis, IN: Hudson Institute

Jones, (1985) The genesis and present status of afﬁrmative action in employment:
economic, legal, and political realities. 70 IOWA L. REV. 901, 903

Jones, (1988) The origins of afﬁrmative action. 21 U. CAL. DAVIS L. REV. 383, 385
Kant, I. (1788/1993). Critigue of Practical Reason. New York: Macmillan.

Kavathatzopoulos, I. (1993). Development of a cognitive skill in solving business ethics
problems: the effect of instruction. Journal of Business Ethics. 12, 379-386.

Kerlinger, F. N. (1986). Foundations of Behavioral Research. Fort Worth, TX: Holt,
Rinehart, and Winston, Inc.

Kinder, D. R., & Rhodebeck, L. A. (1982). Continuities in support for racial equality,
1972 to 1976. Public Opinion Quarterly, 4_6_(2), 195-215.

King, W. C., & Miles, E. W. (1994). The measurement of equity sensitivity. Journal of
Occupational and Organizational Psychology, 6_7, 133-142.

150

Klimoski, R. & Palmer, S. N. (1994). The ADA and the hiring process in organizations.
In S. M. Bruyere,. & J. O’Keefe (Eds.). Implications of the Amerigns with

disabilities act for psychology. (pp. 37-83). New York: Springer.

Kluegel, J. R., & Smith, E. R. (1983). Afﬁrmative action attitudes: effects of self-interest,

racial affect, and stratiﬁcation beliefs on whites’ views. Social Forces. 61, 797-
824.

Kohlberg, L. (1969). In D. Goslin (Ed.). Handbmi of Socialisation Theory and
Research. (pp. 347-480). Chicago: Rand McNally.

Kohlberg, L. (1977). The implications of moral stages for adult education. Religious
Education 3(2), 183-201.

 

Konovsky, M. A., & Cropanzano, R. (1991). The perceived fairness of employee drug
testing as a predictor of employee attitudes and job performance. Journal of

Applied Psychology, 26, 698-707.

Konrad, A. M., & Linnehan, F. (1995). Formalized HRM structures: coordinating equal
employment opportunity or concealing organizational practices? Academy of
Management Journal, 3_8, 787-820.

Kravitz, D. A. (1995). Attitudes toward afﬁrmative action plans directed at blacks:
effects of plan and individual differences. Journal of Applied Social Psychology,
2_5, 2192-2220.

Kravitz, D. A., & Platania, J. (1993). Attitudes and beliefs about afﬁrmative action:
effects of target and of respondent sex and ethnicity. Journal of Applied
Psychology, 78, 928-938.

Kravitz, D. A., Harrison, D. A., Turner, M. E., Levine, E. L., Brannick, M. T., Denning,
D. L., Russell, C. J., Conard, M. A., & Bhagat, R. S. (1996, October). Afﬁrmative
action. a review of ps sychological and behavioral resea_rch. Report of a
subcommittee of the Scientiﬁc Affairs Committee of the Society for Industrial and
Organizational Psychology.

Leventhal, G. S. (1976). The distribution of rewards and resources in groups and
organizations. In L. Berkowitz & E. Walster (Eds.), Advances in experimental
social psychology: Vol. 9, 91-131. New York: Academic Press.

Leventhal, G. S. (1980). What should be done with equity theory? In K. J. Gergen, M. S.
Greenberg, & R. H. Willis (eds. ) Social exchange. advances 1n theory a_nd
r____esearch (pp. 27- 55). New York: Plenum Press.

151

 

Leventhal, G. S., Karuza, J ., Fry, W. R. (1980). Beyond fairness: A theory of allocation
preferences. In G. Mikula (ed.) Justice and social interaction (pp. 167-218). New
York: Springer-Verlag.

Lind, E. A. (1995). Justice and authority relations in organizations. In R. Cropanzano &
M. K. Kacmar (Eds.), Organizational politics. justice, and support: Managing the
social climate of the workplﬂ (pp. 83-96). Westport, CT: Quorum Books.

Lind, E. A., & Tyler, T. (1988). The social psychology of procedural justice. New York:
Plenum.

Madkins, J. B. (1989, August). Afﬁrmative action is necessary and ethical. Personnel
Journal. 29-30.

Matheson, K., Echenberg, A., Taylor, D. M., Rivers, D. & Chow, I. (1994). Women’s
attitudes toward afﬁrmative action: putting actions in context. Journal of Applied
Social Psychology, 24, 2075-2096.

McDonald, G., & Pak, P. C. (1996). It’s all fair in love and war: cognitive philosophies
in ethical decision making. Journal of Business Ethics. ,1_5, 973-996.

Mill, J. S. (1861/1957). Utilitarianism. Indianapolis: Liberal Arts Press.
Mintzberg, H. (1973). The nature of managerial work. New York: Harper & Row.

Mosley, A. G. & Capaldi, N. (1996). Afﬁrmative action: socﬂljustice or unfair
preference. Lanham, MA: Rowman & Littleﬁeld Publishers.

Nacoste, R. B. (1990). Sources of stigma: analyzing the psychology of afﬁrmative action.
Law & Policy, 12, 175-195.

N acoste, R. B. (1993). Procedural justice and preferential treatment: a brief review and
comment. Current Psycholgggv: Research and Reviews, _1_2_, 230-235.

Nacoste, R. W. (1989). Afﬁrmative action and self-evaluation. In Blanchard, F. A., &
Crosby, F. J ., Afﬁrmative action in parspactive. pp. 103-109. New York:
Springer-Verlag.

Nacoste, R. W. (1994). Policy schemas for afﬁrmative action. In Heath, L. et a1
Applications of Heuristics and Biases to Social Issues. pp. 205-221. New York:
Plenum Press.

National Labor Relations Act (1935)

Newman, J. D. (1989) Afﬁrmative action and the courts. In Blanchard, F. A., & Crosby,

152

 

F. J ., Afﬁrmative action in perspective. New York: Springer-Verlag.

Novelli, L., J r., Kirkman, B. L., & Shapiro, D. L. (1995). Effective implementation of
organizational change: an organizational justice perspective. In C. L. Cooper & D.
M. Rousseau (Eds.), Trends in orgzmizational behavior (Vol. 2, pp. 15-36).
Chichester, England: John Wiley & Sons.

Opotow, S. (1996). Afﬁrmative action, fairness, and the scope of justice. Journal of
Social Issues, 62, 19-24.

Ozawa, K., Crosby, M., & Crosby, F. (1996). Individualism and resistance to afﬁrmative
action: a comparison of Japanese and American samples. Journal of Applied
Social Psychology, E, 1138-1152.

Parker, C. P., Baltes, B. B., & Christiansen, N. D. (1997). Support for afﬁrmative action,
justice perceptions, and work attitudes: a study of gender and racial-ethnic group
differences. Journal of Applied Psychology, 82_, 376-389.

Peachment, A., McNeil, M., Soutar, G., & Molster, C. (1995). Means or ends? Ethical
decision frameworks in the Western Australian public service. Journal of
Business Ethics. 2_4, 629-641.

Prentice, D. A., & Miller, D. T. (1992). When small effects are impressive.
Psychological Bulletin. 112. 160-164.

Rest, J. (1986). Manual for the Deﬁning Issues Test 3rd edition. Minneapolis: Center for
the study of Ethical Development, University of Minnesota.

Richard, O. C., Fubara, E. I. & Castillo, M. N. (2000). The impact of explanations and
demographic group membership: Reactions to diversity initiatives. Journal of

Applied Social Psychology, 86, 1039-1055.

Roberts, P. C. (1995, September 11). Quotas: a bad idea whose time has gone. Business
Week, 23.

Rosen, B. & Mericle, M. F. (1979). Inﬂuence of strong versus weak fair employment
policies and applicant’s sex on selection decisions and salary recommendation in

a management simulation. Journal of Applied Psychology, 64;, 435-439.

Sackett, P. R. & Wilk, S. L. (1994). Within-group norming and other forms of score
adjustment in preemployment testing. American Psvchologist. ﬂ, 929-954.

Schmidt, F. L. (1991). Why all banding procedures in personnel selection are logically
ﬂawed. Human Performance. 8, 165-177.

153

 

Schminke, M., Ambrose, M. L., & Noel, T. W. (1997). The effect of ethical frameworks
on perceptions of organizational justice. Academy of Management Journal, ﬂ,
1190-1207.

Sears, D. O., & Allen, H. M., Jr. (1984). The trajectory of local desegregation
controversies and Whites opposition to busing. In M. B. Brewer, & N. Miller
(Eds.), Groups in contact: The psychology of desegregation (pp. 123-151). New
York: Academic Press.

Solomon, R. (1949). An extension of control group design. Psycholcgical Bulletin. 516,
137-150.

Sorrell, T. & Hendry, J. (1994). Business Ethica. Oxford: Butterworth-Heinemann.

Stepina, L. P., & Perrewe, P. L. (1991). The stability of comparative referent choice and
feelings of inequity: A longitudinal ﬁeld study. Journal of Organizational
Behavior, _1_2, 185-200.

Stout, K. D. & Buffum, W. E. (1993). The commitment of social workers to afﬁrmative
action. Journal of Sociology & Social Welfare. E, 123-135.

Summers, R. J. (1995). Attitudes toward different methods of afﬁrmative action. Journal
of Applied Social Psychology, _26, 1090-1104.

Sweeney, P. D., & McFarlin, D. B. (1993). Workers’ evaluations of the “ends” and the
“means”: an examination of four models of distributive and procedural justice.
Organizational Behavior and Human Performance, ﬂ, 23-40.

Sweeney, P. D., McFarlin, , D. B., & Inderrieden, E. J. (1990). Using relative deprivation
theory to explain satisfaction with income and pay level: A multistudy
examination. Academy of Management Journal. 2, 423-436.

Sykes, G. M., & Matza, D. (1957). Techniques of neutralization: a theory of delinquency.
American Sociological Review. 22, 664-670.

Takala, T., & Uusitalo, O. (1995). Retailers’ professional professio-ethical dilemmas: the
case of Finnish retailing business. Journal of Business Ethics, 151, 893-907.

Taylor, B. R. (1991). Afﬁrmative Action at Work: Law Politics and Ethics. Pittsburgh,
PA: University of Pittsburgh Press.

 

Taylor, P. W. (1975). Principles of Ethics: An Introduction. Encino, CA: Dickinson
Publishing Company.

Taylor-Carter, M. A., Doverspike, D., & Alexander, R. (1995). Message effects on the

154

perceptions of the fairness of gender-based afﬁrmative action: a cognitive
response theory-based analysis. Social Justice Reseagﬂ, 8, 285-303.

Taylor-Carter, M. A., Doverspike, D., & Cook, K. (1995). Understanding resistance to
sex and race-based afﬁrmative action: A review of research ﬁndings. Human
Resource Management Review. _5_, 129-157.

Thibaut, J. & Walker, L. (1975). Procedural justice: a psychological analysis. Hillsdale,
NJ: Erlbaum.

Tierney, W. G. (1997). The parameters of afﬁrmative action: equity and excellence in the
academy. Review of Educational ReseaLh, 61, 165-196.

Tomasson, R. F., Crosby, F. J ., & Herzberger, S. D. (1996). Afﬁrmative action: Thgros
and cons of policy and practice. Washington, DC: The American University
Press.

Tougas, F., Crosby, F., Joly, S., & Pelchat, D. (1995). Men’s attitudes toward afﬁrmative
action: justice and intergroup relations at the crossroads. Social Justice Reseaﬂ,
8, 87-71.

Tsalikis, J ., & Fritzsche, D. J. (1989). Business ethics: A literature review with a focus
on marketing ethics. Journal of Business Ethics. 8, 695-743.

Turner, R. (1990). The past and future of afﬁrmative action: a gpide and analysis for
human resource professionals and corporate counsel. Westport, CT: Quorum
Books.

Tyler, T. R. (1990). Why people obey the law: procedural justice. legitimacy, and
compliance. New Haven, CT: Yale University Press.

Tyler, T. R.(1987). Conditions leading to value expressive effects in judgments of
procedural justice: a test of four models. J oumal of Personality and Socia_l
Psychology, 5_2,, 333-344.

Tyler, T. R.(1989). The psychology of procedural justice: a test of the group-value
model. Journal of Personality and Social Psychology, 67, 830-838.

Tyler, T. R., & Bies, R. J. (1990). Beyond formal procedures: The interpersonal context
of procedural justice. In J. S. Carroll (Ed.), Applied social psychology and
organizational settings (pp. 77-98). Hillsdale, NJ: Lawrence Erlbaum Associates.

Tyler, T. R., & Lind, E. A. (1992). A relational model of authority in groups. In M. P.

Zanna (Ed.), Advances in exparimental social psychology (Vol. 25, pp. 115-191).
San Diego, CA: Academic Press

155

 

Tyler, T. R., Rasinski, K., & Spodick, (1985). Exploring the meaning of process control.
Journal of Personalitmd Social Psychology. 48, 72-81.

US. Department of Labor, 1995

Veilleux, F., & Tougas, F. (1989). Male acceptance of afﬁrmative action programs for
women: the results of altruistic or egoistical motives? International Journal of

Psychology, 2_4, 485-496.

Vitell, S. J ., & Grove, S. J. (1987). Marketing ethics and the techniques of neutralization.
Joumaﬂf Business Ethics. 6, 433-438.

Vitell, S. J ., Singhapakdi, A., & Thomas, J. (2001). Consumer ethics: An application and
empirical testing of the Hunt-Vitell theory of ethics. JoumaLof Consumer
Marketg Special Issue: Vol. 18(2), 153-178.

Walster, E., Berscheid, E., & Walster, G. (1973). New directions in equity research.
Journal of personality and social psychology, g, 151-176

Watson v. Fort Worth Bank and Trust, 487 US 977 (I988)

Weick, K. E. (1966). The concept of equity in the perception of pay. Administrative
Science Quarterly. _1_1, 414-439.

 

Wood, J. A., Longenecker, J. G., McKinney, J. A., & Moore, C. W. (1988). Ethical
attitudes of students and business professionals: a study of moral reasoning.
Journal of Business Ethics. 1, 249-257.

156

 

   

'quleljjjljj‘ljjjjljjjlul