.340 35.3",
‘L I .
y .0: $7.:
u :1 IE...

 

u»...
r W*. ff

.9. I|\7.A o

v E} . 1
un'
*
.0.qu am? v

1 My: . . . H V ‘ .

 

' ' '-VU\')

                                                   

llllilllllllHUIWIIIIIHIIHIHUI

1293 01050 0001

 

LIBRARY
Michigan State
University

1—

 

 

 

This is to certify that the

dissertation entitled

Diversity Programs as Moderators of the Effect of
Individual Dissimilarity and Group Heterogeneity
on Work-Related Attitudes

'presented by

Shobha Ramanand

has been accepted towards fulfillment
of the requirements for

Ph.D. degree“, Resource Development
9K Urban Studies

‘11:: Kwfi

Ma jOl' professor

 

 

 

Date :5“ ?- -621

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

—.— '— "*v v‘ ' v 1' ~. ~‘ _. _.._ __ _

v—v—T v v.—

 

--——.v- ‘- ——.—_._‘_ ._.—._.,_._'...

PLACE II RETURN BOX to remove thle checkom from your record.
TO AVOID FINES return on or before dete due.

 

DATE DUE DATE DUE DATE DUE

 

H

9 2004I

 

'0

 

gt

 

 

 

 

 

 

 

082207

 

 

 

 

 

 

 

 

MSU “in?”

 

DIVERSITY PROGRAMS AS MODERATORS OF THE EFFECT
OF INDIVIDUAL DISSIMILARITY AND GROUP HETEROGENEITY 0N
WORK-RELATED ATTITUDES
By

Shobha Ramanand

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Resource Development & Urban Studies

1996

 

ABSTRACT
DIVERSITY PROGRAMS AS MODERATORS OF THE EFFECT OF

INDIVIDUAL DISSIMILARITY AND GROUP HETEROGENEITY ON WORK-
RELATED ATTITUDES

By

Shobha Ramanand

Many organizations have instituted diversity programs to modify and address
organizational systems that prevent diverse employees from reaching their full
potential. However, few studies have examined the effects of these programs on the
workplace despite growing anecdotal evidence and case literature documenting their
mixed impact.

A review of the diversity management literature suggests that investment in
diversity programs contribute to increased employee morale, motivation, and
satisfaction. Researchers have theorized that diversity management should lead to
organizational effectiveness and positive work behaviors. The objective of this
dissertation is to investigate the effect of diversity programs/practices on work-related
attitudes--job satisfaction, job involvement, and organizational commitment.

The literature has revealed that demographic attributes influence group outcomes
and behaviors. Both Pfeffer's (1983) organizational demography model and
Schneider's (1987) Attraction—Selection-Attrition (ASA) framework suggest that
dissimilarity and heterogeneity are negatively related to organizational effectiveness,
thereby influencing work-related attitudes. The second objective of the dissertation is

to explore whether the relationship between individual dissimilarity and group

heterogeneity, on work-related attitudes were likely to be moderated by an individual's
perception of organizational efforts in implementing diversity programs/practices.

Based on survey analyses of 214 employees in three financial institutions, the
study demonstrates that satisfaction with diversity programs/practices significantly
predict job satisfaction, job involvement, and organizational commitment. Individual
dissimilarity and group heterogeneity moderately predict work-related attitudes.
Contrary to prediction, diversity programs did not moderate the effects of individual
dissimilarity and group heterogeneity on work—related attitudes. These results seem to
support Lawrence's (1994) and Smith et al. 's (1994) arguments that process
explanations--long ignored by proponents of organizational demography--might be
relevant in predicting organizational performance. The moderate effects of various
demography variables in predicting work-related attitudes in this research suggests the
need to look beyond the demography model.

Implications for research and theory, implications for practice, and directions

for future research are discussed.

 

Copyright by
Shobha Ramanand
1996

DEDICATION

This dissertation is dedicated to my mother, with her belief that I could do anything
with enough perseverance, my father, who got me through this with his love, wisdom,
and patience, and to the women in my family who have always had confidence in my
ability.

ACKNOWLEDGEMENTS

A great many people have provided me with encouragement, support, financial
assistance, and nurture at differing stages, that have made this endeavor possible. In
this regard, I would particularly like to thank my three mentors and committee
members, Drs. Frank Fear, John Schweitzer, and Michael Moore. I am very grateful
to Frank Fear, my committee chair, who provided me with the vision and support, for
asking the right and important questions, and for motivating me to successfully
complete this difficult journey. I am very grateful to Dr. Moore for his intellectual
insights and wonderful encouragement, and sage advise during my long tenure as a
student. I am also very grateful to Dr. Schweitzer for his outstanding support, spiritual
strength, and for making me overcome my fear of statistics by continually providing
insights and novel ways of examining data. I do not have words to express my
immense gratitude to the three of you for openly expressing such confidence in me,
suffice it say, I owe everything I have achieved in this University to you three. I also
owe special thanks to Drs. George Rowan and James Jay for serving on my committee,
for embracing this research topic, giving me interesting ideas, and providing

encouragement.

vi

I would like to acknowledge and thank my Resource Development colleagues,
friends, and family. Cynthia Fridgen, Eck Dersh, Eric, Linda, Karen and Jeff,
Shannon, Kim, Dan, Shawn, Rick, Karen Williams, Karen, Vince, and James for your
friendship support, and ideas. A special thanks to all the RD staff--Sue, Sarah, Julie,
Jamie, Cheryl and Nancy for your support, encouragement, and friendship.

I would like to acknowledge and thank my Urban Studies colleagues, friends,
and family. Dr. Darden for your financial support and encouragement during my long
tenure as a student. Drs. Richard and June Thomas for your encouragement, support
and getting me started on diversity research. A special thanks to Fran Fowler and John
Melcher for being always there for me, your love and affection, and being part of my
adoptive family. A special thanks to De Bryant, Carolyn, Oumatie, Ruba, Pat,
Julianne, Rick Hurst, Rex LaMore, Mary Ellen, Dr. Bonnie Morrison, Irene, and other
friends for their friendship, comradary, and encouragement.

I also want to thank and acknowledge my SLIR colleagues, friends, and family.
Dr. Kruger and Dr. Gleason for your friendship and confidence in me. My Japan
project family-Joel, Betty, Art, Bill, Takashi, Dr. and Mrs Ishino, Nejib, Mark,
Michio, Hisako, Wen Jeng, Jennifer, Hiroko, Susan and others for all your friendship,
love, intellectual stimulation, provocative perspectives, and ideas.

I would also like to thank my three research sites and gratefully acknowledge
their help in getting this dissertation completed. A special thanks to the staff of the

Food Industry Institute for all their help.

vii

I am exceedingly grateful to all my friends for all their love and support, infinite
patience, listening ears, and kindness. Lisa, Kate, Kathy, Sherry, Bruce, Fantaya,
Memoona, Yvette, Willard, many, many thanks.

A special thanks to the Schweitzer and Melcher family for opening their homes
and heart to me.

I would also like to thank and acknowledge the wonderful MSU community who
have made my stay in this country so very joyful, productive, and happy.

And finally, my entire family deserve many rounds of applause. My sincerest
gratitude for their unwavering confidence, support, patience, and love. And for
constantly bugging me to finish this seemingly endless, sometimes frustrating, but great

journey.

viii

TABLE OF CONTENTS

TABLE OF CONTENTS ...................................... ix
CHAPTER ONE
................................................... 1
INTRODUCTION AND OVERVIEW ......................... 1
RESEARCH NEED ................................. 9
DEFINITIONS OF DIVERSITY ....................... 11
PURPOSE OF THE STUDY ......................... 14
THE RESEARCH CONTEXT: THE BANKING INDUSTRY . . 16
RESEARCH METHODS ............................ 18
CONTRIBUTIONS AND LIMITATIONS OF THIS
DISSERTATION ............................ 19
KEY ASSUMPTIONS .............................. 21
ORGANIZATION OF THIS DISSERTATION ............. 22
CHAPTER TWO ........................................... 24
REVIEW OF LITERATURE ......................... 24
MANAGING WORKFORCE DIVERSITY ............... 24
REVIEW OF RELEVANT THEORIES .................. 32
Attraction-Selection-Attrition Model ............... 32
Organizational Demography Model ................ 38
INTERPRETATION OF THE THEORIES ............... 43
DIVERSITY AND WORK-RELATED ATTITUDES ......... 46
Organizational Commitment ......................... 46
Job Satisfaction .............................. 49
Job Involvement ............................. 52
DIVERSITY PROGRAMS AS A MODERATOR ........... 54
CHAPTER THREE ......................................... 57
RESEARCH METHODOLOGY ............................ 57
DIFFICULTY IN GAINING ACCESS TO A STUDY INDUSTRY
......................................... 57
OVERVIEW OF THE BANKING AND FINANCIAL
INSTITUTION INDUSTRY ..................... 58
FOCUS OF THE STUDY ........................... 60
STUDY SAMPLE ................................. 61
METHOD OF DATA COLLECTION ................... 62
Phase I - Identifying the Sample .................. 63
Phase II - Conducting a Pilot Study ................ 63
Phase III - Collecting Quantitative Data ............. 64
RESEARCH MODEL .............................. 66

RESEARCH QUESTIONS AND STUDY HYPOTHESES ..... 70

Hypotheses Associated with Job Satisfaction .......... 71
Hypotheses Associated with Job Involvement ......... 73
Hypotheses Associated with Organizational or Company
Commitment ........................... 75
ANALYSIS OF DATA .............................. 77
Hypotheses Related To Individual Dissimilarity ........ 79
Hypotheses Related to Group Heterogeneity .......... 80
MEASUREMENT DESIGN .......................... 82
Diversity Programs/Practices as a Variable ........... 82
Attribute Dissimilarity of Individual Team Members . . . . 84
Group Heterogeneity .......................... 85
Outcome Variables ........................... 87
Individual Dissimilarity and Group Heterogeneity Variables
.................................... 89
CHAPTER FOUR .......................................... 90
RESULTS ........................................... 90
INTER-CORRELATIONS OF STUDY VARIABLES ........ 90
BRIEF REVIEW OF THE METHODOLOGY ............. 93
RESULTS OF ANALYSES BY HYPOTHESIS ............. 95
Hypothesis 1 ..................................... 95
Hypothesis 2 ..................................... 96
Hypothesis 3 ..................................... 96
Hypothesis 4 ..................................... 97
Hypothesis 5 ..................................... 98
Hypothesis 6 ..................................... 98
Hypothesis 7 .................................... 100
Hypothesis 8 .................................... 100
Hypothesis 9 .................................... 101
Hypothesis 10 ................................... 102
Hypothesis 11 ................................... 103
Hypothesis 12 ................................... 103
Hypothesis 13 ................................... 104
Hypothesis l4 ................................... 105
Hypothesis 15 ................................... 105
Hypothesis 16 ................................... 106
Hypothesis l7 ................................... 107
Hypothesis 18 ................................... 107
DISCUSSION OF THE STUDY FINDINGS .............. 109
ANOVA ANALYSES OF BACKGROUND VARIABLES ..... 114
SUMMARY OF HYPOTHESES ...................... 118
CHAPTER FIVE .......................................... 121

DISCUSSION ....................................... 121

SUMMARY .................................... 121
IMPLICATIONS FOR RESEARCH AND THEORY ........ 124
IMPLICATIONS FOR PRACTICE .................... 131

Developing & Maintaining Competitiveness Through Training

Systems ............................. 132

Attitude Structuring with HRM Practices ........... 134

SUGGESTIONS FOR FUTURE RESEARCH ............. 136

LIST OF REFERENCES .................................... 140
APPENDICES ............................................ 156
APPENDIX A ....................................... 156
QUESTIONNAIRE ............................... 156
INTRODUCTION ................................ 156

SECTION I .................................... 157

SECTION II ................................... 159

SECTION III ................................... 162

SECTION IV ................................... 165

SECTION V .................................... 167

SECTION VI ................................... 169

APPENDIX B ....................................... 171

An Example Showing the Computation
of a Single Individual Dissimilarity Variable

............................................ 171
APPENDIX C ....................................... 172
Group Heterogeneity (Jackson et al. 1991) .................... 172

xi

 

Table 1.

Table 2
Table 3.1

Table 3.2

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10.1

Table 10.2

LIST OF TABLES

Demographic Profile of Final Sample ................... 62
Internal Consistency Reliability Estimates ................ 89
Correlations of Individual Dissimilarity Variawa

with Dependent Variables ...................... 92
Correlations of Group Heterogeneity Variables

with Dependent Variables ...................... 93

Hierarchical Regression Results for Job Satisfaction (JS), Predicted
from Overall Satisfaction W/Diversity Programs, Individual
Dissimilarity & Cross-Product ........................ 97
Hierarchical Regression Results for Job Satisfaction (JS), Predicted
from Overall Satisfaction W/Diversity Programs, Group
Heterogeneity & Cross-Product .................. 99
Hierarchical Regression Results for Job Involvement (JI), Predicted
from Overall Satisfaction W/Diversity Programs, Individual
Dissimilarity & Cross-Product ....................... 102
Hierarchical Regression Results for Job Involvement (JI), Predicted
from Overall Satisfaction W/Diversity Programs, Group
Heterogeneity & Cross-Product ...................... 104
Hierarchical Regression Results for Company Commitment (CC),
Predicted from Overall Satisfaction W/Diversity Programs,
Individual Dissimilarity & Cross-Product ............... 106
Hierarchical Regression Results for Company Commitment (CC),
Predicted from Overall Satisfaction W/Diversity Programs, Group
Heterogeneity & Cross-Product ...................... 108
Two-Way Analysis of Variance Showing Interaction Effects of
Individual Dissimilarity Related to Marital Status and Between
Groups Having More Positive or Less Positive Attitudes Towards
Diversity Programs with Job Involvement and Company
Commitment ................................... 115
Two-Way Analysis of Variance Showing Interaction Effects of Group
Heterogeneity Related to Gender, Education, and Marital Status,
and Between Groups Having More Positive or Less Positive Attitudes
Towards Diversity Programs with Job Involvement ........ 116

xii

Table 11 Significant Relationships Between, Overall Satisfaction with Diversity
Programs, Individual Dissimilarity and Group Heterogeneity variables,
and Cross-products with Work-Related Attitudes (Job Satisfaction, Job
Involvement, and Company Commitment) ................. 120

 

CHAPTER ONE
INTRODUCTION AND OVERVIEW

Increasing ethnic and gender diversity of the work force is drawing the attention
of organizations. Solomon (1995) reports that, according to the US. Bureau of
Census, minority workers account for up to 23% of today's workforce, compared to
10.7% in 1964 when Affirmative Action went into effect. These minorities comprise a
much greater variety of ethnicities and cultures now than they did in 1964. The number
of women in the workforce has also grown substantially since the Civil Rights Act was
passed 30 years ago. In 1964, women made up of 34% of the workforce. Today they
account for more than 46% of the workforce and comprise about 42% of all the
managers and professionals in the workplace. The changing composition of the
workforce underscores the importance being given by corporate America to the
management of diversity in the workplace (Solomon, 1995).

Leading consultants, academics, and business leaders in the United States have
urged organizations to respond to the demographic trends predicted initially by the
W (Johnston & Packer, 1987) and by adopting a "valuing
diversity" approach (Copeland, 1988). They point out that a well managed, diverse
work force holds potential competitive advantages for the firm (Cox & Blake, 1991).
Towards this goal, many companies have increased their emphasis on hiring,
promoting, and retaining women, minorities, and other persons with diverse
backgrounds. Organizations have also instituted multicultural training and other
activities to modify organizational systems to address root causes of institutional racism

and sexism. Typically, these efforts have sought to enhance relations between members

2
of different ethnic and racial groups by finding ways to sensitize people to intergroup
differences (Thomas, 1990; Cox, 1991; Ferdman, 1992; Kossek & Zonia, 1992).

Managing workforce diversity is a relatively new concept in management. In a
global economy where technology, capital, and other resources are transferable,
developing human resources provides firms with a competitive advantage (Copeland,
1988; Thomas, 1990). The rationale used to justify creating a multicultural
organization is that such a change will result in many benefits, such as better decision
making, greater creativity and innovation, and increased business competitiveness. The
transition from mass production (individualized) to team-based production
(cooperative/participatory) of the US. industry provides further impetus. Growing
research shows heterogenous groups are more creative, innovative, and make better
quality decisions (Houghton, 1988; Cox, 1991). Another reason is that the changing
labor force demographics might motivate firms to change management practices that
promote and effectively use a diverse work force (Johnston & Packer, 1987; Jackson, et
al. 1991). Further, there is a need to add to our knowledge on the affects of a diverse
work force on organizational performance and effectiveness (Cox, Lobel, & McLeod
1991; Watson et al. 1993).

Generally, organizations have been at various stages of transition with regard to
equal opportunities for women and minorities. These stages have ranged from token
compliance to changing the basic organizational systems, values, climate, and practices.
The inevitability of diversity within the work force should provide the necessary

impetus for organizations to change. For lasting and substantive change to take place,

 

3
organizations must recognize that change in societal and employee expectations is
inevitable, and that legal compliance alone is not sufficient. Company policies and
practices that positively change and impact the experiences of women, minorities, and
other interest groups will enhance the work environment and result in increased
employee development and productivity (Copeland, 1988; Cox and Blake, 1991;
Thomas, 1991).

For systematic change to occur, it is necessary to identify and examine barriers
for the development of pluralism and diversity within organizations. These barriers
may exist at the organizational, managerial/supervisory, and individual levels. At the
organizational level, barriers could include policies, practices, pay and reward systems,
personnel practices, communications, accountability, and philosophy. At the
managerial and supervisory levels, barriers can consist of values, expectations,
attitudes, stereotypes and prejudices about women, minorities, and others. The third
set of barriers might include individual and intrapersonal characteristics of minorities
and women. These could include attitudes, behaviors, expectations, values, the
confidence level of each individual, role definitions, and organizational experiences.
Sensitive management will assess the needs of its particular situation and provide
individual or group efforts that result in the necessary support for affected individuals

(Thomas, 1991; Cox, 1993).1

 

Given the literature on managing diversity (Shipper & Shipper, 1987; Gerber, 1990; Copeland, 1988; Jerich et al. 1989; Cox,
1991; Laden &. Rosner, 1991; and Thomas, l990), the factors involved in managing diversity can be grouped as follows:

1-W1 The cognitive model suggests that diversity is viable as it is concerned with the meaningful utilization of all employees,
i.e., women. minorities, and White males within the organization; and views better employee work attitudes like job satisfaction,
involvement, commitment, etc, as important lay-products. Mamging diversity is concemed with promoting knowledge and acceptance
of cultural differences among different groups of employees, and the recognition of the effects of group identity on interactions. (for
e.g., stereotyping, racism, sexism, and ethnocenterism). It involves valuing the participation and contribution made by women and

4

As culturally diverse groups are likely to become the norm within organizations in
the future, the impact of managing diversity on group interaction and performance will
become critical (Jackson, et al. 1991). In an effort to adapt to a heterogenous work
force, and eliminate subtle barriers that prevent different people (individual
dissimilarity) and different groups (group heterogeneity) from working together and
reaching their full potential, organizations are launching diversity programs/practices
that are designed to create climates that are supportive of a heterogenous work
environment (Braham, 1989; Thomas, 1990; Cox, 1991)’. A major organizational
problem with managing diversity stems not so much from an inability to hire women
and minorities at the entry levels, but from the difficulty in making better use of their
potential, and retaining and promoting them after initial recruitment (Thomas, 1990).

Creating a culture of diversity within organizations is not likely to be easy.

Policies to increase or promote diversity seek to enhance the integration of various

groups (particularly racial ethnic minorities, and White women) into organizations that

 

minorities, leading to perceptions of better employee relations and taking advantage of the opportunities that a diversity of employees
provide.

2%: The diversity process leads to greater attaimnent of individual ego needs that lead to increased morale. motivation, am
satisfaction. Due to these reasons, there is increased knowledge, sensitivity. and promotion and acceptance of cultural differences.

3 .W: Mamging diversity should lead to higher levels of organizational effectiveness. posnitive work behaviors, a decrease m
conflict, prejudice. sexism, and racism, and an increase in employee participation and coopera

4. W: The focus here is on intervening variables on the diversity-outcome relationship, diversity programs/practices.
organization culture, prevailutg organizatioml leadership, prevailing value systems, and bias- free humanre sourceemanag ement systems.

1 Based on the literature on managing diversity, the following can be concluded about diversity programs. Diversity programs and
practices are a set of activities W with one or more of the following intentions: (1) To promote knowledge and
acceptance of cultural, gender, and racial differences; (2) To take advantage of the opportunity that heterogeneity presents; (3) To
facilitate a more effective use of all employee resources within the organization; (4) To better facilitate women an! minorities better
access to resources; (5) To better facilitate women and minorities in sharing organizational power and rewards; (6) To facilitate
perceptions of better employee relations thereby resulting in more positive work attitudes and behaviors: (7) To provide a system of
altermte dispute resolution in the workplace; and (8) To resolve conflict in the workplace.

 

5
have historically been dominated by White men. Changing organizations to become
more diverse is also likely to adversely effect the current dominant group by altering
the distribution of power and resources, as well as changing the dominant goals and
values of the firm (Kossek & Zonia, 1992). By promoting multiculturalism and
accentuating differences between groups that are out-of-balance with their larger
organizational systems, it is likely that members of White male groups might resist such
efforts and engage in increased categorization, labeling, and stereotyping. Similarly,
White women and racial ethnic minorities might tighten their own internal group
boundaries and negatively categorize the White males in power and each other
(Ferdman, 1992; Kossek & Zonia, 1992). Such organizational policies and activities
are likely to have different connotations for each group--with interests that may or may
not overlap--thereby heightening intergroup conflict by creating increased competition
for resources and power, and further highlighting differences in goals and values (Berg
& Smith, 1992). In many cases, member resistance to change hampers efforts to
modify recruitment, promotion, and other policies designed to foster a multi-cultural
work environment (Belfry & Schmidt, 1989).

Part of the challenge of diversity management has been the inability to clearly
articulate what it is and what is not diversity management, that has fueled the current
debate on diversity. Lynch (1994) opposes affirmative action and diversity on the
grounds that it makes it impossible in achieving the goal of a "color blind society." He
believes that Affirmative Action, with its focus on race and gender to the exclusion of

other variables, has resulted in society from moving away from trying to achieve equal

Pp,

‘

6
opportunity and non-discrimination towards individuals to a system of proportional
representation. By limiting diversity to issues of race, ethnicity, and gender, the
concept can easily be dismissed as a "warmed over" version of Equal Employment
Opportunity (EEO) and Affirmative Action (Thomas, 1990). Solomon (1995) points
out the differences between Affirmative Action and diversity management. According
to her, AfflxmatiyeActiQn is driven legally and is about trying to achieve equality of
opportunity by focusing on specific groups. On the other hand, diversity management
focuses on managing and handling the workforce that already exists within each
organization. One key difference is that managing and valuing diversity gives an
organization a competitive advantage. One can view affirmative action and diversity
management as a continuum. Affirmative Action means a company will take positive
steps to ensure that it does not discriminate. Diversity management then comes into
play—-a company promotes a diverse and inclusive workforce proactively.

A survey conducted by the Society for Human Resource Management and
Commerce Clearing House, Inc. (SHRM/CCH) in 1993 highlighted some of the
concerns associated with diversity management. Data were solicited from a wide
variety of human resource professionals concerning the perceived attitudes of members
of their organizations, as well as to obtain an understanding of the status of practices
related to diversity. 785 national SHRM members participated in the survey. The
results of the survey indicated that, in terms of priorities, issues related to diversity

were not that important, and that the link between diversity and financial success was

7

not clear. The survey also revealed that the increased focus on diversity has led to a
growing alienation of White males within organizations.

Many companies are facing a quandary that their efforts to recruit a more
diverse workforce is likely to lead to a backlash from White males who might be
"passed over. " Yet, if companies fail to embrace a more diverse workforce then they
risk- becoming vulnerable to global competition. W (1994) magazine ran
a cover story highlighting some of the concerns felt by White males and their
frustrations with diversity management. Many White males feel that, in the rush for a
more diverse workplace, they will lose out to less qualified workers. Further
exacerbating the problem is the perception by some White males that diversity
accommodation efforts are tantamount to preferential treatments for some groups. On
the other hand, women and minorities complain that some White male managers try to
undermine their credibility by attributing their rise as "tokenism. " The dilemma for
companies committed to a corporate culture that embraces diversity is twofold: first,
how to ensure a diverse workforce without antagonizing White males (whose support is
critical for change) or women and minorities (who may resent efforts to win over White
males); and, second, how to reverse historical discrimination without creating new
forms of it.

There have been a number of instances where diversity training workshops have
led to greater race and gender divisions with each group vying with each other, instead
of promoting harmony (Caudron, 1993). Exposure of potential prejudice in some of

these workshops have led to sex and race discrimination law suits, where evidence has

8
been cited as management bias. Some of the other problems with diversity programs
and practices include confrontational trainers who accuse employees of bigotry or pit
them against each other, resentment by employees who feel left out of the process, and
a sense among many executives that training sessions alone are inadequate (Them
W, Sunday, August 1, 1993). In its July 5th, 1993 cover story, Them
Republic criticized the "diversity industry " for charging millions of dollars for
programs that did not always "solve the problems they purport to address"
(MacDonald, 1993).3 Negative outcomes can include the possibility of post-training
participant discomfort, reinforcement of group stereotypes, perceived
disenfranchisement, backlash from White males, and lawsuits based on managers'
exposure of stereotypical beliefs exposed during "awareness-raising" sessions (Rynes &
Rosen, 1994).

One problem that causes ineffective diversity programs is that companies simply
take a prepackaged workshop or program and try to graft it directly into their corporate
culture. Another problem is that decision makers in companies are often middle-aged
White males. Perhaps some of the backlash problems are the result of the exclusion of

White males from the process. How an organization communicates the needed changes

 

Diversity efforts currently in effect in organizations are:

l. Interpersonal training and education, such as cultural awareness training. educational programs targeting women
and minorities;

2. Strategic planning and monitoring, such as multicultural participation councils, and cultural board of directors;

3. Human resources employee recmitmenr, retention, and career planning programs targeting women and minorities;
and

4. Benefit programs, such as alternate work schedules, elder care, child care etc.

A

9
to support a diverse workforce may also contribute to the backlash they experience.
There is a need for a broader definition of diversity that is inclusive, and a need to link
diversity management with business success. Instead of focusing on the shortcomings
of traditional corporate cultures that support White male values, organizations must
emphasize that the changes in society require changes in business strategies. Without
such efforts, mandates to foster diversity are likely to fail (Kossek & Zonia, 1992;
SHRM/CCH Survey, 1993). Therefore, the perceptions that each group is likely to
have towards the management of diversity will be critical in the successful
implementation of diversity programs and practices. The value (usefulness and
importance of diversity programs/practices) that different employees and groups hold
about these programs/practices is also likely to affect work-related attitudes.
RESEARCH NEED

Previously, a large number of studies have focused on issues related to race and
gender in organizations (Riger & Galligan, 1980; Donnell & Hall, 1980; Morrison et
al. 1987; Howard & Bray, 1988). A number of studies have documented the negative
experience of women and minorities within organizations (Kanter, 1977; Thomas &
Alderfer, 1989; Ilgen & Youtz, 1986; Greenhaus et al. 1990; Bell, 1990). However,
the importance of studying the management of workforce diversity from a strategic and
competitive advantage has emerged only during the last five years. Generally,
researchers studying this issue have focused on the theoretical concepts of diversity,
and how it can be linked to competitive advantage and organizational effectiveness.

But, there have been very few empirical studies making such a link explicit. For

10

example, Cox et al. (1991) and Watson et al. (1993) have studied the effects of a
diverse workforce on organizational performance and effectiveness. The study by
Kossek and Zonia (1992) has focused on diversity climate. Research by Pfeffer (1983),
Schneider (1987), and Jackson et al. (1991) support the view that information about a
person's demographic characteristics influences both attributes regarding the person's
psychological character and behavior towards the person and his/her membership
within the group. They contend that demographic attributes are likely to determine
both the perceptions of similarity/dissimilarity and perceptions of person-environment
fit. Based on this assumption, one of the premises for implementing diversity programs
and practices is that these programs are likely to facilitate the adjustment of dissimilar
individuals and heterogenous groups within organizations and effect work-related
attitudes.

However, no studies have examined the effects of diversity programs in the
workplace, although there is growing anecdotal evidence and case literature
documenting the mixed impact of diversity programs. The experience of some
companies has been that diversity training programs do more harm than good. On the
other hand, there is evidence that companies, which have adopted and embraced a
broader definition of diversity that is inclusive of everyone, have reported considerable

success. This assessment on the management of diversity suggests there is a mug:

 

1 1

There are other problems associated with studying cultural diversity within
organizations. Cox (1990) argues that there are serious obstacles, such as sample sizes
of racially and ethnically gender diverse groups that fall below conventional levels, that
hamper research in this area. Another obstacle is the reluctance of organizations to
participate in diversity research due to the fear of legal problems. Because of these
reasons, the specific link between organizational efforts to manage diversity and work-
related attitudes is not explicit. There is a paucity of actual research data supporting
such a link. There is also very little research supporting managing diversity as a
competitive resource. However, there is some research that supports the benefits to
organizations that invest in special programs and human resource practices targeting
women and minorities. Recently, growing research on creativity in groups has
generally supported the notion that heterogeneity along a variety of dimensions leads to
original and high quality ideas and problem solutions. This further highlights the need
for programs and practices that facilitate the working of heterogenous workgroups.

DEFINITIONS OF DIVERSITY

Philosophically, Shipper and Shipper (1987) conceptualize diversity and
pluralism as a prerequisite to the dynamic regeneration of an organization. According
to them, this concept represents a change from the remedial idea of equal opportunity

and affirmative action to the positive concept and benefits of a truly mixed work force

at all levels. They define diversity and pluralism as a W

 

 

"ll .1. I III] 'lfil IE1] . 1
12mm.

Loden and Rosner (1991) have focused on the attributes of diversity. From an
objective viewpoint, diversity is the vast array of physical and cultural differences
among people. From a subjective viewpoint, diversity is "otherness" or those human
qualities that are different from one's own and outside the group to which one belongs.
These authors also discuss the primary and secondary dimensions of diversity. Primary
dimensions are defined as those immutable human differences that are inborn and exert
an ongoing impact throughout peoples lives. These dimensions include age, ethnicity,
gender, physical abilities and qualities, race, and sexual/affectional orientation. The
primary dimensions serve as interdependent, core elements. Secondary dimensions of
diversity are those that can be changed. They are mutable differences that people
acquire, discard, and/or modify throughout their lives. The secondary dimensions
include, but are not limited to, educational background, geographic location, income,
marital status, military experience, parental status, religious and political beliefs, and
work experience. The view of these authors also fosters a management approach that
makes full use of the ideas, talents, experiences, and perspectives of all employees at all
levels of an organization.

Jerich et al. (1989) also focus upon attributes, ranging from differences that can
usually be concealed, such as a particular lifestyle, to differences that cannot be
concealed, such as gender and race. Diversity, therefore, is an umbrella term that

encompasses such employees as single parents, gays, religious fundamentalists, the

13

handicapped, and older workers to name a few. It can also refer to work styles within
the company, such as sales people or accountants, that can be subjected to stereotypes
typical of that profession. In this sense, diversity is an extremely broad term that
people use to refer to all sorts of differences. According to this viewpoint, just about
everyone has a characteristic that makes him/her a bit different from the other.

Although Equal Employment Opportunity (EEO) and Affirmative Action form
the moral social impetus and legal basis for treating people fairly, Copeland (1988)
argues that one of its consequences has been that people have resisted talking about
differences because these differences have implied that someone or something was
either inferior or superior. In this sense, valuing diversity is new in that it begins to
address these differences. Such an approach allows addressing the fact, that in addition
to stereotypes and prejudice, there are real cultural differences that cause
miscommunication and conflicts, which then further compound the stereotype and the
prejudice. Therefore, according to this viewpoint, the underlying premise in managing
heterogeneity effectively is W the differences.
Such an understanding would allow individuals to make their maximum contribution
regardless of gender, color, and race.

Gerber (1990) states that the philosophy of managing diversity suggests that
organizations make whatever changes that are necessary in their systems, structures,
and management practices to eliminate any subtle barriers that might keep people from
reaching their full potential. The goal is to treat people as individuals, recognizing that

each employee has different needs and will need different kinds of help to succeed.

14

This approach is likely to be a big philosophical shift for some organizations, where
paying attention to differences is the antithesis of the melting pot philosophy or, in
other words, fitting in with the dominant White male culture that permeates most
organizations.

Thomas (1991) also takes a systems approach and defines managing diversity as
a comprehensive managerial process for developing an environment that works for all
employees. This process allows organizations to evolve and develop steps that tap the
potential of all employees, women, minorities, including White males. Similarly, Cox
(1991; 1993) has defined "managing diversity " as a variety of human resource
management issues and activities pertaining to the hiring and effective utilization of
personnel from different cultural backgrounds for the purpose of creating a competitive
advantage. He links managing diversity with the business strategy of the firm and with
strategic human resource management. Thus, managing diversity becomes a means for

building community within organizations.

PURPOSE OF THE STUDY

 

mmlmntmdnrgamzahnnalmmrtment Research by Jackson et al. (1991) and
others support the view that information about a person's demographic characteristics
influences both attributions regarding the person's psychological character and behavior

towards the person and his/her membership within the group. Demographic attributes

 

15
are also likely to determine both perceptions of similarity/dissimilarity and perceptions
of person—environment fit. One of the premises for implementing diversity programs
and practices are that these programs are likely to increase the adjustment of dissimilar
individuals and heterogenous groups within organizations. Also, the perceptions,
attitudes, and values that employees hold about these programs/practices are likely to
affect the successful implementation of employer policies towards diversity. Employee
attitudes and behaviors are one component of organizational effectiveness (Kanter &
Birkenhoff, 1982). This research will help identify issues that suggest differences

between different groups and how they perceive managing diversity.

 

drxersitxnmgramslnractiremmanagingdiyersity. The research will attempt to

extend the research by Jackson et al. (1991) on individual dissimilarity and group
heterogeneity as correlates of turnover to work-related attitudes. The authors indicated
the need for research regarding how situational incentives might be used to inhibit
dysfunctional similarity-dissimilarity biases in group processes. This research will
therefore investigate whether diversity programs and practices as situational incentives
help to moderate or overcome similarity-dissimilarity biases. In addition, the research
will study how an individual employee's perceptions of organizational efforts to manage
diversity affects the impact of individual dissimilarity and group heterogeneity on work-

related attitudes.

 

16
THE RESEARCH CONTEXT: THE BANKING INDUSTRY

The banking industry (Standard Industrial Code 602) is a central part of the
US. economy and directly affects the lives of nearly all consumers and the operations
of almost every business. They are several types of banking institutions, commercial
banks, savings and loans, and credit unions, to name a few. All these institutions offer
a wide variety of financial services, including making loans, accepting deposits,
clearing checks, and conducting more complex operations. The employment patterns in
this industry are dominated by women. Women comprise 70 percent of the workforce
in this industry and occupy most of the lower paying jobs. Consequently, banking
institutions see the need for instituting diversity programs that are responsive to the
needs of a largely female workforce. For example, a number of banks have developed
flexible work schedules for employees to help them balance work and family
obligations. Because of these factors, banking institutions are suitable for studying the
effects of diversity programs on work-related attitudes.

Despite the many changes in the economic and legislative climate for banking
institutions in the latter part of the 1980s (due to federal legislation, deregulation,
mergers and acquisitions, commercial bank and savings & loans failures), the industry
maintained its traditional occupational staffing patterns (Kuster, 1993). Annual
employment in the banking sector has increased considerably. Commercial banks
employed 1,552.7 thousand employees in 1989, an increase of 9.9 percent from 1982.
Federal savings institutions employed 244.5 thousand employees in 1989, representing

a growth of 61.9 percent employment from 1982. And, credit unions employed 130.1

 

17

thousand employees in 1989, an increase of 5.8 percent from 1988 (BLS, Employment,
Hours, and Earnings, 1991). The majority of the employees in this sector are females.
In 1989, women constituted 68.8 percent of the total employment. 77 percent of the
employees were White, whereas 22.9 percent were minorities (EEOC Reports, 1989).

Many banking institutions responded to the deregulated environment of the
19805 by speeding up the implementation of new technologies and processes to
streamline operations. Of particular importance, is the growing use of automation and
computer technology to achieve productivity gains. New machines and technologies
permitted greater automation of many positions and operations such as bookkeeping.
As computer terminals were moved to the teller window, bank transactions could be
entered directly into the computer, increasing speed and efficiency of the operations.
Computerization lessened the need for a more educated and trained workforce. The
proliferation of automated teller machines (ATMs) also reduced the need for labor, thus
decreasing operating costs. Customer services, such as deposits and withdrawal of cash
and checks which form a large part of the day-to-day functioning of the banks, could be
easily handled by tellers due to computerization. Tellers form the largest occupation in
this industry. Employment of tellers has also increased due to the increased use of part-
tirne, "peak hour" or prime-time tellers (Kuster, 1993). This occupational category
also receives low wages as compared to other comparable clerical work elsewhere.
Clerical and administrative support occupations as a whole also form the largest

occupational group in the banking industry.

 

18

Although organizational characteristics and other details of each participating
institution will be described, the main focus of the study will be limited to the
individual perceptions of diversity programs and their effect on work-related attitudes.
The research focus will be on indiyiduals who make up the organizations, their
perceptions regarding diversity programs, and their relationship to work-related
attitudes. In this study, three financial institutions representing the industry will be
studied: two mm and a W.

RESEARCH METHODS

The objective of this research is to study the effects of diversity programs on
work-related attitudes. The research focuses on individuals who make up the
organization, their perceptions regarding diversity programs, and their relationship to
work-related attitudes. There are three key objectives to this study:

1.

 

The impetus for the study resulted, in part, from theoretical arguments made by

scholars on the importance of managing diversity and the effective use of people from
different cultural backgrounds to create competitive advantage (Copeland, 1988; Loden

& Rosner, 1991; Thomas, 1991; Cox, 1991, 1993). Another reason for this research is

 

19
to investigate the effects of diversity programs and whether they facilitate the
adjustment of dissimilar individuals and heterogenous groups within organizations,
thereby effecting work-related attitudes of these individuals and groups. Based on this
rationale, diversity programs should help to overcome or moderate similarity-
dissimilarity biases. Specifically, diversity programs and practices would correlate
positively with work-related attitudes.

Data will be collected by means of a survey questionnaire. A hierarchical
regression or moderated regression method will be used to test whether the relationship
between individual dissimilarity and group heterogeneity and work-related attitudes--job
satisfaction, job involvement, and organizational commitment--is moderated by an
individual's perceptions of diversity programs/practices. Hierarchical regression
analysis permits assessment of whether each variable significantly predicts the
dependent variable with variance due to other independent variables controlled (Cohen
& Cohen, 1983). The regression involves three steps; each step will be used to predict
variance in work-related attitudes. Support for the hypotheses is demonstrated by the
amount of variance explained by each step and by the cross-product of the interaction.

CONTRIBUTIONS AND LIMITATIONS OF THIS
DISSERTATION

This dissertation will assess the extent to which an individual's perceptions of
organizational efforts to manage diversity effect work-related attitudes. At present,
there is very little empirical literature on the dynamics of culturally diverse work
groups, the affects of diversity on performance of work groups, and even less literature

on the management of such groups (Kirchmeyer & Cohen, 1992). Also, there is

 

20
paucity of information regarding issues related to diversity. Cox and Blake (1991)
discuss the need for research on the career experiences of different cultural groups,
differences of opinion about the value of diversity based on different cultural groups,
areas where changes are needed, and clues about how to make them to name a few.
Research is also needed to evaluate the change effort and to provide baseline data on
key indicators of a valuing diversity environment. This research is therefore designed
to assess the meaning and interpretation of managing diversity to individuals and
groups.

Pfeffer (1983) and other researchers (Wagner et al. 1984; O'Reilly et al. 1989;
and Jackson et al. 1991) have mainly researched organizational demography as
correlates of turnover. This research will borrow their perspective and further
investigate the affects of individual dissimilarity and group heterogeneity on work-
related attitudes, thereby extending their knowledge base. Jackson et al. (1991) also
indicates there is need for research regarding how situational incentives might be used
to inhibit dysfunctional similarity—dissimilarity biases in group processes. This research
will therefore study whether diversity programs and practices as situational incentives
help to moderate or overcome similarity-dissimilarity biases. Moreover, most studies
(Wagner et al. 1984; Murray, 1989; and Jackson et al. 1991) on individual dissimilarity
and group heterogeneity have involved top-management teams. This research will
involve all levels of the organization, hence making research findings more

generalizable to the population.

21

Practitioners will also benefit from this dissertation. The research results will
have implications for diversity training. By identifying key indicators and baseline data
on employee perceptions of organizational efforts to manage diversity, new programs
can be instituted, current programs can be evaluated, and changes can be made to meet
the needs of a changing workforce. Diversity management has implications also for
human resource management practices. The research will identify organizational
barriers towards diversity that prevent the effective use of women, minorities, and
White males in organizations. With its focus on organizational culture and climate
change and effective utilization of all employees to create competitive advantage,
research in this area will further help integrate human resource management within the
strategic management of the firm.

Finally, diversity research has implications for the theory and practice of
community development. As communities become more demographically diverse,
reconciling the needs of heterogenous groups becomes important. The challenge faced
by many communities is to retain commonly shared values and norms in the face of the
needs and demands of diverse groups. This research hopes to provide data on group
processes and interactions of heterogenous groups, that might be helpful in "building
community" without losing the benefits of diversity.

KEY ASSUMPTIONS
This research is grounded in the following assumptions:

(1) As organizations become more diverse, they will bear increasing costs

22
associated with integrating workers who are dissimilar. One of the consequences may
be poor work-related attitudes and behaviors.

(2) An individual employee's values and needs are relatively stable. This
means that, unless an event occurs to change attitudes and behavior, these attitudes and
behaviors are also likely to remain constant.

(3) Organizations, which have initiated a variety of programs and practices
related to the hiring and effective utilization of personnel who are dissimilar and
heterogenous, are likely to experience a number of benefits. These benefits include an
improved level of creativity, better decision making and problem solving, and positive
work-related attitudes and behaviors.

(4) Human resource management programs on diversity have value, not only in
terms of direct benefit of successful programs and practices, but also in terms of an
indirect beneficial impact on employee perceptions. This, in part, might lead to
positive employee attitudes and behaviors (e.g., job satisfaction, job involvement,
organizational commitment).

(5) The nature of the moderating relationship of these programs and practices is
likely to be positive when organizations are perceived by employees to be making
" good efforts" to effectively integrate and utilize people from dissimilar and
heterogenous backgrounds.

ORGANIZATION OF THIS DISSERTATION
This dissertation contains five chapters. Chapter One includes the rationale

underlying the research objectives, and information about the potential contributions of

23
this dissertation. Chapter Two includes a focused review of the diversity literature.
The purpose of this discussion is to highlight the issues related to the management of
diversity in the workplace. The method of investigation is presented in Chapter Three.
It includes the organizations under study, the subjects for this dissertation, the data
collection procedure, the operationalization of variables, and the method of data
analysis for each hypothesis.

The results of the data analysis are presented in Chapter Four. Results of the
relationship between individual dissimilarity and work-related attitudes are presented
first. Next, results of the relationship between group heterogeneity and work-related
attitudes are presented. This chapter also includes the hierarchical regression results
and the differential effects of individual dissimilarity and group heterogeneity, and
explains some of the interactions and processes that have occurred when these programs
were put in place. The conclusions and suggestions for future research in managing

workforce diversity are presented in Chapter Five.

 

CHAPTER TWO

(‘7
“I

REVIEW OF LITERATURE

As stated in Chapter One, the purpose of this dissertation is to investigate the

effect of diversity programs/practices on work-related attitudes (job satisfaction, job

involvement, and organizational commitment). This research will test whether the
relationship among individual dissimilarity, group heterogeneity, and work-related
attitudes are likely to be moderated by an individual's perceptions of organizational
efforts in implementing diversity programs/practices on managing diversity. The
research seeks to answer three major questions:

1. To what extent does an individual's perceptions of diversity programs/practices
affect the impact of individual similarity or dissimilarity on work-related
attitudes?

2. To what extent does an individual's perceptions of diversity programs/practices

affect the impact of group homogeneity or heterogeneity on work-related

attitudes?

3. To what extent does the experience of diversity programs/practices affect
individual and group interaction and work—related attitudes?

MANAGING WORKFORCE DIVERSITY
The term ‘managing diversity' has been very loosely defined in the literature.
It encompasses managing race and gender within organizations, employee attitudes and
values in the work place, and work practices such as total quality management (TQM)
among others (Thomas, 1991). Managing diversity has also been defined as a
comprehensive managerial process for developing an environment that works for all
employees (Thomas, 1990). It refers to a variety of human resource management

issues and activities associated with hiring and working with personnel from different

24

25
cultural backgrounds (Cox & Blake, 1991). Concerns characterize this issue, such as
how to merge cultures successfully, how to develop interventions that influence the
corporate culture, and how to facilitate the acceptance of women and minorities in
senior positions. This process allows organizations to evolve and develop steps that tap
the potential of all employees--women, minorities, and White males as well. According
to Thomas (1990), managing diversity is a new way of thinking about human resources.
It is different from the traditional business perspective of looking at race and gender
through the lens of Affirmative Action and Equal Employment Opportunity (EEO). It
addresses the many ways that employees are different and similar. It assumes that
adaptation is a mutual process between the individual and organization.

This view is a change from the usual assimilation approach where the burden of
adapting rested solely on the individual who is different. Therefore, managing
diversity approaches diversity from a management perspective. It deals with the way
organizations are managed. It is grounded in the very specific definition of "managing"
i.e., creating an environment that allows the people being managed to reach their full
potential (empowering the work force), linking human resource management with the
strategic needs of business. According to Kossek & Zonia (1992), culture and climate
are likely to influence perceptions of organizational events. The climate for diversity is
likely to be influenced by the way in which organizational policies are carried out at the
unit level. Specifically, policies pertaining to the distribution of organizational
resources and the recruitment and promotion of women and minorities are likely to

influence the perceptions of a diversity climate and culture. Consequently, the focus is

 

26

on the individual, interpersonal, and organizational levels in terms of looking at the
way organizations are managed and structured.

Historically, women and minorities have been excluded from the middle and
upper levels of many organizations (Morrison et al. 1987; Greenhaus et al. 1990).
While Equal Employment Opportunity laws and Affirmative Action programs have
been implemented to stop discrimination in employment, institutional racism and
sexism still persist in the workplace (Kanter, 1977; Fernandez, 1981; Ilgen & Youtz,

1986; Porras, 1991).4

 

In the past. there have been a number of theories that explain the differential treatment of women and minorities in management.
Psychological researchers, like Riger and Galligan (1980), have generally emphasized person-centered variables to explain women's low
job status. Women's traits, behaviors, attitudes, and socialization are cited as some of the reasons that make them inappropriate or
deficient as managers. These factors supposedly make women fear success and cause them to be less willing to take risks. Other
investigations imo sex differences have generally refitted this explanation. Researchers have presented considerable evidence that shows
women and men in management roles have similar aspirations, values, and personality traits as well as job-related skills and behaviors
(Dobbins and Platz, I986; Harlan & Weiss, 1981; Morrison et al. 1987). Howard and Bray (1988) on their research on AT&T
Assessment Centers present dam disputing both sex and race deficiencies. They found race differences to be greater than sex differences.

The human capital theorists, have attempted to explain continued sex and race-related differences in management by
suggesting that individuals are rewarded in their current jobs for their past investment in education and job training (Blau & Ferber,
1987). Workers may chose to accept a wage, or to invest in acquiring new skills and experiences, to qualify for higher paying jobs.
According to this theory, any policy changes that may be used to correct differential treatment should be directed at the educational
process nther than the employment setting, because no difference other than those in human capital are seen. This explanation assumes
that investment pays off equally for all groups. However, recent studies have shown that there are differential payoffs for White males
as compared to women and minorities. Larwood et al. (I984). (1987) have shown that education] levels do not fully account for
discrepancies In job level or pay in studies of sex and race in organizations.

The discrimination theorists suggest bias on the part of the dominant group as a basis for differential treatment. Blau and
Ferber U987) offer a labor nnrket discrimimtion explamtion and point out that employers with discriminatory tastes hire women when
there Is a wage discount large enough to compensate for the loss of utility or level of discomfort associated with employing them. The
dual market theory proposed by Larwood and Gattiker (1987) states that the labor Inarket consists of a set of better or primary jobs, and
a set of worse or secondary jobs with little mobility between the two. Women and minorities are groups most likely to be associated
with the secondary labor market and are oflen confined there, anti discrimimtion is often justified as economic efficiency. Other studies
suggest tint deficiencies are also presumed due to die persistence of stereotypes based upon historical roles (13er et al. 1984: Powell,
1988; Thomas & Alderfer, 1989).

Stnrctural discrimination is suggeswd as another set of barriers to the advamement of women an! minorities in the workforce.
The intergroup theory advanced by Alderfer & Thomas (1989) suggests that two types of groups ex'm in organizations—identity groups
based on race, ethnicity, family, gender, or age, and organization groups based on common work tasks, work experiences and position
in the hierarchy. Tensions may result because organizatioml membership changes, whereas ideraity group membership does not. When
the pattern of group relations within an organization mirrors the pattern in society as a whole. such as Whites predominating' In high
status positions am! Blacks concentrated in low status jobs, then the evaluation of Blacks is likely to be distorted by prejudice or anxiety
as racist assumptions go unquestioned in the organization.

An organizational structure perspective is also taken by Kanter (1977) in her classic research on men and women in
organizations. She emphasizes that women' s lack of opportunity and power in organizations and the sex ratio of groups within
organizations explain the lack of managerial success for women in management. The impact of structural factors Is also by
Irons and Moore (1985) on their study of the banking industry. They identified the three most significant problems faced by Blacks:
not knowing what is going on in the organization or not being in the network; racism, and inability to get a mentor. Ilgen & Youtz
(1986) take a career developan perspective. They suggest tint the interaction of situatioml factors with person centered characteristics
account for some of the differential treatment For example, without opportimities to take challenging assignments, minority managers
may fall behind their White cohorts In terms of knowledge and skill development, or they may internalize negative evaluations and

K

27

Consequently, the current emphasis on diversity is likely to be welcomed by women
and minorities (Kossek & Zonia, 1992). Another reason for creating multicultural
organizations is that change might result in many benefits, such as better decision
making, greater creativity, more innovation, and increased business competitiveness
(Houghton, 1988; Cox, 1991). Changing organizations towards a multicultural model
will not only involve changing policies and practices, but will also mean changing the
way in which power and rewards are distributed across gender, racial, and ethnic
groups. For example, if White men are currently in favored organizational positions,
then they are likely to be the group most negatively affected by the implementation of
diversity policies and practices and, consequently, may view such policies negatively.
They may hold less positive attitudes towards efforts to increase organizational
diversity-partly because such efforts to change the status quo might decrease their
current positions of power and their acquisitions of rewards (Cox, 1991; Alderfer,
1991).

In recent years, workplace diversity management and training programs have
become some of the "hottest" human resources issues around as companies are search
for ways to get the most productivity from their diverse workforces. U.S. Employers

increasingly want to provide diversity training. Diversity represents one of the few

 

stereotypes to the point where they limit themselves and turn down future opportunities for fear they will not succeed. This litre of
research seems to point out that the extent to which organizational structures and practices follow models based solely on how White
men develop, women and minorities are likely to he disadvantaged.

Minorities also struggle with biculturalism or fitting into two distinct cultural worlds (Alderfer and Thomas, 1989). Bell's
(1990) research on bicultural conflict among Black women shows that those from cultures other than that of the domimm group must
chose how to manage the stress of moving physically. cognitively, anti emotionally between the two cultural systems. These results
concur with those of Femandez (1981) who found minorities are excluded from informal work groups.

 

28
areas of training that is growing during a time of shrinking training budgets. A 1991
survey conducted by New York City—based Towers Perrin revealed that 75% of the
companies reported that they may plan or currently have diversity-training programs in
place (Caudron, 1993). Diversity management training programs have usually been
dominated by one or two basic approaches: increasing awareness of policy or
increasing sensitivity to the concerns of people in different groups than one‘s own.
Programs dealing with policy focus almost entirely on legal issues concerning diversity.
They are designed to inform managers (primarily) about equal-employment—opportunity
laws and their applications, as well as affirmative action rules. The objective is to
demonstrate a company's commitment to equal opportunity and ensure that managers
understand why they must comply with policies. The second approach focuses on
increasing the trainee's sensitivity to diversity. Often, these programs try to heighten
the individual's awareness of what it is like to be misunderstood, undervalued, and
stereotyped at work. A great deal of awareness training is designed specifically to
sensitize the White male manager to the changing complexion of the workforce, and to
help him develop more "appropriate" and creative ways of working with his
subordinates (Karp & Sutton, 1993).

This approach of sensitizing the "White male manager" has led to increasing
criticisms of diversity training. That is because very often diversity training turns into
a "curious mix" of psychology and anthropology--the evils of stereotyping are preached
while values of growing ethnicity and other types of diversity within the workforce are

exalted. As many of these issues essentially deal with personality, people's thoughts,

 

29
feelings, behaviors, they are sensitive and potentially touchy issues. Such issues cannot
be dealt over a period of one or two sessions. Diversity efforts often have as their tacit
objective the modification of behavior. When managed effectively, diversity is a long-
term project. The objective is to a develop a culture that breeds harmony and rapport
which, in turn, provides creativity, effective decision making, and better teamwork.
Seeing cultural diversity as the emotional and fervent social issue that it is will help
companies to understand its complexities and potential pitfalls (Thomas, 1994).

Evaluating diversity programs and the factors that cause them to succeed or fail
is important for several reasons. One is the cost of training and, sometimes, the
negative outcomes from the diversity training itself. Negative outcomes include the
possibility of post-training participant discomfort, reinforcement of group stereotypes,
perceived disenfranchisement, backlash from White males, and even lawsuits based on
managers' exposure of stereotypical beliefs exposed during "awareness-raising"
sessions (Rynes & Rosen, 1994).

Caudron (1993) argues that unqualified trainers are not the only cause of the
diversity "horror stories. " For many employers, diversity programs are thoughtful
responses to changing demographics and business needs. For others, the training may
be nothing more than an impulse to participate in the latest "corporate fad." These
companies sandwich diversity training between other contemporary workplace
programs without giving any thought to the importance of developing a comprehensive

workplace diversity policy. On its own, training usually fails unless there are other

30

strategies to support it. Ineffective training can hamper or set back an organization's
efforts to support diversity and use it as a business advantage.

According to Caudron (1993), had training can raise the expectations of women
and minorities in organizations while increasing the fear of Caucasian males. Bad
training also can give managers the wrong clues about what to do in a given situation,
such as stop hiring Caucasian males. The pain that comes from prejudice can
intimidate employers and make them reluctant to deal with diversity again. Ineffective
training can take many forms. One focuses on "fixing the victim" where the focus is on
assimilation into the corporate culture. Another problem occurs when trainers try to
enforce politically correct language into workshop participation. This prevents real
issues from reaching the table. Another reason why training fails is the failure of
companies to understand the issues facing employees. Finally, training is likely to be
ineffective when it focuses on confronting stereotypes without giving corresponding
emphasis to developing the skills needed to bring awareness of these stereotypes back to
the workplace.

To succeed, organizational infrastructures have to be created in the organization
to support the change that is taking place. Companies that have been successful, such
as Avon Products Inc., Prudential, and Hewlett-Packard Co., and Ortho
Pharmaceuticals, know that training is only one part of the diversity equation. For
training to be successful, there also needs to be an environment that supports diversity
in ways that go beyond the classroom. To begin with, diversity efforts must have

commitment and direction from the management. Upper management must understand

31

the value of a diverse workforce, direct the strategies to support diversity, and model
the appropriate behaviors (Caudron, 1993). For example, Nestle's success with its
diversity program was due, in part, to tying a research study to the design of its overall
diversity training and effort. The company felt that an accurate understanding of
peoples' perceptions was necessary to ensure that a great deal of energy would not be
wasted on issues with little payoff. The research helped to focus diversity efforts on
areas that would be of greatest benefit to the company, such as performance
management and informal networking. It also helped determine what focus, language,
and benefits would be most likely to generate commitment from the employees at all
levels. The study helped to discover company values. These included teamwork,
innovation, and genuine care for employees (Morris, 1994).

Rynes and Rosen (1994) found that two other characteristics of the training itself
were significant predictors of overall success: length of training and amount of post-
training evaluation and follow—up. Beyond characteristics of the training itself, two
other factors were statistically associated with reported program success. These were
top management support and the extent to which managers were explicitly rewarded for
supporting diversity. Top management support was the single most important predictor
of training success--more than any characteristic of the training itself. One of the major
ways to which top management indicated strong support for diversity was by linking it
with other key strategic priorities. Caudron (1993) suggests that diversity training
should be connected with specific business outcomes and results, whether it is to

enhance team relationships, increase productivity, cut down absenteeism, or reduce

 

32
turnover. In the absence of strategic linkages, diversity is likely to stay in the limelight
for only a short period of time before being crowded out by other issues. While one
might conclude that many organizations are not receiving the long-term benefits hoped
for from diversity training, training is not at fault. Rather, it is more likely that
training is not succeeding because it is being carried out in an environment that places
low priority on diversity issues (Rynes & Rosen, 1994).
REVIEW OF RELEVANT THEORIES

Organizations are beginning to realize that the diversity of the workforce may
change the patterns of behavior that were established during an era when organizations
and workgroups were relatively homogenous (Jackson, 1991). Schneider's (1987)
attraction-selection-attrition (ASA) model and Pfeffer's (1983) organizational
demography model provide theoretical perspectives that increase an understanding of
groups in organizational settings. Schneider (1983) argues that, through the processes
of attraction, selection, and attrition, organizations evolve towards a state of
interpersonal homogeneity. His interactionist framework proposes that organizations
are functions of the kinds of people they contain, and further, that employees are
functions of an attraction-selection-attrition (ASA) cycle. The ASA cycle is proposed
as an alternative model for understanding organizations, and the causes of the
structures, processes, and technology of organizations, in contrast to the Situationist
interpretations. Schneider's main thesis is that the attributes of people--not the nature

of the external environment, the organizational technology, or the organizational

33

structure--are the fundamental determinants of organizational behavior. This
framework builds on the interactionist assumption that, as a result of natural person-
person interaction, person and situation are frequently integrated. How person and
setting come to be relatively integrated is presented as the outcome of a cycle of
attraction-selection-attrition. Bowers (1973) and Weick (1979) have presented similar
arguments by suggesting that persons are inseparable from environments because most
environments only exist through the people behaving in them and "knowing" them.
More specifically, interactionism argues that situations are as much a function of the
person as the person's behavior is a function of the situation (Bowers, 1973).

Further, Schneider (1987) argues that the kinds of persons in environments
determine the kinds of human environments they are. Although on the surface it might
look as if organizations determine behavior, in reality the way it looks is a result of the
people behaving the way they do. They behave the way they do because they were

' . Different kinds of

 

organizations attract, select, and retain different kinds of people, and it is the outcome
of the ASA cycle that determines why organizations look and feel different to each
other. The theoretical literature in the field of vocational psychology posits that people
are attracted differentially to careers as a function of their own interests and
personalities. The literature on occupational entry and organizational choice suggests
that people are attracted differentially by the attributes and characteristics of particular
kinds of career and work environments. Holland (1976) proposes that occupations can

be clustered into types, and that these types are useful for characterizing both people

34

and work environments. According to him, vocational choice is assumed to be the
result of a person's type or patterning of types and the environment. The character of
an environment emanates from the types of people who dominate that environment. In
essence, he proposes that career environments are characterized by the kind of people
in them, and that people choose to be in environments of a type similar to their own.
Vroom (1966) and Tom (1971) both show that people choose organizations because
they believe that W will be most instrumental in obtaining their valued
outcomes. Their studies support the ideas that peoples' own characteristics (self-image,
desires) are predictive of the kinds of settings to which they will be attracted.

The literature on turnover indicates satisfaction and attrition are meaningfully,
but negatively, related. Studies indicate that people who do not fit an environment will
tend to leave it (Price, 1977; Mobley, 1982; Staw, 1984). The work of Price and
Mobley respectively, emphasizes the critical role of job attitudes and intentions to quit,
and Porter et al. (1974) examine turnover in relation to organizational commitment.
The interactionist perspective suggests that the nature of jobs, interpersonal
relationships, and the reward systems that organizations display to workers must fit
their needs. Otherwise attrition will follow. This perspective further suggests that
organizations, on the surface, might appear to be one thing. Yet, they are in fact
different, and are likely to attract some employees who have needs that do not ”fit"

(W anous, 1980). This conclusion is supported by research done by Porter and Steers
(197 3) that revealed that people who took jobs that did not fit their vocational interests

were more likely to quit. Due to this dynamic, Schneider (1983; 1987) argues that

35

people who remain will be similar to each other as they will constitute a more
homogenous group than those who were initially attracted to the setting. Thus, people
with similar abilities and needs tend to be attracted to particular settings, and people
with similar sets of positive reactions to their experiences tend to remain in those
settings.

Schneider (1983) further states that attrition will produce a restriction in the
range of individual differences. This is a result of the process of career organizational
attraction and attrition that tends to narrow the range of types of people in any one
organization. He further says that, if people with the full range of needs and abilities
are not attracted to each occupation, career, or organization, then there will be a
restriction in the range of abilities and needs represented. Because of attraction to
organizations and attrition from them, similar people are there, and they behave
similarly because they are similar and not because of external factors. This restriction
in range yields people who are not only similar in kind, but who will be similar in
behaviors, experiences, orientations, feelings, and reactions (Alderfer, 1971).
Interactionist thinking leads to the conclusion that this restriction in the range of people
in particular organizations yields organizations, occupations, and careers that are
characterized by the kind of people who are attracted to them and, selected by them,
will remain in them. That is, over time, interpersonal interaction would result in
people and situation becoming integrated.

Schneider (1987) further says that, through recruitment and formal and informal

selection procedures, organizations actually end up choosing people who share many

36

common personal attributes, although they may not share common competencies.
When current members screen potential new members, they too are attracted to similar
others, so they are more likely to admit new members like themselves. After entering
an organization, a new member and the more tenured members become better
acquainted and the similarity-attraction dynamic can again effect the feelings of both
parties. Because the similarity effect operates on attraction, selection, and attrition
processes, organizations evolve to have unique compositions of organizational members
who are relatively similar to each other. The arrangement is likely to be judged
satisfactory to the extent that perceived similarity is maintained (Tsui & O'Reilly,
1989). If a match is judged unsatisfactory, pressures form to encourage dissimilar
members to exit the organization. Over time, these processes create psychologically
homogenous work groups (George, 1990).

The social psychological literature (Byme, 1969; Schneider, 1983; 1987)
suggests that similarity is an important basis of personal attraction, and that
interpersonal attraction is related to integration and, consequently, to turnover (Price,
1977). Kanter (1977) and South et al. (1982) have both argued that it is not one's sex
or race, per se, but the proportion or composition of the organizational unit in terms of
the particular salient ascriptive characteristic that affects the group dynamics. This, in
turn, affects the attitudes and performance of the individuals involved. What is
important, therefore, is the person's attributes in relation to the attributes of others in

the unit in which she or he works. Therefore, similarity is not an attribute of an

37

individual, but rather a relationship between the individual and others in the
organization.

There is ample support across diverse populations on the strong association
between similarity and interpersonal attraction. Werner and Parmlee (1979) showed
that similarity in pastime preferences can predict friendship and attraction. Other
research has shown that dissimilarity can lead to repulsion as the differences between
people increases the distance between them and lowers interpersonal attraction and
liking (Rosenbaum, 1986). In other words, people tend to be drawn to those who are
similar to them in terms of demographic characteristics, activities, or attitudes (Byrne et
a1. 1986).

In addition to interpersonal attraction, similarity has also been found to have a
positive effect on communication and integration in social groups. For example,
Lincoln and Miller (1979) analde the effects of gender, race, and education on work
and friendship ties. They found that greater levels of demographic similarity affected
frequency of communication. Race and gender were positively related to a number of
friendship ties, and education was associated with increased work contacts. More
recently, Zenger and Lawrence (1989) found age and tenure similarity to be positively
associated with the frequency of technical communication in project teams. O'Reilly et
al. (1989) found that increased distance in tenure within work groups lowered social
integration which was, in turn, associated with higher turnover.

Katz and Kahn (1978) argue that social integration can be best thought of as a

multifaceted phenomenon that reflects attraction to the group, satisfaction with other

38

members of the group, and social interaction among the group members. Relative
similarity of group members is likely to increase social integration (Newcomb, 1961).
Similarity in attitudes has been shown to promote group cohesion (Good & Nelson,
1971). There is some support for the association between social integration and
similarity in demographic attributes, such as race, age, and education (Hoffman, 1985;
Ward et al. 1985; Tsui and O'Reilly, 1989). Piper et al. (1983) reported cohesion to be
associated with higher commitment to remaining in the group. Other studies have
suggested that social integration is related to such variables as satisfaction with
coworkers and the degree of inclusion in group communication networks (Roberts and
O'Reilly, 1979; Mobley, 1982). Thus, it appears that demography in organizations can
affect work perceptions and attitudes through both interpersonal attraction and the
frequency of interactions. Recognizing the importance of demography, researchers
have been active over the past ten years documenting the conceptual (Pfeffer, 1983) and
empirical effect of demographic composition or distribution on a variety of
organizational outcomes, such as turnover (McCain et al. 1983) and communication
(Zenger & Lawrence, 1989).
WWW

Wagner et al. (1984) contend that studies emphasizing individuals in isolation
have neglected the effect of relationships among individuals, while other studies of
organizational behavior have tended to neglect individual-level attributes. They further
suggest that organizational demography (Pfeffer, 1983) affords a way of linking

individual and organizational-level attributes. The demography of an organization (or a

39

subunit) reflects the aggregation of characteristics of individual members of its
population. But, at the same time, it is a macro-level property of the unit as a whole
and, as such, incorporates the relationship and distributional properties of the
individuals who are members. The similarity effects provides a rationale for how and
why demographic compositions of organizations are likely to be related to
organizational phenomena.

Pfeffer (1983) uses the term organizaflgnaljgmggmphy to refer to the
demographic composition of formal organizations. Demography refers to composition
in terms of basic attributes, such as age, sex, educational level, length of service, race,
etc. , of the social entity or organization under study. Organizations can be described in
terms of their sex composition, their racial composition, their age composition, the
length of tenure distributions, the educational levels of their workforces, and the socio-
economic origins of their members. Demography of an organization is therefore the
composite aggregation of characteristics of the individual members of that entity. For
instance, the age composition of a organization is the result of the specific ages of its
individual members. However, there are compositional or demographic effects,
resulting from the specific demographic distribution, that are more than the sum of
effects of the individual-level variates.

An example of this effect is provided by Kanter (1977) in her research on men
and women in corporations. Kanter argued that proportion (specifically, the proportion
of men and women in organizations) has important effects on group processes and,

consequently, on what happens to those in minority status. Being in a "token position"

4O

(one's group constitutes less then 20% of the social unit) exposes those in token status
to increased visibility, scrutiny, and pressure which, Kanter argues, can lead to reduced
performance and increased psychological stress. She further contends that there are not
significant sex effects on performance and stress at the individual level of analysis.
Rather, it is the relative proportion of men and women that condition the form and
nature of social interaction and group process. This composition impacts psychological
well being, attitudes, and job performance. Based on such reasoning, Pfeffer contends
that demographic distributions have a theoretical and empirical reality of their own that
are distinct from the aggregation of responses of the individual members within an
organization. Consequently, he makes the case for the demography of formal
organizations as an important explanatory variable in organizational analysis. Pfeffer
(1983) has argued that most organizational theories are premised on a number of
hypothetical constructs that are not directly observable or measurable. He further goes
on to say that underlying process variables are "neither concrete nor unambiguous in
their meaning and interpretation" and often violate "rules of parsimony, producing two-
, three-, and even four-way interactions as explanation of behavior. " Further, the
amount of variance explained by process measures is generally quite small, and Pfeffer
questions whether intervening process variables account for any incremental variation
in dependent variables beyond that already explained by demographic variables.
Furthermore, organizational demography is related to (among other factors) the
rate of growth of the organization and the industry or sector in which it operates, the

technology employed, the employment practices relating to compensation, layoff

41

policies, and factors that affect voluntary and involuntary turnover, and unionization.
Organizational demography can have effect on the frequency and type (insider versus
outsider) of administrator succession; performance, innovation, and adaptability; the
form of control employed and the size of the administrative component; the amount
and form of interorganizational linkages, as well as transaction patterns across
organizations; cohort identity and intercohort conflict that often resulting both in
turnover and the choice of personnel; the distribution of power across cohorts; and
mobility aspirations and expectations resulting from different career processes (Pfeffer,
1983, pg. 229).

In this research, Pfeffer mainly examined tenure and the length of service
distribution of the workforce in organizations to explain behavioral patterns, including
communication, job transfers, promotions, and turnover. But, he also considered
characteristics such as race, age, tenure, religion, gender, and socio—economic
background as important dimensions that explain differences in peoples' attitudes,
values, and behaviors. According to this perspective, the distributional properties of
the demography of an organization are critical--not simply the proportion of the
membership with a given attribute or the potential compositional effects of the
variables. Pfeffer (1985) further elaborates on this theme by proposing that
homogeneity in age and date of entry may increase attraction through similarity and
increased likelihood of communication. Thus, homogeneity or similarity is likely to
enhance social integration and, in turn, lead to a lower likelihood of leaving.

Individuals, who enter a group at the same time or who are of similar age, may become

42

more tightly bound to one another than individuals who are demographically
heterogenous.

Using this perspective to examine turnover, Wagner et al. (1984) speculate that
it is the relative integration within groups that is critical. Arguing that a manager's
attitudes and perspectives are affected by their times of entry, the authors suggest that
demography can affect the relative cohesiveness of groups which may, in turn, enhance
or diminish the probability of turnover. People who leave are likely to be those who
are most different in terms of age and date of entry. They undergo similar experiences,
see things from a common perspective, and have a greater chance to interact (O'Reilly
et al. 1989; Tsui & O'Reilly, 1989). Demographic characteristics, therefore, are clear
examples where both direct attributes, such as an individual's gender or age and
relationship on an attribute between two or more individuals, are important for
understanding social interactions and outcomes (Tsui et al. 1991).

One of the criticisms of the research in organizational demography has been that
researchers have not typically investigated how variations in multiple demographic
variables--in either their simple or relational forms--affect outcomes like attitudes or
work performance. Most demography studies have tended to focus on age or date of
entry as the primary demographic variable affecting turnover. Individuals vary on
multiple demographic characteristics. Analyses of demographic effects have to
consider the full impact of an individual's demographic profile rather than focus on
only one or two demographic characteristics (Tsui & O'Reilly, 1989). Recently,

Lawrence (1994) and Smith et al. (1994) have been critical of organizational

43

demography researchers because they believe they have tended to ignore underlying
processes as an explanation for organizational outcomes. Lawrence (1994) says that,
due to this reason, subjective concepts have become the "black box" of organizational
demography.

Research by Triandis et al. (1965) and Ruhe and Eatrnan (1977) show that there
is a heightened incidence of interpersonal problems associated with heterogenous
groups, including multicultural ones. Interethnic tensions can interfere with members'
interactions (V aid-Raizada, 1985), and communication can be discouraged when
exchange predictability and attraction to others are low (Triandis, 1960). Cultures
themselves affect members' levels of participation and openness (Tang & Kirkbride,
1986). Because free and ongoing communication is an important ingredient of creative
problem solving (Ebadi & Utterback, 1984), overcoming the communication difficulties
of multicultural workgroups will be key to realizing the creative potential of such
groups (Shaw, 1981; Triandis et al. 1965). Individual dissimilarity and heterogeneity
may create equal levels of discomfort for all group members. It may limit the
integration and the development of cohesiveness at the level of the group as a whole,
and thereby influence work-related attitudes and behaviors (Jackson et al. 1991).

INTERPRETATION OF THE THEORIES

Tsui et al. (1991) contend that any individual can be different from, or similar
to, any other individual in a social unit on the demographic attribute being considered.
Thus, "being different" is a relational concept that applies to everyone-the majority as

well as the minority. Demographic attributes, such as gender, race, age, occupation,

44

education, status, authority level, represent information that individuals might use to
infer one's similarity to others on such things as attitudes or beliefs (Byrne, 1971). The
presumed similarity in attitudes or beliefs then influences the individual's attraction
toward other individual(s). Consequences of low attraction include less
communication, low social integration, and eventual turnover (T sui et al. 1991).

Both the organizational demography and the ASA framework indicate that
dissimilarity and heterogeneity are negatively related to organizational effectiveness. In
fact, research indicates that demographic attributes are associated with differences in
attitude, values, beliefs that have the potential to create conflict among team members,
and can influence group outcomes and behaviors (Daft & Weick, 1983; Pfeffer, 1983).
Studies also indicate that dissimilar members also face difficulties in integrating into a
group. They are often made to feel uncomfortable in the group, are perceived as poor
performers, and are pressurized to leave (O'Reilly et al. 1989; Schneider, 1987; Tsui &
O'Reilly, 1989; Wagner et a1. 1984). Individuals in an age-heterogenous group have
also been shown to have higher turnover rates as are individuals distant in age from an
otherwise homogenous group (McCain et al. 1983; Wagner et al. 1984; and Pfeffer and
O'Reilly, 1984).

One of the potential drawbacks of similarity and homogeneity is that
organizations, over time, can become so ingrown in type that they begin to occupy an
increasingly narrow ecological niche (Aldrich, 1979). Schneider (1987) cautions that a
lack of diversity can cause organizations to fail--as its people, structures, and processes

may become so fixed to a particular segment of the environment that, when the

45

environment changes, the existing people, processes, and structures are no longer
viable. He argues that, unless organizations consciously fight restriction in the range of
people they contain, when the environment changes they may not be aware of change in
the first place and may not be capable of changing it.

As Argyris (1976) puts it, organizations are likely to experience "dry rot. " Dry
rot refers to the tendency of organizations to become increasingly unresponsive to
signals from the larger environment that change is necessary. Argyris notes that
organizations tend to attract and retain managerial people who are "right types, " that is,
people who have similar comprehensions, similar experiences, and similar reactions.
This very similarity yields stability, but also decay. Schneider (1983) cautions that, for
organizations to be viable, they need to have people who can comprehend the nature of
the relationship between their organization and the larger environment to carry out the
process of goal redefinition so essential to the continued viability of an organization.

He further states that structures and processes will change when the behaviors of people
change, and the behaviors of people will change when different kinds of people are
attracted to, are selected by, and stay in, an organization (Schneider, 1987).

In practice, organizations typically aim to homogenize their multicultural work
groups rather than attempt to use their inherent diversity (Brown, 1983; Fernandez,
1981; Jones, 1986). According to Brown (1983), cultural differences are ignored due
to "cultural homogenization," and valuable perspectives remain unexplored.
Conformity is encouraged, thereby depriving all sides of resources that are potentially

constructive. The key is to ensure that members are called upon to contribute

46
regardless of their cultural background (Copeland, 1988; Kirchmeyer & Cohen 1992),

without losing a core of similarity among group members. Cox & Blake (1991)
interpret this to mean that all members must share some common values and norms to
promote coherent actions on organizational goals. The need for heterogeneity to
promote problem solving and innovation must be balanced with the need for
organizational coherence and unity of action. Both the organizational demography and
ASA model suggest that, to some extent, personnel policies and practices are partial
determinants of the demographic distributions created in organizations.

Based on this assumption, Jackson et al. (1991) suggest the value of
investigating the multiple influences of personnel practices, group processes, and
individual psychology when attempting to understand behavior in organizations, given
the increasing demographic diversity of the American workforce (Johnston & Packer,
1987). Most of the organizational efforts to manage diversity at present are a
component of personnel policies and human resource efforts. Therefore, organizational
efforts to promote diversity as a process of planned change and effective utilization of
personnel might moderate and facilitate the interaction and integration of diverse
individuals and groups. This could then lead to better work-related attitudes and group
performance.

DIVERSITY AND WORK-RELATED ATTITUDES
Q . |° I C 'I |
March and Simon (1958) theorize that an employee makes two distinct, ongoing

decisions about an organization: the decision to participate, and the decision to produce

47

or perform. The considerations that lead to a decision to participate are based on the
notion of exchange between individual and organization. Employees balance the
inducements the organizations provides against the contributions required to maintain
membership. The considerations that lead to the decision to produce include the
strength of an employee's identification with the goals and values of the organization.
The most widely accepted and used definition of organizational commitment is based on
research done by Porter et al. (1974). They defined organizational commitment as: (1)
a strong belief and acceptance of the organization's goals and values; (2) a willingness
to exert considerable effort on behalf of the organization; and (3) a strong desire to
maintain membership in the organization.

There have been a number of models of organizational commitment based on
either the decision to participate or produce. Models of the decision to participate are
relevant for predicting decisions to join, attend, and remain a member of the
organization. Models of the decision to produce are relevant for predicting outcomes
such as performance, citizenship behaviors, and satisfaction with work. Definitions of
commitment have differed to the extent to which they emphasize the considerations
relevant to participation or production. The decision to participate is reflected in the
”desire to remain in the organization," and the decision to produce is well described by
a "willingness to exert considerable effort on behalf of the organization, " and "belief in
the acceptance of the goals and values of the organization" (Mayer &

Schoorman, 1992).

48

The literature has consistently reported a strong negative relationship between
organizational commitment and absenteeism and turnover (Bluedorn, 1982). Data show
that turnover and absenteeism are often higher for women and minorities as compared
to White males (Rhodes & Steers, 1990). Turnover among women in professional jobs
is double that of men (Schwartz, 1989; Thomas, 1990). When actual turnover is not a
feasible alternative, individuals may engage in other forms of withdrawal behavior,
such as lower levels organizational attachment (Tsui et al. 1991). Absence (Rhodes and
Steers, 1990), reduced commitment (Porter et al. 1972), and thoughts of leaving the
social unit may be considered as intervening mechanisms between demography
differences and actual turnover behavior. In general, age and company tenure have
been found to be positively related to commitment, while level of education has been
shown to have a negative effect (Morris and Steers, 1980; Morris and Sherman, 1981).
A recent meta-analysis (Cohen et al. 1988) reports a small negative correlation between
gender and organizational commitment. Other studies have found conflicting results.
Hrebiniak and Alutto (1972) showed that women were more committed than men, while
Aranya et al. (1986) reported the opposite finding. The study by Tsui et al. (1991)
demonstrates that the extent to which an individual is different from all other
individuals in the unit had a systematic effect on an individual's psychological and
behavioral attachment to the organization. The greater the differences on age, gender,

and race, the lower the individual's attachment to the organization.

49

Wu

Job satisfaction may be defined as a pleasurable or positive emotional state
resulting from the appraisal of one's job or job experiences (Locke, 1983). Job
satisfaction emphasizes the specific task environment where an employee performs his
or her duties. Daily events have been hypothesized as affecting the level of job
satisfaction on one's own experiences, reflecting more immediate reactions to specific
and tangible aspects of the work environment (Mowday et al. 1979). A meta-analyses
by Premack & Wanous (1985) demonstrated that the better the fit between individual
expectations and the reality of organizational life, the higher the job satisfaction and the
longer the tenure. Kanter's (197 7) research points out that women and minorities have
the least favorable experiences in the workplace. Their job satisfaction levels are often
lower. Kanter (197 9) argues that minority members, women, and other "token
employees" often have less access to opportunity and power within organizations. She
further views opportunities as growth prospects stemming from a present job, and
suggests that employees with restricted opportunities ultimately lower their aspirations
and commitment, and often engage in behaviors that reinforce negative opinions about
their potential contributions to an organization. There is evidence of discrimination
against women in compensation (Terborg and Ilgen, 1975), prospects for promotion
(Olsen and Becker, 1983), assignments to challenging tasks (T erborg and Ilgen, 1975),
access to authority and responsibility on the job (Harlan and Weiss, 1982), and
opportunities to cultivate developmental relationships with mentors, sponsors, and peers

(Fernandez, 1981). Miller and Wheeler (1992) found evidence that suggested work-

50

related factors--specifically job satisfacfion--contribute to women's intentions to leave
their jobs and actual turnover. They found women less satisfied than their male
counterparts with their opportunity for promotion, pay increases, pay relative to others
in the same organization (internal pay equity), and pay relative to people in similar
positions in other organizations (external pay equity). They are also significantly less
satisfied than men with the recognition they received for their accomplishments.

Ilgen and Youtz (1986) also suggest that minorities and women face
considerable treatment discrimination in organizations. Treatment discrimination not
only affects position assignments, training opportunities, salary increases, promotions,
terminations, and layoffs, but also relates to issues such as acceptance into a work
group and the availability of career-enhancing and psychosocial support from others.
The authors further posit that treatment discrimination and unfavorable work
experiences faced by minorities can have dysfunctional consequences for their career
success due to reduced job performance, fewer opportunities to enhance work-related
skills, and limited chances to develop supportive relationships within an organization
with other employees that can result in diminished job performance. They further
propose that race differences in job performance can be explained-at least in part--by
the differential treatment people in different groups experience. Minority members in
an organization may internalize an organization' s negative evaluations of them and
engage in "self-limiting behaviors. " For example, they may refuse a challenging job
assignment or decline an opportunity for additional training that perpetuate performance

differences between minority and nonminority employees.

51
Both Ilgen and Youtz (1986) and Kanter (1979) posit that race influences job

performance evaluations through its effects on the organizational experiences of Black
and White managers. There is considerable evidence that raters evaluate the job
performance of Blacks less favorably than the job performance of Whites, especially
when the raters are themselves White (Kraiger and Ford, 1985). There is also evidence
that Black managers experience restricted advancement opportunities and report
extensive dissatisfaction and frustration with their careers. Jones (1986) reported that
only 15 percent of the Blacks in his sample described their organizational climates as
supportive for Black managers. Black managers are less likely than White managers to
feel they have been provided with important career-related information (Alderfer et al.
1980; Fernandez, 1981), and more likely to experience low levels of job discretion,
autonomy, power, and influence as a result of their status as out-group members in
their organization (Ilgen and Youtz, 1986; Fernandez, 1975; 1981). Nixon (1985)
found that 56 percent of the Black managers in her sample perceived themselves as
either partially or totally alienated from the formal or informal aspects of corporate life.
Fernandez (1981) also observed that many Blacks believe that minority managers are
likely to be excluded from informal work groups, further compounding their job
success. Greenhaus et al. (1990) found evidence that Blacks, compared to White
managers, felt less accepted in their organizations, perceived themselves as having less
discretion on their jobs, received lower ratings from their supervisors on their job
performance and promotability, were more likely to have reached career plateaus, and

experienced lower levels of career satisfaction. One of the major reasons women and

52

minorities experience considerable frustration over career growth is due to cultural
conflict with the dominant group (Cox & Nkomo, 1990; Morrison et al. 1987).
W

Job involvement is the psychological importance of work to the individual.
Lodahl and Kejner (1965) define job involvement as the degree to which a person's
work performance affects his or her self-esteem. It also refers to the identification with
one's work (J ans, 1985). It has been found to influence a number of important career
issues and outcomes such as organizational commitment (Blau, 1985), career
satisfaction (Lodahl and Kejner, 1965; Rabinowitz and Hall, 1977), expected or actual
turnover (Lee and Mowday, 1987), and desire for upward mobility (Hall et al. 1978).
Other studies have found positive relationships between high job involvement and work
outcomes, such as productivity, commitment, and time spent on work. Others have
found high job involvement to be related to negative outcomes, such as work conflict,
family conflict, and work-family conflict (V oydanoff, 1987).

Cox and Nkomo (1991) found very little empirical research on race-group
differences in job involvement and male-female comparisons on job involvement.
However, there are related findings in other streams of research. Employees who
withdraw from organizations have been found to be less involved with their jobs than
employees who remain (Mirvis and Lawler, 1977; Parasuraman, 1982). Data show
that turnover and absenteeism are often higher for women and minorities as compared
to White males (Rhodes and Steers, 1990). Turnover among women in professional

jobs was found to be twice that of men (Schwartz, 1989; Thomas, 1990). Chadwick-

53

Jones et al. (1982) found that, among blue collar workers, women felt their families
placed more pressure on them to stay at home rather than go to work. Thus, when a
child is ill, women are more likely to remain at home with a child. Child care
obligations may create a role conflict between work and family obligations and
contribute to absenteeism among women. Cheloha and Farr (1980) found a negative
relationship between job involvement and the number of days absent from the job. One
interpretation of this research is that women are likely to express less job involvement
due to work-family conflict than men and, subsequently, may be absent more
frequently. In his study on work-nonwork conflict Wiley (1987), found higher levels of
job involvement to be associated with higher levels of family-work conflict. Cox and
N komo (1991) found Blacks and women to have lower job involvement than White
males. They further observed, that in a society where family responsibilities are more
strongly associated with women than men, the family-work role conflict is likely to be
more severe for women. In many instances, women reduce their job involvement
levels as a conflict reduction strategy.

Runyon's (197 3) locus of control research found that externals to the group
show low degree of job involvement. Furthermore, Helms and Giorgis (1980) reported
data showing that Blacks, when compared to Whites, have a world view that is
consistently external in locus-of-control. Cox (1988) found that Blacks rated external
influences on promotion opportunities higher than Whites. These findings seem to
suggest that Blacks may have lower job involvement than Whites. Further, Rabinowitz

and Hall (1977) argue that self-esteem deriving from the fulfillment work-related

54

expectations is key to high levels of job involvement. Because the effects of negative
stereotypes and discrimination add an additional source of frustration to work-related
goals for women and minorities, it is reasonable to expect their self-esteem, and
ultimately their job involvement, will suffer.

Based on the above literature this research will attempt to answer these two

questions:

1. To what extent does. an WWW
I-ii‘ut 0 it... u o mW‘O.‘ .

attitudes.
2. To what extent does an WW

ul‘ |' "I O ‘00l9H0' 0‘ 0|

attitudes.

DIVERSITY PROGRAMS AS A MODERATOR

Schneider's (1987) ASA model and Pfeffer's (1983) organizational demography
model both assume that organizations are composed of individuals who are relatively
homogenous with respect to psychological attributes, such as attitudes, values,
personality, and demographic characteristics. A person's attitudes and psychological
attributes may be influenced by group socialization processes which, in turn, might
bolster within-group homogeneity. Schneider (1987) argues that, in organizational
settings, the tendency of people to be attracted to similar others and to feel
uncomfortable among dissimilar others leads them to seek membership in groups
composed of people perceived to be similar to themselves; select into their groups

people they perceive to be similar to themselves (the similarity-selection effect); and

discontinue their membership in groups whose members are uncomfortably dissimilar

55

to themselves (the dissimilarity-attrition effect). Based on this reasoning, dissimilar
individuals are most likely to feel less satisfied with their jobs, less likely to be
involved in their jobs, and less likely to be committed to the organization. The same
reasoning can be extended to groups where group heterogeneity can be the cause of
conflict and dissatisfaction and, in turn, affect work-related attitudes.

Jackson et al. (1991) view the perspectives of Schneider (1983, 1987) and
Pfeffer (1983, 1985) as complimentary, and suggest an integration that can be used to
improve understanding of groups in organizational settings. The research on individual
dissimilarity and group heterogeneity indicates that, with the exception of age, personal
attributes per se did not predict turnover from the group. They argue that this finding
provides strong support for the interactionist perspective, which highlights the role of
person-environment fit in determining behavior that takes into consideration the
importance of interpersonal contexts when explaining behavior or work-related
attitudes. For this reason, diversity programs and practices have the potential to
facilitate interaction and understanding among individuals and groups--thereby
overcoming some of the effects of individual dissimilarity and group heterogeneity.
Companies with liberalized benefits and work schedules reportedly experience lower
absenteeism and sick days (USA Today, December 2, 1987). Other studies have shown
companies, which provide child care assistance, not only experience less absenteeism
and turnover, but were able to improve worker morale, organizational commitment,

and job satisfaction (Y oungblood & Chambers-Cook, 1984). Therefore, it is likely that

56

diversity programs and practices as a moderator variable might provide a situational
incentive for maximizing the effectiveness of individuals and groups.
Based on the above literature this research will attempt to answer a third

question:

3. To what extent does the W
'l"lll . . ll-ll'l.

CHAPTER THREE
' RESEARCH METHODOLOGY
The purpose of this chapter is to introduce the organizations under study,
describe the sampling design, review the data collection process, present the research
hypotheses; provide a review of measurement issues, and operationalize the variables.
DIFFICULTY IN GAINING ACCESS TO A STUDY INDUSTRY
Before presenting an overview of the banking and financial institution industry,
it is necessary to discuss some of the difficulties associated with conducting diversity
research. Despite assertions made by managers regarding the importance of diversity
efforts within organizations and the need for research in this area, they are extremely
reluctant to provide researchers with access to their organizations for the purpose of
conducting diversity research. The sensitivity of diversity issues, the fear of possible
"political fallout, " and the real threat of being sued by employees if diversity data are
misused, has contributed to this reluctance. This researcher spent nearly two years
trying to get access to various organizations. Initially, a number of companies within
the food industry were contacted with the help of the Food Industry Institute on
campus. The food industry was selected because research would be important to the
College of Agriculture and Natural Resources, the researcher's home college. Through
the help of the Staff in the Food Industry Institute, a number of corporate diversity
managers were contacted. All the companies contacted were very interested in the
proposed research. However, none agreed to allow actual research to be conducted

within their organizations. When the food industry did not prove to have potential as a

57

58

possible research site, many months were spent in trying to get research access to other
industries such as the automotive, media, and service industries. Finally, a Mid-
Michigan area bank consented to provide access to conduct this research. This was
used as leverage to get access to the other two financial institutions. Perhaps, the trend
of "workforce feminization" within financial institutions, and the need to coordinate
work-family issues as a result of this feminization, could have been the reason why the
financial institution industry was agreeable to provide research access.

OVERVIEW OF THE BANKING AND FINANCIAL
INSTITUTION INDUSTRY

The banking industry (Standard Industrial Code 602) is a central part of the
U.S. economy and directly affects the lives of nearly all consumers and the operations
of almost every business. They are several types of banking institutions, such as
commercial banks, savings and loans, and credit unions to name a few. All these
institutions offer a wide variety of financial services, including making loans, accepting
deposits, clearing checks, and conducting more complex operations.

The employment patterns in this industry are dominated by women. Women
comprise 70 percent of the workforce in this industry, and occupy most of the lower
paying jobs. Due to these reasons, many banking institutions see the need for
instituting diversity programs that are responsive to the needs of a largely female
workforce. For example, a number of banks have developed flexible work schedules
for employees to balance work and family obligations. Because of these factors,
banking institutions are immensely suited for studying the effects of diversity programs

on work-related attitudes.

59

Despite the many changes in the economic and legislative climate for banking
institutions in the latter part of the 1980s due to federal legislation, deregulation,
mergers and acquisitions, commercial bank and S&L failures, the industry has
maintained its traditional occupational staffing patterns (Kuster, 1993). Annual
employment in the banking sector has increased considerably. Commercial banks
employed 1,552.7 thousand employees in 1989, an increase of 9.9 percent from 1982.
Federal savings institutions employed 244.5 thousand employees in 1989. This
represented a growth of 61.9 percent employment from 1982. Credit unions employed
130.1 thousand employees in 1989, an increase of 5.8 percent from 1988 (BLS,
Employment, Hours, and Earnings, 1991). In 1989, women constituted 68.8 percent of
the total employment. 77.03 percent of the employees were White, whereas 22.9
percent were minorities (EEOC Reports, 1989).

Many banking institutions responded to the deregulated environment of the
19803 by speeding up the implementation of new technologies and processes to
streamline operations. Of particular importance is the growing use of automation and
computer technology to achieve productivity gains. New machines and technologies
permitted greater automation of many positions and operations, such as bookkeeping.
As computer terminals were moved to the teller window, bank transactions could be
entered into the computer directly, increasing speed and efficiency of the operations.
Computerization lessened the need for a more educated and trained workforce. The
proliferation of automated teller machines (ATMs) also reduced the need for labor, and

thus decreased operation costs. Customer services, such as deposits and withdrawal of

60

cash and checks, form a large part of the day-to—day functioning of the banks and can
be easily handled by tellers due to computerization.

Tellers form the largest occupation in this industry. Employment of tellers has
also increased due to the increased use of part-time, peak hour, or prime-time tellers
(Kuster, 1993). This occupational category also receives low wages as compared to
comparable clerical work elsewhere. Clerical and administrative support occupations
as a whole also form the largest occupational group in the banking industry.

FOCUS OF THE STUDY

Although organizational characteristics and other details of each participating
institution will be described, the main focus of the study will be limited to the
individual perceptions of diversity programs and the effect on work-related attitudes.
The research focus will be on imh'mluals who make up the organizations, their
perceptions regarding diversity programs, and their relationship to work-related
attitudes. In this study, three banking institutions will be studied: two commercial
banks and a federal credit union. All institutions operate in the Mid-Michigan area.
The credit union and one of the commercial bank are similar in size with 150
employees each. The other commercial bank is larger with approximately 300
employees. The majority of the employees work at the main branch in all the three
institutions. All institutions have small branch offices at a few locations. The majority
of employees in all institutions are non-unionized, white collar-clerical workers. There

is also a management team in each of the financial institutions that directs operations.

61

STUDY SAMPLE

A total of 550 employees were surveyed in the three institutions. Among these,
214 individuals provided usable survey responses. The sample was predominantly
female and Caucasian. The following contains a discussion of sample characteristics.
214 employees who provided completed surveys

The total sample includes 156 females (78.8%) and 42 (21.2%) males. The
mean age for the sample was 27.8 years. Twelve employees (6.1%) identified
themselves as African-American. One hundred seventy-one employees (86.8%)
identified themselves as Caucasian. Nine (4.6%) employees identified themselves as
Hispanic, and four (2%) as Asian. Approximately eleven percent of the total sample
completed high school. Forty-one percent had some college education. Thirty-three
percent of the employees had completed a college program, and thirteen percent had
been to graduate school. About sixty-four percent of the sample were married at the
time of the survey, and about forty-seven percent had children living at home. Ninety-
three (47.9%) employees identified themselves as the primary wage earners in their
families, and 100 (51.5%) were not the primary wage earners in the family. Less than
twenty percent of the sample consisted of part-time employees with 160 (81.6%)
identifying themselves as full-time employees. The average length of service in the
three institutions was 7 .2 years. Table 1 shows the demographic profile for this

sample.

62

Table 1

Demographic Profile of Final Sample

 

 

 

 

 

 

 

 

 

Variable Total Percent II
Gander 198
Female 156 78.8
Male 42 21 .2
W 27:8
Standard Deviation 1.9
Race
African American 12 6.1
Caucasian 171 86.8
Hispanic 9 4.6
Asian 4 2.0
E I |° [I . I | l l
mmnletedl
High School 23 11 .6
Some College 82 41.4
College Degree 66 33.3
Graduate School 26 13.1
[1 . I' I I
Average Years 7.2
Standard Deviation 5.0

 

 

METHOD OF DATA COLLECTION
This section describes the data collection procedure used in this study. The
methodology encompasses three phases:
1. Identifying the sample;

2. Conducting a pilot study; and
3. Collecting quantitative data;

63

El [-11 IT. II S l

A survey of each of the bank's central offices was taken. Interviews with key
people were held to find out about the various diversity programs in place and the
perceived effects of these programs. The interviews were used to develop a profile of
each of the organizations under study. Archival records, such as organizational charts,
job descriptions, materials on diversity programs, and about various functional
departments, were also examined. These meetings were utilized to discuss the
proposed survey questionnaire and method of data collection for the study. These
interviews were utilized to identify a sample for the pilot study. For the pilot study,
branch offices of each of the banks were chosen to test the instrument.
W

The pilot study was used to strengthen the W of the survey
instrument (Appendix A) to be used in the quantitative phase of the study. Specifically,
the data collection effort involved the collection of demographic and attitudinal factors
using a structured questionnaire. W is defined as representing the
correspondence between a construct and the operational procedure to measure or
manipulate that construct (Schwab, 1980). It is based on the way a measure relates to
other variables within a system of theoretical relationships. On the other hand, content
nudity refers to the degree to which a measure covers the range of meanings included
within the concept (Babbie, 1989). To strengthen both the construct and content
validity of the instrument, the participants in the pilot study were asked to fill out the

questionnaire. After they had filled out the questionnaire, they were asked how

64

strongly or comfortable they felt about each operationalized statement, and whether the
statement had been placed correctly. The scale used to measure the agreement level
ranged from 1 to 3, where 3 = no doubt, 2 = agree, 1 = doubtful. Statements were
ranked according to mean agreement ratings so that the best items could be determined.

The instrument's [eliahility or its internal consistency was evaluated in three
ways. First, an item-level analysis of means and standard deviations were calculated.
Items with high or low means and low standard deviations were eliminated. Second,
item-level and scale-level intercorrelations were examined. Third, scale-level reliability
statistics, i.e. , Cronbach's alpha internal consistency reliability coefficient, were run to
set a level for alpha after examining the data.

Participants in the pilot study were also asked to participate in focus groups
where they were asked about their perceptions and effects of diversity programs that
their banks had instituted. They were also asked whether these programs facilitated
individual and group interaction thereby affecting work-related attitudes and
performance. Feedback from these focus group discussions was used to refine the
survey instrument.

[I III-Ell I' 0 I'll' DI

The objective of this phase was to collect more specific information to test the
hypotheses. The data collection effort involved collecting demographic and attitudinal
factors using a structured questionnaire. For the collection of data for this phase, the

following was undertaken:

65

Several group meetings were held between the researcher and groups of
employees (preferably departments) on site. As individual dissimilarity is being
measured in relationship to the individual ' s functional group or department and
group heterogeneity computed for each separate department, this process
ensured that each department received a separate set of surveys to facilitate the
data analysis process. Additionally, employees were also asked to self-report
the in home department as a means to increase the accuracy of the data
collected. This was an opportunity for the researcher to get acquainted with
various departments and establish credentials as a bona fide researcher.

Before handing out the survey, employees were told that the purpose of the
research was to help the researcher complete her doctoral research on the
relationship between diversity programs and work-related attitudes.
Confidentiality of responses was promised. The participants were told that only
the researcher would have access to each person's completed survey. To
minimize potential social desirability bias, it was emphasized that participation
in this study was mluntanz.

During the survey administration, individuals were asked to complete the
surveys but that they were not required to identify themselves. At the end of the
survey administration, they handed in a completed, anonymous survey. They
were told that a box would be placed in a central location for them to return the
completed survey or they could mail-in their surveys. Stamped envelopes were

provided to employees with each survey.

66

4. The researcher also explained how to complete the questionnaire. For each
section, she read the instructions to increase the chance that participants
understood the purpose of each section. She was also available to answer any
questions or respond to comments participants had while they looked at the
survey.

RESEARCH MODEL
Conceptually, the research model in this study is based upon Pfeffer's (1983)
organizational demography theory. Demography refers to the composition of an
organization in terms of basic attributes, such as age, sex, educational level, length of
service, race, and other characteristics of an organization under study. Pfeffer contends
that organizational demography has the benefit of being objectively measurable, and has

the potential to account for a wide range of organizational behaviors and attitudes in a

parsimonious and straight-forward fashion. However, there are compositional or

demographic effects, which result from the specific demographic distribution, that are
more then the sum of the effects of the individual level variates. Organizational
demography affords a way of linking individual level and organizational level
attributes. In this sense, the demography of an organization or subunit reflects the
aggregate characteristics of individual members of its population. But, at the same
time, it is also a macro-level property of the unit as a whole, incorporating the
relationship and distributional properties of the individuals who are members (Wagner

et al. 1984).

67

Kanter (1977), in her study on men and women in corporations, provides an
example of demographic effects. She has argued that proportions of men and women in
organizations can have important effects on group processes and consequently on what
happens to those in minority status. Pfeffer (1983) points out there are not significant
sex effects on performance or stress at the individual level of analysis. Rather, it is the
relative properties of men and women which condition the form and nature of social
interaction and group processes that occurs with subsequent impacts on psychological
well-being, attitudes, and even job performance. Pfeffer further says that it is in this
sense that demographic distributions have a theoretical and empirical reality distinct
from the aggregation of responses of the individual members that might be predicted on
the basis of some demographic attribute at the individual level of analysis.

Wagner et al. ( 1984) contend that examining the effect of education, age, or
length of service or other attributes on an individual's turnover or other outcome misses
the possibility that what may be critical is not an individual's characteristic in isolation
but, rather, the relationship of his or her attributes to others in the organization. For
instance, the social psychological literature emphasizes that similarity is an important
basis of interpersonal attraction (Byrne, 1969) and interpersonal attraction is related to
selection and attrition (Schneider, 1987) and, consequently, to turnover (Price, 1977).
But similarity is not an attribute of an individual; it is a relationship between the
individual and others in the organization (Wagner et al. 1984). Kanter (1977) and
South et al. (1982) have both argued that it is not one's gender or race, per se, but the

proportion or composition of the organizational unit in terms of the particular salient

68

ascriptive characteristic that affects group dynamics and the attitudes and performance
of the individual's involved. Therefore, what is important is the person's attributes in
relation to the attributes of others in the unit a person works (Wagner et al. 1984).

Pfeffer (1981) has argued that organizational demography researchers must use
a methodology that enables them to reflect the variations in demographic distribution
across organizations or organizational subunits. Based on this argument, Wagner et al.
(1984) developed measures for similarity/dissimilarity at both the individual and
subunit levels. For group level of analysis, they have suggested using Blau's (1977)
index of heterogeneity or the Gini Index to assess the extent to which there are a
number of significant categories in a distribution and how persons are dispersed over
such categories. They also defined a distance measure or measure of isolation for an
individual-level variable, such as birth date or date of entry into a metric, the purpose
of which is to capture membership in a cohort. This procedure is consistent with the
recommendations of Burt (1982), Lincoln and Miller (1979), and Blau and Alba (1982),
who have described the distance between any two individuals in terms of the Euclidian
distance between each of them and every other person in the population of interest.
Following this reasoning, then, individuals who are roughly the same from everybody
else are close to one another. In this way, grouping individuals according to
similarities allows cohorts to be identified.

A number of studies have tested hypotheses deriving from Pfeffer's demography
model of organizational behavior. All these studies have used organizational groups,

generally top management teams, as a target for their study. These studies have used

69

age and tenure, in particular, as a basis of defining predictor variables. The outcomes
studied have included turnover (McCain, O'Reilley, & Pfeffer, 1983; O'Reilley et al.
1989; Pfeffer & O'Reilley, 1987; Wagner et al. 1984; & Jackson et al. 1991);
communications (Zenger & Lawrence, 1989); innovation (Bantel & Jackson, 1989);
liking, role ambiguity, and performance ratings (Tsui & O'Reilley, 1989); and team
performance (Smith et al. 1994).

Recently, two studies done by Lawrence (1994) and Smith et al. (1994)
challenge Pfeffer's assumptions. Both studies suggest links between organizational
demography and organizational performance to unmeasured social psychological
concepts. Lawrence (1994) contends that organizational demography researchers have
accepted Pfeffer' s central argument without debate or discussion, and have ignored
subjective concepts that explain the significant association between demographic
predictors and outcomes, usually leaving these concepts unmeasured or untested. As a
result, subjective concepts and their relationships within research models have become
the "black box" of organizational demography. She argues, however, that Pfeffer's
instrumental approach, emphasizing prediction over explanation, poses serious
limitations and that process explanations may prove more appropriate. She therefore
emphasizes the need to go beyond the variance model proposed by Pfeffer.

The research of Smith et al. (1994) found evidence partially supporting process
as a mediator of the relationship between demography and performance, in which
demography and process variables each effect performance separately. They found

little support for the argument that demography, rather than process, affects

70

performance. This contradicts Pfeffer's (1983) contention that process measures can be
dismissed because they account for little of the variation in outcomes. Instead Smith et
al. (1994) found that team processes account for variability left unexplained by
demography. Their data also suggest that some forms of heterogeneity, such as
functional background heterogeneity, may have neither a direct nor indirect effects on
firm performance.

In their study, Jackson et al. (1991) found that age and tenure heterogeneity
were not the only demography variables found to be predictive of group turnover rates.
They suggested that attributes, other than age and tenure, can be successfully
incorporated into the research on organizational demography. Given the emerging
diversity of the American workforce, they suggest the inclusion of additional
background characteristics. According to them, this would well serve the demography
literature as it would produce a closer match between Pfeffer's theoretical model and
provide empirical tests for his model. Accordingly, they suggest that the particular
demographic attributes, which affect the feelings and behaviors of group members, are
likely to depend upon the distribution of large sets of demographic attributes. The
above arguments provide the background and impetus for this study.

RESEARCH QUESTIONS AND STUDY HYPOTHESES

The purpose of this dissertation is to investigate the effect of diversity
programs/practices on work-related attitudes--job satisfaction, job involvement, and
organizational commitment. This research proposes to test whether the relationship

among individual dissimilarity, group heterogeneity, and work-related attitudes are

71

likely to be moderated by an individual's perceptions of organizational efforts in

implementing diversity programs/practices on managing diversity. The research will

attempt to answer three major questions:

1. To what extent does an individual's perceptions of diversity programs/practices
affect the impact of individual similarity or dissimilarity on work-related
attitudes? ’

2. To what extent does an individual's perceptions of diversity programs/practices
affect the impact of group homogeneity or heterogeneity on work-related

attitudes?

3. To what extent does the experience of diversity programs/practices affect
individual and group interaction and work-related attitudes?

H ll ! 'II'IIIISI'E‘I'
Job satisfaction is defined as a pleasurable or positive emotional state resulting
from the appraisal of one's job or job experiences (Locke, 1983). A meta-analyses by
Premack & Wanous (1985) demonstrated that the better the fit between individual
expectations and the reality of organizational life, the higher the job satisfaction and
longer the tenure. Kanter's (1977) research points out that women and minorities have
the least favorable experiences in the workplace. Their job satisfaction levels are often
lower. Frustration over career growth and cultural conflict with the dominant group
are some of the major reasons (Cox & Nkomo, 1990; Morrison et al. 1987). There is
evidence of discrimination against women in compensation (Terborg and Ilgen, 1975),
access to authority and responsibility on the job (Harlan and Weiss, 1982), and
opportunities to cultivate developmental relationships with mentors, sponsors, and peers
(Fernandez, 1981). Greenhaus et al. (1990) found evidence that Blacks felt less

accepted in their organizations, perceived themselves as having less discretion on their

72

jobs, received lower ratings from their supervisors on their job performance and
promotability, were more likely to have reached career plateaus, and experienced lower
levels of career satisfaction.

Other researchers have indicated that demographic attributes are associated with
differences in attitude, values, beliefs that have potential to create conflict among team
members, and thus influence group outcomes and behaviors (Daft & Weick, 1983;
Pfeffer, 1983). Studies also indicate dissimilar members also face difficulties in
integrating into a group. They are often made to feel uncomfortable in the group, are
perceived as poor performers, and feel pressured to leave (O'Reilly et al. 1989;
Schneider, 1987; Tsui and O'Reilly, 1989; and Wagner et al. 1984). Individual
dissimilarity and heterogeneity may create equal levels of discomfort for all group
members. It may limit the integration and development of cohesiveness at the level of
the group as a whole, and thereby influence work-related attitudes and behaviors
(Jackson et al. 1991). Research by Jackson et al. (1991) and others support the view
that information about a person's demographic characteristics influences both
attributions regrading the person's psychological character and behavior towards the
person, and his/her membership within the group. Demographic attributes are also
likely to determine both perceptions of similarity/dissimilarity and perceptions of
person-environment fit. They also indicate the need for research regarding how
situational incentives might be used to inhibit dysfunctional similarity-dissimilarity
biases in group processes. As managing diversity is concerned with the meaningful

utilization of all employees, i.e. , women, minorities, and White males within the

73

organization, diversity programs should help to overcome or moderate similarity-
dissimilarity biases. Specifically, diversity programs and practices will correlate

positively with job satisfaction. Based on this discussion, the following hypotheses are

proposed:

W: Individual perceptions of diversity programs and practices will
predict job satisfaction. °

Hymthesisj: Individual dissimilarity will predict job satisfaction beyond
individual perceptions of diversity programs and practices.

W: The relationship between individual dissimilarity and job
satisfaction will be moderated by individual perceptions of
diversity programs and practices.

HypetliesisA: Group perceptions of diversity programs and practices will
predict job satisfaction.

W5: Group heterogeneity will predict job satisfaction beyond group
perceptions of diversity programs and practices.

Hymtlieeisj: The relationship between group heterogeneity and job satisfaction

will be moderated by group perceptions of diversity programs
and practices.

H ll ! 'II'IIIIII |
Job involvement is the psychological importance of work to an individual.
Specifically, job involvement is the degree to which a person's work performance
affects his/her self-esteem (Lodahl & Kejner, 1965). Employees who withdraw from
organizations have been found to be less involved with their jobs than employees who
remain (Mirvis & Lawler, 1977; Parasuraman, 1982). Data show that turnover and

absenteeism are often higher for women & minorities as compared to White males

74

(Rhodes & Steers, 1990). Turnover among women in professional jobs was found to
be twice that of men (Schwartz, 1989; Thomas, 1990). Chadwick-Jones et al. (1982)
found that blue collar women workers felt their families placed more pressure on them
to stay at home rather than go to work. Thus, when a child is ill women are more
likely to remain at home with a child. Child care obligations may create a role conflict
between work and family obligations and contribute to absenteeism among women. In
his study on work-nonwork conflict, Wiley (1987) found higher levels of job
involvement to be associated with higher levels of family-work conflict.

Cox and Nkomo (1991) found Blacks and women to have lower levels of job
involvement than White males. They further observe that, in a society where family
responsibilities are more strongly associated with women than men, the family-work
role conflict is likely to be more severe for women. In many instances, women reduce
their job involvement levels as a conflict-reduction strategy. Cox (1988) found that
Blacks rated external influences on promotion opportunities higher than did Whites.
These findings seem to suggest that Blacks may have a lower job involvement than
Whites. Further, Rabinowitz and Hall (1977) argue that self—esteem deriving from the
fulfillment of work-related expectations is key to high levels of job involvement.
Because the effects of negative stereotypes and discrimination add an additional source
of frustration to work-related goals for women and minorities, it is reasonable to expect
their self-esteem and ultimately their job involvement will suffer. As managing
diversity is concerned with promoting knowledge, accepting cultural differences among

different groups of employees, and valuing the participation and contributions made by

75

women, minorities, and dissimilar others, organizational efforts to implement diversity
programs should lead to a greater degree of job involvement for women, minorities,
and others. Specifically, diversity programs and practices will correlate positively

with job involvement. Based on this discussion, the following hypotheses are

proposed:

Hymthesisj: Individual perceptions of diversity programs and practices will
predict job involvement.

flymtliesisj: Individual dissimilarity will predict job involvement beyond
individual perceptions of diversity programs and practices.

W: The relationship between individual dissimilarity and job
involvement will be moderated by individual perceptions of
diversity programs and practices.

Hymtheeisjll: Group perceptions of diversity programs and practices will
predict job involvement.

W1: Group heterogeneity will predict job involvement beyond group
perceptions of diversity programs and practices.

W2: The relationship between group heterogeneity and job

involvement will be moderated by group perceptions of diversity
programs and practices.

 

Porter et al. (1974) define organizational commitment as: ( 1) a strong belief
and acceptance of the organization's goals and values; (2) a willingness to exert
considerable effort on behalf of the organization; (3) a strong desire to maintain
membership in the organization. The literature has consistently reported a strong

negative relationship between organizational commitment and absenteeism and turnover

76

(Bluedom, 1982). Data show that turnover and absenteeism are often higher for
women & minorities as compared to White males (Rhodes & Steers, 1990). Turnover
among women in professional jobs was double that of men (Schwartz, 1989; Thomas,
1990). A study by Tsui et al. (1991) demonstrated that individual differences in a unit
has a systematic effect on an individual's psychological and behavioral attachment to
the organization. The greater the differences on age, gender, and race, the lower the
individual's attachment to the organization. Individuals in an age-heterogenous group
have also been shown to have higher turnover rates, as do individuals distant in age
from an otherwise homogenous group (McCain et al. 1983; Wagner et al. 1984; and
Pfeffer & O'Reilly, 1984). There is evidence that companies with liberalized benefits
and work schedules reportedly experienced lower absenteeism and sick days (USA
Today, December 2, 1987). Other studies show that when companies provide child
care assistance, they not only experience less absenteeism and turnover, but also show
higher levels of worker morale, organizational comnritrnent, and job satisfaction
(Y oungblood & Chambers-Cook, 1984). As diversity programs are intended to
facilitate a better use of organizational resources by all employees, they should lead to
better organizational commitment. Specifically, diversity programs and practices will
correlate positively with organizational comnritrnent for individual dissimilarity and
group heterogeneity. Based on this discussion, the following hypotheses are suggested:
W: Individual perceptions of diversity programs and practices will
predict company commitment.

HypethesisJA: Individual dissimilarity will predict company commitment beyond
individual perceptions of diversity programs and practices.

77

W: The relationship between individual dissimilarity and company
commitment will be moderated by individual perceptions of
diversity programs and practices.

W6: Group perceptions of diversity programs and practices will
predict company commitment.

W: Group heterogeneity will predict company commitment beyond
group perceptions of diversity programs and practices.

Hymthesisjfi: The relationship between group heterogeneity and company

commitment will be moderated by group perceptions of diversity
programs and practices.

ANALYSIS OF DATA

A hierarchical regression or moderated regression method was used to test
whether the relationship between individual dissimilarity and group heterogeneity and
work-related attitudes (job satisfaction, job involvement, and organizational
commitment) would be moderated by an individual's perceptions of diversity
programs/practices. Hierarchical regression analysis permits assessment of whether
each variable significantly predicts the dependent variable with variance due to other
independent variables controlled (Cohen & Cohen, 1983). Hierarchical regression is
also called moderated regression (Stone & Hollenbeck, 1984). The term inedemter
ladable refers to an independent variable that potentially enters into interaction with
"predictor" variables, while having a negligible correlation with the criterion itself
(Cohen & Cohen, 1983; & Stone & Hollenbeck, 1984).

The order of entry of the independent variables becomes very important in
studying moderating effects (Stone & Hollenbeck; 1984). In this research, diversity

programs and practices was entered first in the regression equation. This is because

78

diversity programs and practices are only a small part of each organization's general
human resource management policies. There might be a number of other factors,
which can influence job-related attitudes, that might have led to individuals feeling
dissimilar or similar. To some extent, diversity programs have been instituted in
response to some employee dissatisfaction and felt organizational need. Therefore, to
test whether diversity programs and practices predict variance in work-related attitudes
and have a moderating effect on work-related attitudes, they were entered in the first
step. The regression involves three steps, and each step was used to predict variance in
work-related attitudes. Support for the hypotheses would be demonstrated by the
amount of variance explained by each step and by the cross-product.

In addition, the weak effects of diversity programs might pose potential
methodological problems due to the violation of WW.
Statistical conclusion validity addresses the problem of drawing potentially incorrect
conclusions about the relationship between two variables based on the assumptions for
probability distributions in statistical hypothesis testing (Cook & Campbell, 1979).
Correlation and regression models assume linearity, normality in the marginal
distributions, and homogeneity of variance in the conditional distributions of the
dependent variable (i.e. , homoscedasticity) around points of the independent variable(s)
(Hays, 1981). Because skewness and leptokurtosis can result from heteroscedasticity, a
nonnormal sample distribution indicates that the assumption of homoscedasticity is
violated when correlation and regression models are used in data analyses (Cohen &

Cohen, 1983).

79
H II RIIIIIIHIID"°I'I

Job satisfaction, job involvement, and organizational commitment were
separately regressed with individual perceptions of diversity programs and practices in
the first step to test whether individual perceptions of diversity programs and practices
would predict a significant amount of variance in work-related attitudes. The amount
of variance predicted by the regression would support the hypotheses.

HypetbeeisJ: Individual perceptions of diversity programs and practices will
predict job satisfaction.

W: Individual perceptions of diversity programs and practices will
predict job involvement.

Hymtbesm: Individual perceptions of diversity programs and practices will
predict company commitment.

Job satisfaction, job involvement, and organizational commitment were
separately regressed with individual dissimilarity in the second step to test whether
individual dissimilarity would predict a significant amount of variance in work-related
attitudes beyond that predicted by individual perceptions of diversity programs and

practices. The amount of variance predicted beyond that in the first step would support

the hypotheses.

W2: Individual dissimilarity will predict job satisfaction beyond
individual perceptions of diversity programs and practices.

Hmtbesiefi: Individual dissimilarity will predict job involvement beyond
individual perceptions of diversity programs and practices.

W14: Individual dissimilarity will predict company commitment beyond

individual perceptions of diversity programs and practices.

80

To test whether the cross-product of individual perceptions of diversity
programs and practices and individual dissimilarity would predict a significant amount
of variance in work-related attitudes above and beyond that predicted by the first two
hypotheses, individual perceptions of diversity programs and practices were regressed
by individual dissimilarity cross-product in the third step. This cross-product would
carry the interaction and provide support regarding whether diversity programs and
practices has a moderating effect on individual dissimilarity and work-related attitudes.
The amount of variance predicted above and beyond the first six hypotheses would
support these hypotheses.

W: The relationship between individual dissimilarity and job
satisfaction will be moderated by individual perceptions of
diversity programs and practices.

W: The relationship between individual dissimilarity and job
involvement will be moderated by individual perceptions of
diversity programs and practices.

W15: The relationship between individual dissimilarity and company

commitment will be moderated by individual perceptions of
diversity programs and practices.

WNW

Job satisfaction, job involvement, and organizational commitment were
separately regressed with group perceptions of diversity programs/and practices in the
first step to test whether group perceptions of diversity programs and practices would
predict a significant amount of variance in work-related attitudes. The amount of
variance predicted by the regression would support the hypotheses.

WM: Group perceptions of diversity programs and practices will
predict job satisfaction.

81

W9: Group perceptions of diversity programs and practices will
predict job involvement.

W: Group perceptions of diversity programs and practices will
predict company commitment.

Job satisfaction, job involvement, and organizational commitment were
separately regressed with group heterogeneity in the second step to test whether group
heterogeneity would predict a significant amount of variance in work-related attitudes
beyond that predicted by group perceptions of diversity programs and practices. The
amount of variance predicted beyond that in the first step would support the hypotheses.

W5: Group heterogeneity will predict job satisfaction beyond group
perceptions of diversity programs and practices.

Hymtbesile: Group heterogeneity will predict job involvement beyond group
perceptions of diversity programs and practices.

W: Group heterogeneity will predict company commitment beyond
that predicted by group perceptions of diversity programs and
practices.

To test whether the cross-product of group perceptions of diversity programs
and practices and group heterogeneity would predict a significant amount of variance in
work-related attitudes above and beyond that predicted by the first two hypotheses,
group perceptions of diversity programs and practices were regressed by group
heterogeneity cross-product in the third step. This cross-product would carry the
interaction and provide support regarding whether diversity programs and practices
have a moderating effect on group heterogeneity and work-related attitudes. The

amount of variance predicted above and beyond the first two steps would support these

hypotheses.

82

W: The relationship between group heterogeneity and job satisfaction
will be moderated by group perceptions of diversity programs
and practices.

HmtbesisJZ: The relationship between group heterogeneity and job
involvement will be moderated by group perceptions of diversity
programs and practices.

W: The relationship between group heterogeneity and company
commitment will be moderated by group perceptions of diversity
programs and practices.

MEASUREMENTDESIGN
D. °| B [B I. If . II

There have been no constructs suggested to measure diversity. There is also no
reported information on construct validity or specification regarding which paradigm is
being utilized for reliability issues. Researchers have applied different definitions to
diversity. There are many aspects to diversity programs and practices at the individual
(perceived, actual, desired) and the organizational level (perceived, actual, and
desired). Therefore, research must specify which aspects of diversity are being studied.
The data collection method that was used in this research is a questionnaire, which is
subject to problems associated with salience, attribution, and priming (Babbie, 1989).
There was also the issue of interaction between different levels of analysis. Diversity
programs and practices are naturally embedded within the context of the larger
organization. Subjective employee perceptions could be confounded with other job-
related factors or organizational experiences might not correspond with individual ones.

Diversity programs/practices as a variable were operationalized in. the following

manner. A list of diversity programs and practices currently in use by various

83

organizations across the U.S. were presented as part of the survey instrument. This list
consisted of eighteen diversity programs and practices in current usage in organizations
across the country and was adapted from the Society of Human Resource Management
Survey/Commerce Clearing House Inc. (SHRM/CCH) Survey (1993). These
programs/practices pertained to sexual harassment policy, physical access for
employees with disabilities, flexible work schedules, days off for religious holidays
beyond legal requirement, parental leave policy, job redesign for handicapper
employees, support for employees who are care-givers to elderly and ill relatives,
interpersonal skills training, subsidized day care benefits, cultural awareness training,
alternate career-pathing for employees who sacrifice advancement for more family
time, support groups and networking assistance to women, support groups and
networking assistance to minorities, support groups and networking assistance to single
parents, mentoring programs to women, mentoring programs to minorities, fast track
career programs for women, and fast track career programs for minorities.

Participants were asked about the knowledge of each program and the
effectiveness of each program. Knowledge was assessed with a single item that asked
whether a program existed. A four-point Likert-type scale, with scale points ranging
from. excellent to poor, was used to rate the effectiveness of each program that the
participant had expressed knowledge as existing in the organization. Asking about each
of these programs was important because it provided important feedback to each
individual organization regarding whether employees have knowledge about various

organizational efforts to manage diversity and the effectiveness of each of their

84

diversity programs. Based on the perceived knowledge and effectiveness of each
program, individuals were asked to rate their individual organization's overall efforts in
this area. Again, a four-point Likert-type scale ranging from excellent to poor was
used to rate the overall efforts in this area. In this study, the most widely reported
diversity programs/practices were sexual harassment policy, interpersonal skills
training, physical access for employees with disabilities, flexible work schedules, days
off for religious holidays beyond legal requirement, and parental leave policy. The
least reported were support and networking programs for women, fast track programs
for women and minorities, cultural sensitivity training, and programs supportive of
work-family issues such as day care assistance. The mean overall satisfaction with

diversity programs/practices was 2.7 on a four-point scale with a standard deviation of

.809.

 

According to Pfeffer (1983), demography needs to be assessed in ways that
enable the researcher to capture the distributional and compositional effects of
variations in group demography. Typically, such approaches attempt to capture the
relative homogeneity or the heterogeneity of the group or organization. Wagner et al.
(1984), O'Reilly et al. (1989), and Jackson et al. (1991) all used the Euclidean distance

measure to measure an individual's dissimilarity from the group for a given year. The

85

same method was used in this study. For each personal attribute, individual's

dissimilarity equals:

In this equation, ‘ n' is the number of group members, ‘s. ' is the individual's
value on the attribute, and g is the jth member's ‘ 5' value on the attribute. In other
words, this measures the square root of the mean squared distance for each individual
attribute, such as age, tenure, or gender etc. , of individual i from all other group
members. The squaring and square-root operations make this measure insensitive to
the direction of an individual's distance from the others in the group, without giving
disproportionate weight to greater distances (Burt, 1982; O'Reilly et al. 1989).
Appendix B shows the computation of a single dissimilarity variable.
Gmnnfletemgmin

Group heterogeneity was computed for each demographic characteristic
assessed. Two types of heterogeneity indices were computed. Allison (1978), in his
review of measures of inequality in social aggregates, observes that most measures of
dispersion can be converted into scale invariant measures of inequality by dividing by
the mean or some function of the mean. The W, for instance,
is just the standard deviation divided by the mean:

V=o

it .
This measure provides the most direct and scale invariant measure of dispersion.

He further noted that for variables, such as age, where utility is neither strictly

86

increasing nor especially relevant, the flat sensitivity of the coefficient of variation
makes it the appropriate choice. For W, the coefficient of variation
(standard deviation divided by the mean) will be computed. For WW,
Blau's (1977) indexJiLbetemgeneity will be computed. He suggests the index as one
of the methods to measure the integration of a social system. According to Blau, the
operational criterion of the degree of heterogeneity in a population is that two randomly
chosen persons do not belong to the same group. For any nominal parameter, the
larger the number of groups and the more evenly the population is divided among them,
the greater is the heterogeneity. Thus, there is more ethnic heterogeneity if there are
many ethnic groups than if there are few. For example, if nine-tenths of the population
belong to the same ethnic group and merely one-tenth to the others, ethnic
heterogeneity is less then if the population is more evenly distributed among ethnic
groups. This index, therefore, assessed the extent to which there are a number of
significant categories in a distribution and how persons are dispersed over such
categories. This measure, like the Gini index, is defined for aggregations as it seeks to
assess to some degree the extent to which there is dispersion within a distribution. The
index of heterogeneity varies from a low of 0 (if all group members are the same) to a

theoretical of high 1. Heterogeneity was defined in the following manner:
Heterogeneity = (1 - Epiz) .
In this equation, ‘ p' is the proportion of group members in a category and ‘i' is
the number of different categories represented in a group. If all persons are in one

group, there is no heterogeneity (1 - 1.00 =0); if all groups have the same size,

87

heterogeneity approximates unity with increasing numbers of groups. Appendix C
shows the computation of a group heterogeneity variable.
Qutcomelariables

Given the exploratory nature of the investigation and the wide range of variables
that might relate to the diversity-work attitude relationship, it was necessary to limit the
number and type of variables to the study. The study focused upon three work-related
correlates, namely, job satisfaction, job involvement, and organizational commitment.
These three variables were the eiiteemexariables for the study. When groups or
functional departments were the focus of analysis, gmmmeans were computed for
each work—related attitudes to test the effects of group heterogeneity.
Matisfaetign: Job satisfaction is defined as a pleasurable or positive emotional state
resulting from the appraisal of one's job or job experiences (Locke, 1983). Job
satisfaction was measured with the short form of the mm
W (Weiss et al. 1967). The MSQ contains items with two sub-
scales: intrinsic satisfaction (12 items) and extrinsic job satisfaction (8 items). The
MSQ contains a seven-point Likert scale that ranges from very dissatisfied to very
satisfied. The manual for MSQ reports median internal consistency reliability
coefficients of .86 for intrinsic satisfaction and .90 for extrinsic satisfaction. The
extrinsic and intrinsic sub-scales correlate .60 (Weiss et al. 1967). Thus a total score
can be used as a general satisfaction index. Based on completed surveys (N =212),
coefficient alpha was .92 for this study. The mean job satisfaction rate for this study

was 4.93 with a standard deviation of 1.02.

88

,leblnmlment: Job involvement is the psychological importance of work to an
individual. Specifically, job involvement is the degree to which a person's work
performance affects his/her self-esteem (Lodahl & Kejner, 1965). Job involvement was
measured with Lodahl & Kejner's (1965) 20-item job involvement scale. The lab
InmlxemenLSefle contained a seven-point Likert scale ranging from strongly disagree
to strongly agree. This instrument had demonstrated adequate psychometric properties
with internal consistency reliability estimates (i.e. , coefficient alpha) ranging from .72
to .89. Based on completed surveys (N =214), coefficient alpha was .86 for this study.
The mean job involvement for this study was 4.3 and standard deviation was 0.82.
thanizafienaLCennninnent: Porter et al. (1974) define organizational commitment
as: (1) a strong belief and acceptance of the organization's goals and values; (2) a
willingness to exert considerable effort on behalf of the organization; and (3) a strong
desire to maintain membership in the organization. Organizational commitment was
measured with Mowday et al. 's (1979) Organizationalflmmrtmentfluestinnnaire
(QQQ). The OCQ contains 15 items. Each question has a seven-point Likert scale
ranging from strongly disagree to strongly agree. The OCQ has demonstrated adequate
reliabilities with internal consistency reliability ranging from .82 to .93. Based on
completed surveys (N =210), coefficient alpha was .89 for this study. Table 2 shows
the means, standard deviations, and internal consistency reliability estimates for the

appropriate scales based on the final sample.

Internal Consistency Reliability Estimates

89
Table 2

 

 

 

 

 

 

 

 

 

Variable Mean Standard Deviation Coefficient Alpha
Job Satisfaction(J S) 4.93 1.02 0.92
Job 4.3 0.82 0.86
Involvement(JI)
Organizational 4.5 0.91 0.89 ll
Commitment (CC) i

 

J8, J1, CC through subjective norms based on a 7-point scale: 1= low on
measure; 7 = high on measure

Indixidnal Dissimn' ' anty' and Quinn Hetemgeneity' Variables: Individual

dissimilarity and group heterogeneity as variables were operationalized to include

demographic factors. Each functional department, such as the loan or mortgage

department, was be counted as one functional group for the purposes of computing

individual dissimilarity and group heterogeneity. All the variables involved in

computing individual dissimilarity and group heterogeneity were based upon each

employee's own self report. The variables used to compute individual dissimilarity and

group heterogeneity were employee age, gender, organizational and departmental

tenure, race, level of education, marital status, children living at home, primary wage

earner, part-time employment. The employees were also asked to report current

department and job titles to facilitate assignment into functional groups.

CHAPTER FOUR
RESULTS

In this chapter, the results of each hypothesis are presented. The chapter also
includes a discussion of the results of the hypotheses, as well as additional analysis of
the data with results.

INTER-CORRELATIONS OF STUDY VARIABLES

Before testing specific hypotheses, correlation analyses between each of the
independent variables and the dependent variables were done to test for significant
relationships. Table 3.1 and 3.2 take an initial look at the research data for significant
relationships between the independent and the dependent variables. The correlation
analyses was undertaken to confirm the order of entry of variables in the regression
analyses for hypotheses testing as discussed in Chapter Three of this study. The
analyses also helped to test whether any of the independent variables needed to be
controlled.

The results of the correlation analyses confirmed the order of entry of the
independent variables. The overall satisfaction with diversity programs/practices were
very significantly correlated with the three dependent variables, and they were entered
in the first step of the hierarchical regression analyses to test the study hypotheses.
Table 3.1 shows the results of the correlation analyses. It shows that overall
satisfaction with diversity programs/practices was significantly correlated with the
dependent variables. The table also shows the inter-correlations between each of the

individual dissimilarity (DI) variables and the dependent variables. It shows the

90

91

correlation for individual dissimilarity for age with Company Commitment (CC) was
statistically significant (p < .01). Individual dissimilarity for children living at home
was significantly correlated (p < .01) with Job Satisfaction (JS), Job Involvement (II),
and Company Commitment (CC).

Table 3 .2 shows the inter—correlations between each of the group heterogeneity
variables and the dependent variables. This table shows that the correlations between
each of the group heterogeneity variables and the dependent variables are not
significant. These analyses confirmed that it was not necessary to control for other

variables used in the study.

92

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.1
Correlations of Individual Dissimilarity Variables
with Dependent Variables
Dependent Variables
Individual Dissimilarity Variables (D1) .18 J1 CC
Satisfaction with Diversity Prtgrams (OA) .44" .30" .42"
Individual Dissimilarity (mg—Gender -.02 .03 -.09
DIZ—Age .10 .16 .17"
DI3-Race .03 .06 .16
Dli—Education .04 .11 .12
Dlg—Marital Status -.01 -.08 -.02
_D_I[-Children Livingit Home .19" .20"I .22"I
D17—Primary Wge Earner .01 -.08 -.02
Dla—Tenure in Company -.04 -.04 -.08
DI,-Part-time Employment .16 .08 .95
#Tenure in Department .16 .87 . 14

 

 

 

 

 

P < * .01, ** .001

Where: J S = Job Satisfaction; JI = Job Involvement;

CC = Company Commitment

 

93

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.2
Correlations of Group Heterogeneity Variables
with Dependent Variables
=
Dependent Variables I
Grou Hetero enei Variables G
P g 0' ( H) J S J1 C C

Group Heterogeneity (GH1)-Gender .04 .07 -.05
GHz—Age -.01 .04 .08
GH3-Race .11 .03 . 13
GHr-Education -.14 -.06 -.16
GIL-Marital Status -.05 -.09 -.01

, fl—Children Living at Home .15 .01 .13
GIL—Primary Wage Earner -.01 -.01 .01
GHl—Tenure in Company -.00 -.05 -.00
GH,-Part-time Employment .09 -.04 .05

. GHQ—Tenure in Department 1 .07 .07 . 10 ,

 

 

P < "' .01, " .001

Where: J S = Job Satisfaction; J] = Job Involvement;

CC = Company Commitment

BRIEF REVIEW OF THE METHODOLOGY
Hierarchical regression was used to analyze the data obtained in this study. The
dependent variables in the study were work-related attitudes, Job satisfaction (J S), Job
Involvement (J1), and Company Commitment (CC). The independent variables in the

study were Overall Satisfaction with Diversity Programs (0A); a range of Individual

94

Dissimilarity (DI) variables and Group Heterogeneity (GH) variables using ten
background variables (age, gender, level of education, race, organizational tenure,
departmental tenure, marital status, children living at home, primary wage earner, and
part-time employment); and a range of cross-products or interaction variables (CPDI &
CPGH) computed for the interaction between Individual Dissimilarity (D1) or Group
heterogeneity (GH) with Overall Satisfaction with Diversity Programs (OA) based on
each of the ten background variables.

Attribute dissimilarity for individual team members from the group was
calculated using the Euclidean distance measure. For each personal attribute, an

individual's dissimilarity equals:

In this equation, ‘n' is the number of group members, ‘s ' is the individual's
value on the attribute, and sJ is the jth member's ‘ 5' value on the attribute. In other
words, this measures the square root of the mean squared distance for each individual
attribute, such as age, tenure, or gender etc. , of individual i from all other group
members (Wagner et a1. 1984; Jackson et al. 1991). See Appendix B.

Group heterogeneity was computed for each demographic characteristic
assessed. Two types of heterogeneity indices were computed. For W,
the coefficient of variation (standard deviation divided by the mean) were computed.

For categoricauariahles. BlaU'S (1977) indrxnfhetemgeneity were computed. This

index, therefore, assessed the extent to which there are a number of significant

95

categories in a distribution and how persons are dispersed over such categories. The

index of heterogeneity varies from a low of 0 (if all group members are the same) to a

theoretical of high 1. Heterogeneity was defined in the following manner:
Heterogeneity = (l - Zpiz).

In this equation, ‘ p' is the proportion of group members in a category and ‘ i' is
the number of different categories represented in a group. If all persons are in one
group, there is no heterogeneity (1 - 1.00 =0). If all groups have the same size,
heterogeneity approximates unity with increasing numbers of groups (Jackson et al.
1991). See Appendix C.

Hierarchical regression was used to test the eighteen hypotheses. In the first
step, overall satisfaction with diversity programs (CA) was regressed with each of the
work-related attitudes. In the second step, a range of ten individual dissimilarity (DI)
or group heterogeneity (GH) variables were separately regressed with each of the work-
related attitudes. Finally, the cross-products were regressed with the outcome
variables. The model proposed that each step of the regression would predict a
significant amount of variability in work-related attitudes that is left unexplained by the
previous step. When all the eighteen study hypotheses were tested, it was found that
the overall regression model was highly significant.

RESULTS OF ANALYSES BY HYPOTHESIS
Hmthesm
Hypothesis 1 states that individual perceptions of diversity programs and

practices will predict job satisfaction. In other words, satisfaction with diversity

96

programs and practices (OA) will result in employees who are more satisfied with their
jobs. Table 4 includes the results of the regression analyses. This hypothesis was
tested in Step 1 of the regression analyses. In this step, CA was regressed with job
satisfaction. The results indicate that, as overall satisfaction with diversity programs
(OA) increases, individual attitudes towards Job Satisfaction (J S) improve. This
relationship was significant at the .000 level. Hypothesis 1 is therefore supported.
W

This hypothesis states that individual dissimilarity will predict job satisfaction
beyond individual perceptions of diversity programs and practices. In other words,
employees who are more dissimilar in their background demographic attributes are
likely to be less satisfied. This relationship will predict a more significant amount of
variability in job satisfaction beyond that explained by perceptions of diversity
programs and practices. The results of the regression analyses are reported in Table 4.
This hypothesis was tested in Step 2 of the regression analyses. In this step individual
dissimilarity (DI) was regressed with job satisfaction. When the ten individual
dissimilarity variables were collectively analyzed, individual dissimilarity (DI) had no
impact on job satisfaction. This relationship was not statistically significant.
Therefore, hypothesis 2 was not supported.
W

Hypothesis 3 states that the relationship between individual dissimilarity and job
satisfaction will be moderated by individual perceptions of diversity programs and

practices. In other words, satisfaction with diversity programs is likely to moderate the

97

effects of individual dissimilarity resulting in more satisfied employees. Table 4 shows
the result of the regression analyses. This hypothesis was tested in Step 3 of the
regression analyses. In this analysis, the cross-product variables (CPDI) were
regressed with job satisfaction. The cross—product or interaction term was not
statistically significant. This suggests that, when the ten (10) individual dissimilarity
measures were analyzed collectively, OA did not moderate the relationship between

individual dissimilarity and job satisfaction. There was no support for hypothesis 3.

Table 4

Hierarchical Regression Results for Job Satisfaction (J S), Predicted
from Overall Satisfaction W/Diversity Programs, Individual
Dissimilarity & Cross-Product

 

 

 

 

 

 

 

 

 

 

 

Variables R2 (J S) Ch. R2 (J S) F Sig’ .
SteiLl: 0A .190 .190 43.710 .000
Satisfaction with Diversity

'mgrms (OA)
$19.12: D11 ,0 m .256 .066 1.555 .124
Individual Dissimilarity
Variables (DI)
Sign}: CPDI1 ,0 m .279 .023 .537 .862
Cross-Products of OA * DI
Variables (CPDI)

Hawthesisfi

Hypothesis 4 states that group perceptions of diversity programs and practices

will predict job satisfaction. In other words, satisfaction with diversity programs and

98

practices will result in a group of employees that are more satisfied with their jobs.
The results of the regression analyses are reported in Table 5. This hypothesis was
tested in Step 1 of the regression analyses. In this step, 0A was regressed with job
satisfaction. The results indicate that, as overall satisfaction with diversity programs
(OA) increases, group attitudes towards Job Satisfaction (J S) improve. This
relationship was significant at the .000 level. Hypothesis 4 is therefore supported.
Hmthesiss

This hypothesis states that group heterogeneity will predict job satisfaction
beyond group perceptions of diversity programs and practices. In other words, groups
that are more dissimilar in their background demographic attributes are likely to be less
satisfied. This relationship will predict a more significant amount of variability in job
satisfaction beyond that explained by perceptions of diversity programs and practices.
Table 5 shows the results of the regression analyses. This hypothesis was tested in Step
2 of the regression analyses. In this step, group heterogeneity (GH) was regressed with
job satisfaction. When the ten group heterogeneity variables were collectively
analyzed, group heterogeneity (GH) had a moderate impact on job satisfaction. The
nature of the impact will be further examined through AN OVA analyses when
individual group heterogeneity variables are further examined. This relationship was
statistically significant at the .04 level. Thus, hypothesis 5 was supported moderately.
methesiu

Hypothesis 6 states that the relationship between group heterogeneity and job

satisfaction will be moderated by group perceptions of diversity programs and

99

practices. In other words, satisfaction with diversity programs is likely to moderate the
effects of group heterogeneity, thereby resulting in more satisfied employees. Table 5
shows the result of the regression analyses. This hypothesis was tested in Step 3 of the
regression analyses. In this analysis, the cross-product variables (CPDI) were
regressed with job satisfaction. The cross-product or interaction term was not
statistically significant. This suggests that, when the ten (10) group heterogeneity
measures were analyzed collectively, OA did not moderate the relationship between
group heterogeneity and job satisfaction. Therefore, there was no support for

hypothesis 6.
Table 5

Hierarchical Regression Results for Job Satisfaction (J S), Predicted
from Overall Satisfaction W/Diversity Programs, Group
Heterogeneity & Cross-Product

 

 

 

 

 

Variables R2 (J S) Ch. R2 (J S) F Srg' .
SteiLl: 0A .183 .183 40.680 .000
Satisfaction with Diversity

Programs (OA)

SteiLz: GHl ,0 m .266 .084 1.963 .040
Group Heterogeneity Variables

(GH)

$19.13: CPGHl ,0 m .320 .053 1.272 .250

Cross-Products of OA "' GH
Variables (CPGH)

 

 

 

 

 

 

 

 

100

Hmthesisl

Hypothesis 7 states that individual perceptions of diversity programs and
practices will predict job involvement. In other words, satisfaction with diversity
programs and practices will result in employees who are more involved in their jobs.
The results of the regression analyses are reported in Table 6 . This hypothesis was
tested in Step 1 of the regression analyses. In this step, 0A was regressed with job
involvement. The results indicate that, as overall satisfaction with diversity programs
(OA) increases, individual attitudes towards Job Involvement (II) will improve. This
relationship was significant at the .000 level. Hypothesis 7 is therefore supported.
W

This hypothesis states that individual dissimilarity will predict job involvement
beyond individual perceptions of diversity programs and practices. In other words,
employees who are more dissimilar in their background demographic attributes are
likely to be less involved with their jobs. This relationship will predict a more
significant amount of variability in job involvement beyond that explained by
perceptions of diversity programs and practices. Table 6 shows the results of the
regression analyses. This hypothesis was tested in Step 2 of the regression analyses. In
this step, individual dissimilarity (DI) was regressed with job involvement. When the
ten individual dissimilarity variables were analyzed collectively, individual dissimilarity
(DI) had a moderate impact on job involvement. This relationship was statistically

significant at the .07 level. The nature of this impact will be further examined through

101

AN OVA analyses when individual dissimilarity variables are further analyzed. Thus,
hypothesis 8 is supported moderately.
methesfl

Hypothesis 9 states that the relationship between individual dissimilarity and job
involvement will be moderated by individual perceptions of diversity programs and
practices. In other words, satisfaction with diversity programs is likely to moderate the
effects of individual dissimilarity, thereby resulting in more involved employees. . Table
6 shows the result of the regression analyses. This hypothesis was tested in Step 3 of
the regression analyses. In this analysis, the cross-product variables (CPDI) were
regressed with job involvement. The cross-product or interaction term was not
statistically significant indicating that, when the ten (10) individual dissimilarity
measures were analyzed collectively, OA did not moderate the relationship between
individual dissimilarity and job involvement. Thus, there was no support for

hypothesis 9.

102
Table 6

Hierarchical Regression Results for Job Involvement (JI), Predicted from Overall
Satisfaction W/Diversity Programs, Individual Dissimilarity & Cross-Product

 

Variables Rz (JI) Ch. R201) F
Step4: 0A .109 .109 22.801

 

i
Srg' .
000

Satisfaction with Diversity
Programs (OA)

m2: Dr, m, .191 .082 1.785 .066

 

Individual Dissimilarity
Variables (DI)

$199.3: CPDI, ,, ,, .225 .034 .727 .698

 

Cross-Products of OA * DI
Variables (CPDI)

 

 

 

 

 

 

 

W19

Hypothesis 10 states that group perceptions of diversity programs and practices
will predict job involvement. In other words, satisfaction with diversity programs and
practices will result in a group of employees who are more involved with their jobs.
Table 7 shows the results of the regression analyses. This hypothesis was tested in Step
1 of the regression analyses. In this step, 0A was regressed with job involvement.
The results indicate that, as overall satisfaction with diversity programs (OA) increases,
group attitudes towards Job Involvement (JI) improve. This relationship was significant

at the .000 level. Hypothesis 10 is therefore supported.

103

Hmthesisll

This hypothesis states that group heterogeneity will significantly predict job
involvement beyond group perceptions of diversity programs and practices. In other
words, groups that are more dissimilar in their background demographic attributes are
likely to be less involved. This relationship will predict a more significant amount of
variability in job involvement beyond that explained by perceptions of diversity
programs and practices. Table 7 shows the results of the regression analyses. This
hypotheses was tested in Step 2 of the regression analyses. In this step, group
heterogeneity (GH) was regressed with job involvement. When the ten group
heterogeneity variables were collectively analyzed, group heterogeneity (GH)) had no
impact on job involvement. The relationship was not statistically significant. Thus,
hypothesis 11 was not supported.
methesisJZ

Hypothesis 12 states that the relationship between group heterogeneity and job
involvement will be moderated by group perceptions of diversity programs and
practices. In other words, satisfaction with diversity programs is likely to moderate the
effects of group heterogeneity resulting in more involved employees. Table 7 shows
the result of the regression analyses. This hypothesis was tested in Step 3 of the
regression analyses. In this analysis, the cross-product variables (CPGH) were
regressed with job involvement. The cross-product or interaction term was not

statistically significant. This suggests that, when the ten (10) group heterogeneity

104

measures were analyzed collectively, OA did not moderate the relationship between

group heterogeneity and job involvement. There was no support for hypothesis 12.

Table 7

Hierarchical Regression Results for Job Involvement (JD, Predicted from Overall

Satisfaction W/Diversity Programs, Group

Heterogeneity & Cross-Product

 

 

 

Variables R2 (JI)

 

Ch. R2 (JI)

 

Steal: 0A .098

Satisfaction with Diversity
Programs (OA)

.098

19.694

Sig. ,

 

$129.2: GH1 ,0 10 .152

Group Heterogeneity Variables
(GH)

.054

1.099

.365

 

Stems: CPGH,,,,, .215

Cross-Products of OA "‘ GH
Variables (CPGH)

 

 

 

.063

 

Hypothesisn

 

1.294

.238

 

 

 

Hypothesis 13 stated that individual perceptions of diversity programs and

practices will predict company commitment. In other words, satisfaction with diversity

programs and practices will result in employees who are more committed to their jobs.

Table 8 shows the results of the regression analyses. This hypothesis was tested in Step

1 of the regression analyses. In this step, CA was regressed with company

commitment. The results indicate that, as overall satisfaction with diversity programs

105

(OA) increases, individual attitudes towards Company Commitment (CC) improve.
This relationship was significant at the .000 level. Hypothesis 13 is therefore
supported.
Hxnothesisld

This hypothesis states that individual dissimilarity will predict company
commitment beyond individual perceptions of diversity programs and practices. In
other words, employees who are more dissimilar in their background demographic
attributes are likely to be less committed to their jobs. This relationship will predict a
more significant amount of variability in company commitment beyond that explained
by perceptions of diversity programs and practices. Table 8 shows the results of the
regression analyses. This hypothesis was tested in Step 2 of the regression analyses. In
this step, individual dissimilarity (DI) was regressed with company commitment. When
the ten individual dissimilarity variables were collectively analyzed, individual
dissimilarity (DI) had a strong impact on company commitment. This relationship was
statistically significant at the .001 level. The nature of this impact will be further
examined through AN OVA analyses when individual dissimilarity variables are further
analyzed. Thus, hypothesis 14 is supported.
methesisls

Hypothesis 15 states that the relationship between individual dissimilarity and
company commitment will be moderated by individual perceptions of diversity
programs and practices. In other words, satisfaction with diversity programs is likely

to moderate the effects of individual dissimilarity, thereby resulting in more committed

106

employees. Table 8 shows the result of the regression analyses. This hypothesis was
tested in Step 3 of the regression analyses. In this analysis, the cross-product variables
(CPDI) were regressed with company commitment. The cross-product or interaction
term was not statistically significant. This suggests that, when the ten (10) individual
dissimilarity measures were analyzed collectively, OA did not moderate the relationship
between individual dissimilarity and company commitment. Thus, there was no

support for hypothesis 15.

Table 8

Hierarchical Regression Results for Company Commitment (CC), Predicted from
Overall Satisfaction W/Diversity Programs, Individual Dissimilarity & Cross-

 

 

 

 

 

 

 

 

 

 

 

Product

Variables R2 (CC) Ch. R2 (CC) F Srg' .
StepJ: 0A .184 .184 41.955 .000
Satisfaction with Diversity

Imam (OA)
81:12: D11 ,0 m .305 .121 3.077 .001
Individual Dissimilarity
Variables (DI)
Sign}: CPDIl ,0 m .327 .022 .540 .860
Cross-Products of OA * DI
Variables (CPDI)

Hmthesislfi

Hypothesis 16 states that group perceptions of diversity programs and practices

will predict company commitment. In other words, satisfaction with diversity

107

programs and practices will result in a group of employees who are more committed to
their jobs. Table 9 shows the results of the regression analyses. This hypothesis was
tested in Step 1 of the regression analyses. In this step, 0A was regressed with
company commitment. The results indicate that, as overall satisfaction with diversity
programs (OA) increases, group attitudes towards Company Commitment (CC)
improve. This relationship was significant at the .000 level. Hypothesis 16 is therefore
supported.
W

This hypothesis states that group heterogeneity will predict company
commitment beyond group perceptions of diversity programs and practices. In other
words, groups that are more dissimilar in their background demographic attributes are
likely to be less committed to their jobs. This relationship will predict a more
significant amount of variability in company commitment beyond that explained by
perceptions of diversity programs and practices. Table 9 shows the results of the
regression analyses. This hypothesis was tested in Step 2 of the regression analyses. In
this step, group heterogeneity (GH) was regressed with company commitment. When
the ten group heterogeneity variables were collectively analyzed, group heterogeneity
(GH)) had no impact on company commitment. The relationship was not statistically
significant. Thus, hypothesis 17 was not supported.
Hmthesislfl

Hypothesis 18 states that the relationship between group heterogeneity and

company commitment will be moderated by group perceptions of diversity programs

108

and practices. In other words, satisfaction with diversity programs is likely to
moderate the effects of group heterogeneity, thereby resulting in more committed
employees. Table 9 shows the result of the regression analyses. This hypothesis was
tested in Step 3 of the regression analyses. In this analysis, the cross-product variables
(CPGH) were regressed with company commitment. The cross-product or interaction
term was not statistically significant. This suggests that, when the ten (10) group
heterogeneity measures were analyzed collectively, OA did not moderate the
relationship between group heterogeneity and company Commitment. There was no

support for hypothesis 18.

Table 9

Hierarchical Regression Results for Company Commitment (CC), Predicted from
Overall Satisfaction W/Diversity Programs, Group
Heterogeneity & Cross-Product

 

 

 

 

 

Variables R2 (CC) Ch. R2 (CC) F Srg’ .
Steel: 0A .172 .172 37.815 .000
Satisfaction with Diversity

._Pr_ograms (OA)
Steel: GHl u, m .231 .059 1.309 .229
Group Heterogeneity
Variables (GH)
Step1: CPGHl ,0 1, .269 .038 .852 .580
Cross-Products of OA "' GH 1
Variables (CPGH) l

 

 

 

 

 

 

109

DISCUSSION OF THE STUDY FINDINGS

This study elaborated and tested hypotheses based upon Schneider's (1987) ASA
model, and Pfeffer's (1983) organizational demography model. As suggested in
Jackson et al. 's (1991) research, a series of hypotheses were developed about how
interpersonal contexts related to organizational behavior. Hypotheses were stated on
both the individual and group level of analysis. Similarly, it was assumed that
demographic attributes are powerful determinants of both perceptions of similarity or
dissimilarity and work-related attitudes that in turn could influence organizational
outcomes. This assumption is consistent with the organizational demography
perspective that asserts a person' s demographic characteristics influence both
attributions regarding the person's psychological character and behavior towards the
person (Jackson et al. 1991).

Previously, a number of studies have tested hypotheses deriving from Pfeffer's
demography model of organizational behavior. All these studies have used
organizational groups, generally top management teams, as a target for their study.
These studies have used age and tenure in particular as a basis of defining predictor
variables. The outcomes studied have included turnover (McCain, O'Reilley, &
Pfeffer, 1983; O'Reilley et al. 1989; Pfeffer & O'Reilley, 1987; Wagner et al. 1984;
& Jackson et al. 1991); communications (Zenger & Lawrence, 1989); innovation
(Bantel & Jackson, 1989); liking, role ambiguity, and performance ratings (Tsui &

O'Reilley, 1989); and team performance (Smith et al. 1994).

110

In their study, Jackson et al. (1991) found age and tenure heterogeneity were not
the only demography variables found to be predictive of group turnover rates. They
suggested that attributes other than age and tenure can be successfully incorporated into
the research on organizational demography. Given the emerging diversity of the
American workforce, they suggest the inclusion of additional background
characteristics. According to them, this would well serve the demography literature as
it would produce a closer match between Pfeffer's theoretical model and provide
empirical tests for his model. Accordingly, they suggest that the particular
demographic attributes that affect the feelings and behaviors of group members are
likely to depend upon the distribution of large sets of demographic attributes.

Using the methodology suggested by demography researchers (Wagner et al.
1984; O'Reilley et al. 1989; and Jackson et al. 1991) a range of ten background
variables--gender, age, race, level of education, organization and departmental tenure,
marital status, primary wage- earner, children living at home, part-time employment
status- were used to compute individual dissirnilarities of group members from their
group, as well as group heterogeneity indices for each demographic characteristic
assessed. Six hypotheses were tested using individual dissimilarity (3 hypotheses) and
group heterogeneity (3 hypotheses) as predictors of job satisfaction (J S), job
involvement (II), and company commitment (CC).

The results show that, when the ten individual dissimilarity variables were
collectively regressed with the outcome variables, individual dissimilarity moderately

predicted job involvement, had a strong impact upon company commitment, but had no

111

impact upon job satisfaction. Similarly, when ten group heterogeneity variables were
regressed collectively with the outcome variables, group heterogeneity moderately
predicted job satisfaction while having no effect on job involvement and company
commitment. The study also found none of the six hypotheses pertaining to the
interaction of satisfaction with diversity programs and the ten individual dissimilarity
and group heterogeneity variables to be significant when analyzed collectively. These
results suggest that diversity programs had no moderating effect on individual
dissimilarity and group heterogeneity.

Overall, this research suggests there is a very moderate support for the
contention that demography affects work-related attitudes. Pfeffer (1983) had
contended that demography variables would prove to be superior because they were
easily measurable and produce more parsimonious explanations. In his opinion,
demography is an important causal variable that effects a number of intervening
variables and processes and, through them, a number of organizational outcomes. He
has consistently maintained that the study of the underlying processes through which
group demography affects organizational performance was unnecessary, and questioned
whether intervening process variables could account for any incremental variation in
dependent variables beyond that explained by demographic measures alone. The
variance explanation assumes that, if the right demographic predictors are identified,
they will consistently explain some percentage of the variation in organizational

outcomes (Lawrence, 1994).

112

This research finds modest support for Pfeffer' s contention that demographic
measures, alone, significantly explain the variation in the dependent variables.
Recently, two studies done by Lawrence (1994) and Smith et al. (1994) challenge
Pfeffer's assumptions. Both studies suggest links between organizational demography
and organizational performance to unmeasured social psychological concepts.
Lawrence (1994) contends that organizational demography researchers have accepted
Pfeffer' 5 central argument without debate or discussion, and have ignored subjective
concepts that explain the significant association between demographic predictors and
outcomes--usually leaving these concepts unmeasured or untested. As a result,
subjective concepts and their relationships within research models have become the
”black box" of organizational demography. Lawrence further argues that Pfeffer's
instrumental approach, emphasizing prediction over explanation, poses serious
limitations and that process explanations may prove more appropriate. She emphasizes
the need to go beyond the variance model proposed by Pfeffer. For instance, in this
research population, heterogeneity explained little of the groups job involvement or
commitment. Perhaps, in combination with other factors such as personnel practices or
other group processes, different results might have been produced.

Smith et al. 's (1994) research found evidence partially supporting "process" as a
mediator of the relationship between demography and performance. In this research,
demography and process variables each effect performance separately. They found
little support for the argument that demography, rather than process, affects

performance. This contradicts Pfeffer's (1983) contention that process measures can be

113

dismissed because they account for little of the variation in outcomes. Instead, Smith et
al. (1994) found that team processes account for variability left unexplained by
demography. Their data also suggest that some forms of heterogeneity, such as
functional background heterogeneity, may have neither a direct nor indirect effect on
firm performance.

For reasons that are not entirely clear in this research, individual dissimilarity
had no effect upon job satisfaction and group heterogeneity also had no impact upon job
involvement and company commitment when a range of demographic variables were
analyzed collectively. This "non" finding may suggest Smith et al. 's (1994) contention
that all forms of functional background demography may have neither direct nor
indirect impacts on work-related attitudes that could, in turn impact organizational
outcomes. Even though one has to be very cautious against searching for ”the most
important type" of organizational demography (a concern raised by Jackson et al.
1991), the variance explanation makes such an assumption. Therefore, for statistical
reasons, it is necessary to consider each demographic attribute separately to explain the
effects of demography upon work-related attitudes. Each background variable related
to individual dissimilarity and group heterogeneity was separately analyzed to see
which variables were more significant predictors of work-related attitudes. AN OVA
analyses of the data to examine two-way interactions between groups (homogenous and
heterogenous) and individuals (similar and dissimilar) having less positive or more
positive attitudes towards diversity programs and practices for each background

attribute were undertaken.

114

AN OVA ANALYSES OF BACKGROUND VARIABLES
AN OVA analyses of the data to examine two-way interactions between

individuals (similar and dissimilar) having less positive or more positive attitudes
towards diversity programs and practices on each background attribute were tested with
each of the dependent variables (J S, J1, CC). For example, for each background
attribute such as race, similar and dissimilar individuals, having more positive or less
positive attitudes towards diversity programs, were tested for interaction effects with
each dependent variable. Similarly, homogenous and heterogeneous groups for each of
the background attributes were also tested. Significant interaction effects were found
only for one individual dissimilarity variable related to marital status (DIS) involving
job involvement and company commitment. None of the other individual dissimilarity
variables were found to be significant. Similarly, significant interaction effects were
found only for group heterogeneity variables pertaining to gender (GHl), education
(GH,), and marital status (GHs) involving job involvement and company commitment.
None of the other group heterogeneity variables were found to be significant.
AN OVAs involving job satisfaction showed no significant interaction effects. The
AN OVAs related to the overall satisfaction with diversity programs (OA) was
significant at the .000 level for both the groups having less positive or more positive
attitudes towards diversity programs.

The interaction effects for individual dissimilarity related to marital status (D1,)
for groups having less and more positive attitudes towards diversity programs with job

involvement (JD are reported in Table 10.1. Individual dissimilarity related to marital

1 15
status (DIS) for groups that were less similar or more similar on this attribute was not
found to be significant. However, the interaction effect was significant at the .02 level,
indicating that CA has a moderating effect on dissimilarity related to marital status.
Table 10.1 describes the interaction effects for individual dissimilarity related to
marital status (D1,) for groups having less and more positive attitudes towards diversity
programs with company commitment (CC). Individual dissimilarity related to marital
status (D1,) for groups that were less similar or more similar on this attribute was not
found to be significant. However, the interaction effect was significant at the .06 level,
indicating that CA has a moderating effect on dissimilarity related to marital status.
Table 10.1
Two-Way Analysis of Variance Showing Interaction Effects of Individual
Dissimilarity Related to Marital Status and Between Groups Having More Positive

or Less Positive Attitudes Towards Diversity Programs with Job Involvement and
Company Commitment

 

 

 

 

 

 

 

 

 

 

 

 

 

Source F-Ratio Significance I

Mm

Individual Dissimilarity-

Marital StatusL(DI,) .441 .51 u

Interaction 5.63 .02
W

Individual Dissimilarity—

Marital Status, (D19 .18 .68 ll

Interaction 3.64 .06 II

 

 

 

Two-Way Analysis of Variance Showing Interaction Effects of Group

116

Table 10.2

Heterogeneity Related to Gender, Education, and Marital Status, and Between
Groups Having More Positive or Less Positive Attitudes Towards Diversity

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Programs with Job Involvement
Source F-Ratio Significance
W

Group Heterogeneity-

Gender (GH,) 1 .06 .305
Interaction 5.71 .02

Group Heterogeneity-

Education (Gig) .86 .36
Interaction 3.72 .05

Group Heterogeneityu "
Marital Status (GHS) .831 .36
Interaction 4.63 .03

MW

Group Heterogeneity--

Gender (GH) .65 .42
Interaction 4.98 .05

Group Heterogeneity-

Education (GHQ 4.22 .04
Interaction 3.37 .07

Group Heterogeneity-

Marital Status (GHQ 4.22 .04 u
Interaction _ 3.37 1 .07 II

 

Table 10.2 describes the interaction effects for group heterogeneity related to

gender (6H,) for groups having less and more positive attitudes towards diversity

117

programs with Job Involvement (J1). Group heterogeneity (GHl) for groups that were
less heterogenous or more heterogenous on this attribute was not found to be
significant. However, the interaction effect was significant at the .02 level, indicating
that CA has a moderating effect on heterogeneity related to gender.

Table 10.2 describes the interaction effects for group heterogeneity related to
education (GH,,) for groups having less and more positive attitudes towards diversity
programs with Job Involvement (II). Group heterogeneity (GI-1,) for groups that were
less heterogenous or more heterogenous on this attribute was not found to be
significant. However, the interaction effect was significant at the .05 level, indicating
that CA has a moderating effect on heterogeneity related to education.

Table 10.2 describes the interaction effects for group heterogeneity related to
marital status (GHs) for groups having less and more positive attitudes towards
diversity programs with Job Involvement (J1). Group heterogeneity (GH,) for groups
that were less heterogenous or more heterogenous on this attribute was not found to be
significant. However, the interaction effect was significant at the .03 level, indicating
that CA has a moderating effect on heterogeneity related to marital status.

Table 10.2 describes the interaction effects for group heterogeneity related to
gender (GH1) for groups having less and more positive attitudes towards diversity
programs with Company Commitment (CC). Group heterogeneity (GH,) for groups
that were less heterogenous or more heterogenous on this attribute was not found to be
significant. However, the interaction effect was significant at the .05 level, indicating

that CA has a moderating effect on heterogeneity related to gender.

118

Table 10.2 describes the interaction effects for group heterogeneity related to
education (GI-1,.) for groups having less and more positive attitudes towards diversity
programs with Company Commitment (CC). Group heterogeneity (GIL) for groups
that were less heterogenous or more heterogenous on this attribute was found to be
significant at the .04 level. The interaction effect was significant at the .07 level,
indicating that CA has a weak moderating effect on heterogeneity related to education.

Table 10.2 describes the interaction effects for group heterogeneity related to
marital status (GH5) for groups having less and more positive attitudes towards
diversity programs with Company Commitment (CC). Group heterogeneity (GI-1,) for
groups that were less heterogenous or more heterogenous on this attribute was found to
be significant at the .04 level. The interaction effect was significant at the .07 level,
indicating that CA has a weak moderating effect on heterogeneity related to marital

status.
SUMNIARY OF HY POTI-IESES

Table 11 summarizes the significant relationships between work-related
attitudes and overall satisfaction with diversity programs and practices, various attribute
dissimilarity and group heterogeneity variables, and cross-products or interaction
variables. Overall, it was assessed whether diversity programs and practices predicted
job satisfaction, job involvement, and company commitment and moderated the effects
of individual dissimilarity and group heterogeneity on work-related attitudes. Table 11
shows that overall satisfaction with diversity programs and practices significantly
predicted job satisfaction, job involvement, and company commitment. The hypotheses

related to individual dissimilarity and group heterogeneity were moderately supported,

1 19
thereby indicating the efficacy of organizational demography in predicting work-related
attitudes in this study has been moderate. In addition, none of the hypotheses related to
the cross-products was significant, indicating that overall satisfaction with diversity
programs did not moderate the effects of individual dissimilarity and group
heterogeneity on work-related attitudes. A detailed summary and discussion of the

results will be presented in Chapter Five.

120
Table 11

Significant Relationships Between, Overall Satisfaction with Diversity Programs,
Individual Dissimilarity and Group Heterogeneity variables, and Cross-products
with Work-Related Attitudes (Job Satisfaction, Job Involvement, and Company

Commitment)

 

 

Variable

Actual

 

W

 

Individual Overall Satisfaction with Diversity Programs (OA)

 

Individual Dissimilarity (DI)

 

Cross-Product of OA‘DI (CPDI)

 

Group Overall Satisfaction with Diversity Programs (OA)

 

Group Heterogeneij' (GH)

 

Cross-products of OA*GH (CPGH)

 

 

W

 

Individual Overall Satisfaction with Diversity Programs (OA)

 

Imiividual Dissimilarity (DI)

 

Cross-products of OA‘DI (CPDI)

 

Group Overall Satisfaction with Diversity Programs (OA)

 

Group Heterogeneity (GH)

 

Cross-Products of OA‘GH (CPGH)

 

 

W

 

Individual Overall Satisfaction with Diversity Programs (OA)

 

Illiividual Disshnilam Q!)

 

Cross-Products of OA‘DI (CPGH)

 

Group Overall Satisfaction with Diversity programs (OA)

 

‘ Gm Heteggeneig (GH)

 

 

 

SRepresentsesignifleentreletionship.

 

CHAPTER FIVE
DISCUSSION

This chapter includes a discussion of the results of the dissertation research.
The discussion contains an overall summary of the findings presented in Chapter Four
compared to the problem as stated in Chapter One, the implications for research and
theory, the implications for practice, and suggestions for future research.

SUMMARY

Despite increasing gender and ethnic diversity of the workforce, organizations
have generally been at various stages of transition with regard to equal opportunities for
women and minorities. In response to the changing demographic trends, many
organizations have adopted a "valuing diversity" approach (Copeland, 1988) as a key to
competitive advantage. Towards this goal, many companies have increased their
emphasis on hiring, promoting, and retaining women, ethnic minorities, and people
with diverse backgrounds. They have also instituted multi-cultural training and other
activities to modify and address organizational systems that prevent diverse employees
from reaching their full potential (Thomas, 1990; Cox, 1991, 1993). However, few
studies have examined the effects of diversity programs in the workplace, despite the
fact that there is growing anecdotal evidence and case literature documenting the mixed
impact of diversity programs. Based on this assessment on the management of
diversity, the purpose of this study was to investigate the effect of diversity
programs/practices on work-related attitudes (job satisfaction, job involvement, and

organizational commitment). This research also tested whether the relationship

121

122

between individual dissimilarity and group heterogeneity and work-related attitudes was

likely to be moderated by an individual's perceptions of organizational efforts in

implementing diversity programs/practices.

The study had three objectives. The first objective was to examine the extent to
which an individual's perception of diversity programs and practices affects the impact
of individual similarity and dissimilarity on work-related attitudes. The second purpose
was to examine the extent to which an individual's perceptions of diversity
programs/practices affects the impact of group heterogeneity on work-related attitudes.
Finally, the research investigated the extent to which the experience of diversity
programs/practices affects individual and group interaction and work-related attitudes.

Specifically, the study findings suggest:

1. The correlation analyses of overall satisfaction with diversity programs/practices
with the three dependent variables--job satisfaction, job involvement, and
company commitment--were highly correlated showing a significant
relationship. Except for a single dissimilarity variable (children living at home)
none of the other dissimilarity variables or the group heterogeneity variables
was significantly correlated with the dependent variables. Subsequent
regression analyses revealed that overall satisfaction with diversity
programs/practices was a significant predictor of job satisfaction, job
involvement, and company commitment. This finding is consistent with

literature on diversity management (Thomas, 1990; Cox, 1991, 1993; Copeland,

123

1988, Caudron, 1993) in that organizations that invest in diversity programs
have positive organizational outcomes.

Individual dissimilarity was not a significant predictor of job satisfaction when
the ten individual dissimilarity variables were collectively analyzed.
Consequently, individual dissimilarity had no impact upon job satisfaction. This
finding is not consistent with Pfeffer's (1983) organizational demography theory
and Schneider's (1987) ASA literature.

Consistent with Pfeffer (1983) and Schneider (1987) findings, group
heterogeneity was a moderate predictor of job satisfaction when the 10 group
heterogeneity variables were analyzed collectively. This finding suggests that
group heterogeneity impacts job satisfaction.

Consistent with Pfeffer (1983) and Schneider (1987) findings, individual
dissimilarity was a moderate predictor of job involvement when the ten
dissimilarity variables were analyzed collectively. This suggests that individual
dissimilarity impacts job involvement.

Group heterogeneity was not a significant predictor of job involvement when the
ten group heterogeneity variables were collectively analyzed. This suggests that
group heterogeneity had no impact upon job involvement. This finding is not
consistent with Pfeffer's ( 1983) organizational demography theory and
Schneider's (1987) ASA literature.

Consistent with the work of Pfeffer (1983) and Schneider (1987), individual

dissimilarity was a strong predictor of company commitment when the ten

124
dissimilarity variables were analyzed; individual dissimilarity impacts company
commitment.

7. Using Pfeffer' s (1983) argument that demography affects organizational
outcomes, demography variables modestly enhanced the percentage of variance
explained in job satisfaction, job involvement, and company commitment.
Subsequent AN OVA analyses, which tested each dissimilarity and heterogeneity
variables separately, produced similar results for both homogenous and
heterogenous groups.

8. The study also found that there was no interaction of satisfaction with the ten
individual dissimilarity and group heterogeneity variables. This finding
suggests that diversity programs had no moderating effect on individual
dissimilarity and group heterogeneity. Subsequent AN OVA analyses, which
tested for interaction effects for each dissimilarity and heterogeneity variables,
found little difference for homogenous and heterogenous groups.

IMPLICATIONS FOR RESEARCH AND THEORY
Chapter Two described many reasons why managing diversity is important to
organizations, and presented two theories that may explain why dissimilar individuals
and heterogenous groups have difficulty fitting into work environments. The results of
this study suggest that diversity programs predict work-related attitudes. Specifically,
the overall satisfaction with diversity programs was strongly related to job satisfaction,
job involvement, and company commitment. Individual dissimilarity and group

heterogeneity moderately predicted work-related attitudes. However, diversity

125

programs showed no moderation effects on the relationship between individual
dissimilarity and group heterogeneity and work-related attitudes.

The finding that overall satisfaction with diversity programs/practices are
strongly related to work-related attitudes is consistent with scholars' assumptions which
suggest that investment in diversity programs lead to greater achievement of individual
ego needs and contribute to increased morale, motivation, and satisfaction (Shipper &
shipper, 1987; Gerber, 1990; Copeland, 1988; Cox, 1991, 1993; Loden & Rosner,
1991; and Thomas, 1990). Due to these reasons, there is an increased knowledge,
sensitivity, promotion, and acceptance of differences. Gerber (1990) has argued that
one of the goals of diversity is to treat people as individuals, recognizing that each
employee has different needs and will need different kinds of help to succeed.
Similarly, both Thomas (1991) and Cox (1991, 1993) have defined managing diversity
as a comprehensive managerial process for developing an environment that works for
all employees. These researchers have theorized that managing diversity should lead to
higher levels of organizational effectiveness and positive work behaviors. However,
none of the above mentioned researchers directly assessed the effects of diversity
programs on work-related attitudes. This research suggests that there is strong
relationship of satisfaction with diversity programs and work-related attitudes.

This study also included several other hypotheses that linked individual
dissimilarity and group heterogeneity to work-related attitudes. These hypotheses were
based upon current thought in Schneider's (1987) ASA framework and Pfeffer's (1983)

organizational demography literature and empirical evidence based on these two

126

theories. Jackson et al. (1991) have posited that, as culturally diverse groups are likely
to become the norm within organizations of the future, the impact of managing
diversity on individual and group interaction and performance would become critical.
Tsui et al. (1991) had argued that any individual can be different from, or similar to,
any other individual in a social unit on a demographic attribute (e.g. , race, gender)
being considered. Demographic attributes are information that individuals might use to
infer one's similarity to such things as attitudes and beliefs (Byrne, 1971). This is
similar to Schneider's (1987) emphasis on the role of "person effects" as determinants
of behavior in organizations, where personality, interests, and values are the
dimensions of similarity assumed to influence attraction to organizations and the people
in them. The homogeneity of personalities, values, and interests that characterize
members within an organization are what accounts for their unique quality.

Jackson et al. (1991) assert that Schneider's (1987) perspective refutes the often-
made assumption that people (and their personal attributes) are randomly distributed
across organizations. Instead, a restriction in range occurs with similar kinds of
people, who exhibit similar behaviors, are clustered together. The similarity effect
provides a rationale for how and why demographic compositions of organizations are
likely to be related to organizational phenomenon. In this respect, Jackson et al. (1991)
contend both models are similar in phenomenon and process, and therefore need to be
integrated. However, the focus on organizations as a unit of analysis clearly
distinguishes Pfeffer's sociological perspective from Schneider's psychological

perspective with its focus on individual level constructs of similarity and attraction.

127

The presumed similarity in attitudes or beliefs might influence the individual's
attraction toward the other individuals. Consequences of low attraction include less
communication, lower social integration, and eventual turnover (T sui et al. 1991).
Both the organizational demography and the ASA framework suggest that dissimilarity
and heterogeneity are negatively related to organizations effectiveness. In fact,
research indicates that demographic attributes are associated with differences in
attitudes, values, and beliefs that have the potential to create conflict among team
members which, in turn, can influence group outcomes and behaviors (Daft & Weick,
1983; Pfeffer, 1983). Individual dissimilarity and heterogeneity may create equal levels
of discomfort for all group members. It may limit the integration and development of
cohesiveness at the level of the group as a whole, and thereby influence work-related
attitudes and behaviors (Jackson et al. 1991).

Not all the hypotheses related to individual dissimilarity and group heterogeneity
were supported. For example, individual dissimilarity moderately predicted job
involvement and had a strong impact upon company commitment, but showed no
impact on job satisfaction. Similarly, group heterogeneity moderately predicted job
satisfaction while having no effect upon job involvement and company commitment.
Pfeffer (1983) has contended that demographic variables are superior because they are
easily measurable and produce parsimonious explanations. In his opinion, demography
is an importantly causal variable that effects a number of intervening variables and
processes and, through them, a number of organizational outcomes. He has

consistently maintained that the study of the underlying processes through which group

128
demography affects organizational performance was unnecessary. He has questioned
whether intervening process variables could account for any variation in dependent
variables beyond that which could be explained by demographic variables alone.
Lawrence (1994) points out that, from the variance explanation presented by Pfeffer, it
can be assumed that if the right demographic predictors are identified they will
consistently explain some percentage of the variation in organizational outcomes.

Smith et al. '8 (1994) recent study challenges Pfeffer's assumptions. Their study
suggests links between organizational performance to unmeasured social psychological
concepts. Smith et al. 's (1994) research found little evidence for the argument that
demography, rather than process, affects performance. Instead, they found that team
processes account for variability left unexplained by demography. They also found that
all forms of functional demography did not have a direct or indirect impact on
organizational outcomes. Underlying processes could, in turn, influence values,
attitudes, and beliefs limiting the integration of the group (Schneider, 1987). In this
research, individual dissimilarity had no effect upon job satisfaction, and group
heterogeneity had no impact upon job involvement and company commitment, when a
range of demographic variables were analyzed collectively. This suggests that all forms
of functional demography might not have a direct or indirect impact on work-related
attitudes that might, in turn, influence organizational outcomes.

Clearly, the percentage of variance explained was enhanced modestly by the use
of individual dissimilarity and group heterogeneity. In this study, it is possible that a

combination of other factors, such as personnel practices or other group processes (not

129

measured in the study) might have combined to moderate the effects of individual
dissimilarity and group heterogeneity. Another possible explanation could be that the
"right" demographic predictors were not identified in this research--although such an
explanation would undermine the theoretical assumptions underlying the organizational
demography literature. It is also possible that functional demography might not be the
best predictor of what Schneider (1987) calls "person effects"-personalities, interests,
values, beliefs, and attitudes as determinants of behavior in organizations. If one
assumes demographic diversity in current organizations to be a fact, factors like
personality, interests, and values might be other important dimensions that need to be
considered along with demographic variables. They could very well moderate the
effects of individual dissimilarity and group heterogeneity.

This study also included several hypotheses relating to the interaction of
satisfaction with diversity programs and the ten individual dissimilarity and group
heterogeneity variables to work-related attitudes. These hypotheses were based upon
the current literature on diversity that contends that company policies and practices,
which positively change and impact the experiences of women, minorities, and other
interest groups, will enhance the work environment and result in employee development
and productivity (Copeland, 1988; Cox & Blake, 1991; Thomas, 1991). Based on this
assumption, many organizations have launched a number of diversity
programs/practices that have been designed to eliminate subtle barriers that prevent
different people from working together and to create climates that are supportive of a

heterogenous work environment (Braham, 1989; Thomas, 1990; Cox, 1991, 1993).

130

Research by Pfeffer (1983), Schneider (1987), and Jackson et a1. (1991) supports the
view that information about a person's demographic characteristics influences both
attributes regarding the person's psychological character and behavior towards the
person and his/her membership within the group. They contend that demographic
attributes are likely to determine both the perceptions of similarity/dissimilarity and
perceptions of person-environment fit. Based on this assumption, one of the premises
for implementing diversity programs is that they are likely to facilitate the adjustment
of dissimilar individuals and heterogenous groups within organizations and, in turn,
affect work-related attitudes. In this research, none of the hypotheses related to cross-
products was significant. This indicates that satisfaction with diversity programs did
not moderate the effects of individual dissimilarity or group heterogeneity on work-
related attitudes.

In most organizations, the institution of diversity programs is relatively new.
Many organizations have instituted these programs because of the changing
demographics within their organizations that correspond to trends within the workforce.
In some cases, diversity programs have been instituted due to ”social desirability"
factors and the current emphasis being given to diversity in the management literature.
Due to these reasons, in this research it is possible that management may not have
properly assessed the needs of its individual members and different employee groups
and therefore has not provided them with the necessary support. For example, in this

research sample, a majority of the employees were women who were concentrated at

131

lower-level clerical positions. Diversity programs in the organizations under study
might not have addressed specific needs of this population.

Another explanation deals with methodological and measurement issues. In this
study, participants were given a list of eighteen diversity programs/practices in
common usage across the country. Participants were asked whether they had
knowledge of each program and how effective each program was. Based on this,
participants were asked to rate their overall satisfaction with diversity
programs/practices within their organizations. Perhaps the single measure used to rate
the overall satisfaction with diversity programs was not robust enough to moderate the
effects of individual dissimilarity and group heterogeneity on work-related attitudes.
Another explanation could be the moderate effect of individual dissimilarity and group
heterogeneity as predictors of work-related attitudes that accounts for the lack of
interaction effects in this research.

IMPLICATIONS FOR PRACTICE

Johnston & Packer (1987) highlighted the demographic and organizational
forces that have led to the changing face of the workforce with new conditions under
which work is being accomplished. They point out the high costs of ignoring
demographic changes in the workforce of 2000. The changing demographics are being
accompanied by modifications in employee attitudes, motivations, and values (Odiorne,
1986), and an increased desire for autonomy, self-development, balance between work
and family life, and more meaningful work experiences (Hall, 1986). As the workforce

continues to become more diverse, issues of motivation and attitudes towards work will

132

become more and more critical. Organizations will need to be more attuned to the
potential expectations of individual employees and of different groups. Organizations
need to replace their traditional focus on conformity through assimilation by a true
understanding of integration (Offerrnann & Gowing, 1990). The diversity of the
workforce warrants the need for research that will shed light on ways to manage a more
diverse workforce.

This dissertation provided support for the idea that both individual and group
perceptions of satisfaction with diversity programs are related to positive work-related
attitudes. There are two key implications for these findings. The first implication deals
with the expectation that diversity training programs will serve to maximize the
potential of each person. The second implication deals with structuring employee

attitudes about diversity through conventional human resource management (HRM)

practices. Each of these implications will be addressed in turn.

 

In the three financial institutions researched for this study, there was no
evidence that diversity training interventions were a component embedded within the
larger set of organizational systems. The diversity programs instituted seemed to be
rather ad hoc and piecemeal interventions that were not based upon systematic need
assessment. If diversity training is to maximize its impact and result in positive
organizational behaviors, a careful needs assessment is required. Morris (1994)
reported that Nestle Company's success with its diversity programs was due, in part, to

the needs assessment research done to accurately understand people's perceptions, to

133

link diversity training with organizational goals and values, and to ensure that a great
deal of energy would not be wasted on issues that would have little payoff. In essence,
Nestle's research helped to focus diversity efforts on areas that would be of greatest
benefit to the company.

Effective diversity training also requires that part of the assessment process
focus on organizational analyses. This includes an examination and audit of
organizational systems that function as barriers. These barriers could potentially
interact with a program's design, development, and effectiveness. Goals, values,
climate, and external and internal constraints present within the organizational
environment also need to be examined as they have a potential to interact with the
effectiveness of training programs.

Currently, diversity training programs are being widely viewed as interventions
to change work behaviors and attitudes. If diversity training is to be successful, there is
a need to collect data regarding the effectiveness of these programs. Data should also
be collected in order to provide feedback that can be used for continuous improvement
in diversity training. Evaluation of diversity training should determine whether
changes have occurred during training, how the training is being transferred to the
work environment, and the extent to which the impacts are felt at the individual and
group level.

This research was not designed to study and evaluate the effectiveness of
diversity training programs. But this research has shown ample evidence that

investment in diversity training and the institution of practices that accommodate

134

diversity have large payoffs in work-related attitudes. The research data showed that,
at the macro-level, satisfaction with diversity programs led to positive work-related
attitudes. However, at the micro-level, they did not have the desired impact on
individual dissimilarity and group heterogeneity. This finding indicates that the
benefits of training could be further enhanced by embedding diversity training and
management within the context of larger organizational systems.

The results of this study also suggest that there are benefits to organizations that
invest in diversity programs and practices. The results of the study show that diversity
programs/practices lead to positive work-related attitudes. The results also suggest that
organizations have yet to benefit from the interaction of these programs with dissimilar
individuals and heterogenous groups. Specifically, in order to influence behavior and
reap the benefits of diversity, management needs to develop a wide variety of effective
programs and practices. Such practices include an audit of all HRM practices that
inhibit the effective use of all employees and prevent diverse individuals from reaching
their full potential.

Management also needs to link diversity management to organizational goals,
values, and business strategy. Such a link requires top management commitment and
accountability to achieve measurable goals pertaining to the management of diversity.
To the extent that management is unable to articulate the linkage of its diversity
management to goals, values, and business strategy, diversity programs are likely to

remain ad hoc interventions without any long-term impacts. One of the potential

135

downsides of piecemeal interventions is unfulfilled and unmet expectations among
employees that could lead to negative work-related attitudes.

Management should also be concerned with external demographics in the
workforce, as well as be concerned about internal organizational demography of its
workforce. External workforce trends will influence long-term recruitment and hiring
practices within the firm. Organizational demography, on the other hand, has
consequences for relationships of individuals within a group. Based on personal
attributes, interpersonal and intergroup conflicts might arise that could lead to attrition
of "dissimilar" individuals from the group. Given our global economy, consequences
of homogeneity could mean less creativity, less innovation, less flexibility to change,
less of a customer focus and, ultimately, the loss of an economic market share.
Organizational demography can be helpful in identifying important personal attributes
that effect organizational outcomes which, in turn, have implications for the types of
diversity interventions that fit organizational need. To the extent that interventions can
create effective and diverse teams and groups, management stands to benefit through
creativity, innovation, and adaptability to change in a global environment.

Finally, this research suggests that employees value organizations that invest in
diversity programs. Data from the research show that the three organizations in the
study were not sufficiently vested in diversity programs related to work-family issues.
Research by Chadwick-J ones et al. (1982) shows that flexible work schedules have
been effective reducing absenteeism. Other studies have shown that companies, which

provide childcare assistance not only experienced less absenteeism and turnover, but

136

were able to improve worker morale, organizational commitment, and job satisfaction
(Y oungblood & Chambers-Cook, 1984). In our society, most women tend to have a
disproportionate burden and share of managing work-family problems and issues. To
the extent management in these organizations focus on work-family issues, and vests in
programs that deal with such issues, it stands to benefit from a more committed female
workforce.

SUGGESTIONS FOR FUTURE RESEARCH

The results of this study have demonstrated, to a sufficient degree, the efficacy
of satisfaction with diversity programs to predict work-related attitudes. Although it
has shed light on key issues in the diversity literature, it has raised many more
questions, than answers. Several suggestions for future research are offered.

First, the results of this study are generalizable to other organizations that have
diversity programs. However, future research should be conducted that examines
diversity programs in organizations located in different industrial sectors, organizations
with different diversity programs/practices, and organizations that have both unionized
and non-unionized workforces.

Second, because the link between diversity programs and work-related attitudes
was established, research into organizational policy that fosters diversity may be a
worthwhile avenue of pursuit. Specifically, research that provides insights into how
each employee interprets these policies and its effects upon organizational outcomes can

be a fruitful area of research.

137

Third, the relationship between individual dissimilarity and group heterogeneity
and work-related attitudes in this study raises questions regarding the power of
organizational demography to predict work-related attitudes and influence other
organizational outcomes when a range of demographic attributes are analyzed
collectively. This research focused upon lower-level organizational groups using ten
different background demographic variables. It assumed demographic attributes are
information individual's infer similarity to such things as attitudes, values, and beliefs
(Byrne, 1971; & Schneider, 1987). Smith et al. (1994) have suggested that all forms of
functional background demography may have neither direct nor indirect impacts on
organizational outcomes. The modest impact of organizational demography in
predicting work-related attitudes in this research seems to attest to Smith et al. 's (1994)
suggestion. Another explanation could be that functional demography might not fully
explain people's underlying values, attitudes, and beliefs that could impact work-related
attitudes. Perhaps future research, which focuses on other lower level organizational
teams and uses different background variables (demographic and psychological), might
produce results that confirm the efficacy of organizational demography and the ASA
model to predict work-related attitudes.

Finally, this dissertation studied whether diversity programs function as
moderators of the effects of individual dissimilarity and group heterogeneity on work-
related attitudes. This study produced no moderating effects. Diversity programs are
still very young in organizations. Their long-term impact has not yet been felt within

most organizations.

LIST OF REFERENCES

LIST OF REFERENCES

Alderfer, C. P. (1971). Effect of Individual, Group, and Intergroup Relations
on Attitudes Towards a Management Development Program. W

Annfiedxsxchnlne! V01 55 pp 302- 311

Alderfer, C. P. Alderfer, C. 1., Tucker, L. A, Tucker, R. (1980). Diagnosing

Race Relations Management.1an:nal_of_Anplied_Behazioral_Science.
Vol. 27, pp. 135-166.

Alderfer, C. P. (1991, August). Remarks Made at the W

W Annual
Meetings of The Academy of Management. Miami, Florida.

Aldrich, H. E. (1979). QrganizafionsaniEnzimnments. Englewood Cliffs,
NJ: Prentice-Hall.

Allison, Paul D. (1978). Measures of Inequality. WW
Rulers, Vol. 43, pp. 865-880.

Aranya, N. , Kushnir, T. , & Valency, A. Organizational Comnritrnent in a
Male-Dominated Profession. (1986). HumanBelatiens. Vol. 39, pp.
433-448.

Argyris, C. (1976). Problems and New Directions for Industrial Psychology. In

Dunnette M D (ed) Handhmknflndnstrialandflrganizational
Bsxcholoex Pp 151- 184, Chicago. Rand, McNally.

Babbie, Earl. (1989). W. (5th ed.). Belmont,
CA: Wadsworth Publishing Company.

Bantel, Karen A., & Jackson, Susan E., (1989). Top Management and
Innovations in Banking. Does the Demography of the Top Team Make a

Difference?. SnategieManagementJQumal. Vol. 10, pp. 107-124.

Belfry, M. & Schmidt, L. (198889). Managing the Diverse Workforce.
EmploymenLerationledax Winter pp 335-339

Bell, E. (1990). The Bicultural Life Experience of Career-Oriented Black
Women .Ionrnalnfflganizationalfiehazinr Vol 11. pp 459-478.

140

141

Berg, D. N., & Smith, K. K. (1990). Paradox and Groups. In J. Gillette, & M.
McCollum (eds). W. Reading, MA. Addison-Wesley.

Blau. F. D.. & Ferber. M. A. (1986). Want!
Work. Englewood Cliffs, NJ: Prentice-Hall.

Blau, G. J. (1985). The Measurement and Prediction of Career Commitment.
.Imrrnalnfnccunatinnallsycholm V01 58 pp 277-288

Blau, Judith R., & Alba, Richard D. (1982). Empowering Nets of

Participation. WW V01 27 PP 363-
379.

Blau, Peter M. (1970). A Formal Theory of Differentiation in Organizations.
AmericanSnciologicaLRexiew. vol 35 PP 201-218-

(1977). W. New York: Free

 

Press.

Bluedom, Allen C. (1982). The Theories of Turnover: Causes, Effects, and

Meanings. In Samuel B. Bacharach (ed.). WW
Qrganizatiens. Vol. 1, pp. 75—128, Greenwich, CT: JAI Press.

Bowers, K. S. (1973). Situationism in Psychology. An Analysis and Critique.
W, Vol. 80, pp. 307- 336.

Braham, J. (1989). No You Don't Manage Everyone the Same. Industry
Week, Iss. 238. Pp. 28-35.

Brown. L. D. (1983). MW.
Reading, MA: Addison-Wesley.

Burt Ronald S (1982) W
W. New York:

Academic Press.

Byrne, D. (1969). Attitudes and Attraction. In Leonard Berkowitz (ed. ),
AdlanminExperimentaLSociaIMholngy. 4: 35- 89. New York:

Academic Press.

Byrne, Donn E. (1971). IheAttraetien Paradigm. New York: Academic
Press.

142

Byrne, D. Clore, G. L., Jr. & Smeaton, G. (1986). The Attraction
Hypothesis: DoSimilar Attitudes Affect Anything?. ,[QnrnaLof

WW V01 51 PP 1167- 1170

Campbell, Donald T., & Fiske, D. W. (1959). Convergent and Discriminant
Validation by the Multitrait-Multirnethod Matrix. Psychological
Bulletin, Vol. 56, pp. 81- 105.

Caudron, Shari. (1993). Training Can Damage Diversity Efforts. Personnel
,lnnrnal, April, pp. 51-62.

Chadwick-Jones, J. K., Nicholson, N., & Brown, C. (1982). 5min]
RsyebelegLQLAbsenteeism. New York: Praeger.

Cheloha, R. S., & Farr, J. L. (1980). Absenteeism, Job Involvement, and Job

Satisfaction, in an Organizational Setting. ,[mirnalbLAnnlied
25351119192! Vol. 65, pp. 467-473.

Cohen, A., Lowenberg, G., & Rosenstein, E. (1988). A Re-Examination of the
Side-Bet Theory as Applied to organizational Commitment: A Meta-
Analysis. Unpublished Manuscript.

Cohen, Jacob & Cohen, Patricia. (1983). AW

 

ed.). Hillsdale, NJ: Lawrence Erlbaurn Associates.

Cook, T., & Campbell, D. (1979) ansL-Expcnmentatinnamsignanu
W Chicago Rand MCDaHY-

Copeland, Lennie. (1988). Learning to Manage a Multicultural Workforce.
Iraim'ng, May, pp. 48-56.

Cotton, 1. L., & Tuttle, J. M. (1986). Employee Turnover: A Meta-Analysis

and Implications for Research. AeadeinLQLMamgemenLReyieg, Vol.
11, pp. 55-70.

Cox, Taylor H. (1988). WWW.
Unpublished Working Paper, University of Michigan, Ann Arbor.

Cox, Taylor H. (1991). The Multicultural Organization. IbeExeentiye. Vol.
5, Iss. 2, pp. 34-47.

143

Cox, Taylor H. (1993).£u|t11ral_DJ1ersitJLin_Qmanlzations (lst ed). San
Francisco, CA: Berrett-Koehler Publishers.

Cox, Taylor H. (1991). The Multicultural Organization. Amdmnf
Managementlxecufile, No. 5, pp. 3447.

Cox, Taylor H., & Blake, Stacy. (1991). Managing Cultural Diversity:
Implications for Organizational Competitiveness. Agadmgf

ManagmenLExesufixe, Vol. 5. No. 3. pp- 45-56-

Cox, Taylor H., Lobel, Sharon A., McLeod, Poppy Lauretta. (1991). Effects
of Ethnic Group Cultural Differences on Cooperative and Competitive

Behavior on a Group Task. WM.
Vol. 34, Iss. 4, December, pp. 827-847.

Cox, Taylor H., & Nkomo, Stella M. (1991). A Race and Gender-Group
Analysis of the Early Career Experiences of MBAs. W
W, Vol. 18, No. 4, November, pp. 431-446.

Daft, R. L., & Weick, Karl E. (1984). Toward a Model of Organizations as

Interpretation Systems. MW, Vol. 9, pp.
284-296.

Denzin, Norman K. (1978). W. (2nd ed.). New York:
McGraw-Hill.

Dobbins, G. H., & Platz, S. J. (1986). Sex Differences in Leadership: How

Real are They?. MW. Vol. 11, pp. 118-
127.

Donell, S. M., & Hall, J. (1980). Men and Women as Managers: A
Significant Case of no Significant Difference. Organizational
Dxnamiss. Vol. 8, pp. 60—77.

Ebadi, Y., & Utterback, J. M. (1984). The Effects of Communication on
Technological Innovation. ManagemgnLSgiencg, Vol.30, pp. 572-585.

Equal Employment Opportunities Commission. WW
1282.

144

Ferdman, B. (1992). The Dynamics of Ethnic Diversity in Organizations:
Toward Integrative Models. In K. Kelly (ed). Wand

WWW Amsterdamz North
Holland, pp. 339 383.

Fernandez, J. P. (1975). W. New
York: Wiley.

Fernandez, J. P. (1981). RacismandSexismjnLormrmLife Lexington.
MA: Lexington Books.

Galen, Michele, & Palmer, Ann Therese. (1994). White Male and Worried.
Biminessfleek. January 31,1994.

Gerber, B. (1990). "Managing Diversity." IrainingMagazine. July.

George, J. M, (1990). Personality, Affect, and Behavior 1n Groups. ,InurnaLof
W, Vol. 75, pp. 107- 116.

Good, Lawrence R. & Nelson, Don A. (1971). Effects of Person-Group and
Intra-Group Attitude Similarity on Perceived Group Attractiveness and

Cohesiveness. W, Vol. 25, pp. 215-217.

Greenhaus, J. H., Parasuraman, S., & Wormley, W. (1990). Effect of Race on
Organizational Experiences, Job Performance Evaluations and Career

Outcomes. ArademJLoLManagemenLloumal. Vol. 33. pp. 64-86.

Guthrie, J. P., Olian, J. D. &Gupta, A. K. (1990). W

W Unpublished
Manuscript, University of Kansas.

Hall, D. T., Goodale, J. G., Rabinowitz, S., & Morgan, M. A. (1978). Effects
of Top-Down Departmental and Job Change Upon Perceived Employee
Behavior ad Attitudes: A Nation Field Study. MAW

1253521191921, Vol. 63, pp. 62-72.

Hall, D. T. (1986). An Overview of Current Career Development Theory,
Research, and Practice. In D. Hall and Associates (Eds. ). Caren

WWW (pp. 1-20). San Francisco, CA: Jossey-

Bass.

145

Harlan, A., & Weiss, C. L. (1982). Sex Differences in Factors Affecting
Managerial Career Advancement. In P. A. Wallace (Ed.). Min
Wakes (pp. 59-100). Boston, MA: Auburn House.

Hays, W. L. (1981). W. New York: Holt, Rinehart & Winston.

Helms, J. E., & Giorgis, T. W. (1980). A Comparison of the Locus of Control
and Anxiety Level of African, Black Americans and White American

College Students. WWW. pp. 503-509.

Hoffman, Eric. (1985). The Effect of Race-Ratio Composition on the Frequency

of Organizational Communication. W Vol.
48, pp. 17-26.

Holland, J. L. (1976). Vocational Preferences. In M. D. Dunnette (Ed..)

W. Chicago:
Rand McNally.

Houghton, James R. (1988). To Add Corporate Value, Break the Glass Ceiling.
W, July, pp. 32-36.

Howard. A., & Bray. D. W. (1988). Managerialmmlransition New
York: Guilford Press.

Hrebiniak, L. G. & Alutto, J. A. (1972). Personal and Role-Related factors 1n
the Development of Organizational Commitment. Administratixe

Scienccflnarterlx Vol 17 PP 555 573

Ilgen, D. R., & Youtz, M. A. (1986). Factors Affecting the Evaluation and
Development of Minorities in Organizations. In K. Rowland and G.
Ferris (eds.). Research in Personnel and Human Resource
Management: A Research Annual, Greenwich, CT: JAI Press, pp.
307-337.

Irons, E., & Moore, G. (1985). MW
Industry. New York: Praeger.

Jackson, S. E. (1991). Team Composition in Organizational Settings: Issues in
Managing an Increasingly Diverse Workforce. In S. Worchel, W.

Wood, & J. Simpson (Eds.). W- pp. 138-
173, Beverly Hills, CA: Sage.

146

Jackson, Susan E., Brett, Joan E, Sessa, Valerie 1., Cooper, Dawn M., Julin,
Johan A., & Peyronnin, Karl. (1991). Some Differences Make a
Differences: Individual Dissimilarity and Group Heterogeneity as
Correlates of Recruitment, Promotions, and Turnover. Jnnrnainf

Appliedxmhnlngx, Vol. 6, No. 5, pp. 675-689.

J ans N. A. (1985). Organizational Factors and Work Involvement.
. - . . . ' . ‘ .V.ol 35, pp.

 

Jerich Barbara, Copeland, Lennie, Boyles, Monica, & Robert, T., (1989).
"How do You Manage a Diverse Workforce?". Irainingand

W. Vol. 43, Iss. 2, pp. 13-21.

Johnston, W. B., & Packer, A. H. (1987). W
W. Indianapolis, IN: Hudson Institute.

Jones, E. W. (1986). Black Managers: The Dream Deferred. Hanan!
W. Vol 64 188 3 Pp 84-93

Kanter. Rosabeth Moss. (1977). Menandflomennflhefimation.
New York: Basic Books.

Kanter, R. M. (1979). Differential Access to Opportunity and Power. In R.
Alvarez (Ed..) Dismm’matinnjnflrganizations pp 52-68 San

Francisco. J ossey-Bass

Kanter, R. M., & Birkenhoff, D (1981). Organizational Performance: Recent

Developments 1n Measurement. MW, Vol. 7,
pp. 321-349.

Karp, H. B., Sutton, Nancy. (1993). Where Diversity Training Goes Wrong.
Training. July, pp. 30-34.

Katz, Daniel, Kahn, Robert L. (1977). WWW
Organizations. New York: Wiley Press.

Kirchmeyer, Catherine, & Cohen, Aaron. (1992). Multicultural Groups: Their
Performance and Reactions with Constructive Conflict. .Gmum

QrganizatinnManagement Vol 17 No 2 June pp 53-170

147

Kossek, Ellen E., & Zonia Susan C. (1992). Assessing Diversity Climate: A
field Study of Reactions to Employer Efforts to Promote Diversuy.

I lffl'l’lBl'.

Kraiger, K, & Ford, J. K. (1985). A Meta-Analyses of Ratees Race Effects in

Porforrlaanoe Ratings WW V01 70 PP 56-
65.

Kuster, Jeffrey C. (1993). Occupational Employment in Commercial Banking,
1987-90. W. Vol. 116. No. 4. PP- 2145-

Larwood, L. A., Gattiker,U .(1984). Study Shows First Job 13 Key to a Career
. . ' - l. Vol. 73, Iss.

 

Larwood, L., Gattiker, U. (1987). A Comparison of the Career Paths Used by
Successful Women and Men. In B. A. Gutek & L. Larwood (Eds.).

flamenlLCaneeLDexelnnment. Beverly Hills. CA: Sage-

Lawrence, Barbara S. (1994). The Black Box of Organizational Demography.
Unpublished Working Paper. University of California at Los Angeles.
pp. 1-51.

Lee, T. N ., & Mowday, R. T. (1987). Voluntarily leaving an organization:
An Empirical Investigation of Steer's & Mowday's Model and Turnover.

AcademLQLManagemenLIoumal. Vol. 30.913. 721-743-

Lincoln, J. R, & Miller, J. (1979). Work and Friendship ties in Organizations:
A Comparative Analysis of Relational Networks. Administrafixe

W. Vol. 24, pp. 181- 199.

Locke, E. A. (1983). The Nature and Causes of Job Satisfaction. In M. D.

Dunnette (Ed...) Handlmkoflndustrialandflrganizatinnl
Psychology PP. 1297- 1349. Chicago: Rand McNally.

Lodahl, T. M, & Kejner, M. (1965). The Definition and Measurement of Job
Involvement. WW, Vol. 49, pp. 24-33.

Loden M &Rosener J B (1991) W
W (lsted) Homewood IL

Business One Irwin.

148

Lyrroh Frederick R (1991) W
W. (Paperback ed.) New York: Praeger.

MacDonald, Heather (1993). The Diversity Industry. W, July
5, pp. 22-25.March, J. G.., & Simon, H. A. (1958). meanizations.
New York: Wiley.

Mayer, Roger C. , & Schoorman, David F. (1992). Predicting Participation and
Production Outcomes Through a Two-Dimensional Model of

Organizational Commitment. “WW. Vol.
35, No. 3, pp. 671-684.

McCain, B. R., O'Reilly, C. A. III., & Pfeffer, J. (1983). The Effects of

Departmental Demography on Turnover. W111
journal, Vol. 26, pp. 626-641.

Miller, Jane G. & Wheeler, Kenneth G. (1992). Unraveling the Mysteries of
Gender Differences 1n Intentions to Leave the Organization. .[QnrnaLQf

Organizationaljielmior Vol 13 PP 465-473

Mirvis, P. H., & Lawler, E. E. III. (1977). Measuring the Financial Impact of

Employee Attitudes. MAW. Vol. 62, pp. 1-
8.

Mitchell, T. (1985). An Evaluation of the Validity of Correlation Research

Conducted in Organizations. MW. Vol.
10, pp. 192-205.

Mobley. W H (1982) WWW
Cnntml. Reading, MA: Addison-Wesley.

Morris, J. H, & Steers, R. M. (1980). Structural Influences on Organizational
Commitment Wham. Vol 17 50-57

Morris, J. H., & Sherman, J. D. (1981). Generalizability of an Organizational

Commitment Model. AcademynflManagmenLlanrnal. Vol. 24., pp.
512-526.

Morris, Linda. (1994). Diversity Effort Linked to Clirnate-Study Results.
IrainingLDexelonment Vol 48.1ss 1 pp. 64-66

149

Morrison, A. M., White, R. P., & Van Velsor, E. (1987). Executive Women:
Substance Plus Style. W, August, pp. 18-25.

Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The Measurement of

Organizational Commitment. WW, Vol. 14,
pp. 224-247.
Mowday, R. T., Porter, L. W., & Steers, R. M. (1982). Employee;

W
W. New York: Academic Press.

Muchinsky, P. M., & Tuttle, M. L. (1979). Employee Turnover: An Empirical

and Methodological Assessment. WW, Vol.
14, pp. 43 -77.

Murray, Alan I (1989). Top Management Group Heterogeneity and Firm
Performance W. Vol 10 PP 125 141

Murray, Kathleen (1993). The Unfortunate Side Effects of Diversity Training. "
W, Sunday, August 1,
Bussiness Section, pp. F 5.

Newcomb, T. M. (1961). W. New York: Holt,
Reinhart & Winston.

Nixon, R. (1985). ' . ' . e. ,
mmgration. Washington, D. C: National Urban League.

 

Nunnally, J. C. (1978). Psychometric Theory, New York: McGraw-Hill.

Odiorne, G. S. (1986) The Crystal Ball of HR Strategy. Personnel
Administrator Vol 31 PP 103-106

Offermann, Lynn R., & Gowing, Marilyn K. (1990). Organizations of the

Future: Changes and Challenges. Americanmmlogist. Vol. 45,
No. 2, pp. 95-108.

Olson, C. A., & Becker, B. E. (1983). Sex Discrimination 1n the Promotion

Process IndustrialandLahoLRelationsRexim Vol 36. pp 624-
641.

0' Reilly, Charles A., Caldwell, David F., & Barnett, William P. (1989). Work
Group Demography, Social Integration, and Turnover. Administratirg

Scienoefluarttrly Vol 34. PP 21-37

150

Parasuraman, S. (1982). Predicting Turnover Intentions & Turnover Behavior.

A Multivariate Analysis lounalflflocationalmm. Vol 21
pp. 111-121.

Pfeffer, J. (1981). Some Consequences of Organizational Demogiaphrltial
Impacts of an Aging Workforce on Formal Organizations. In S.B.

Kiesler, J .N. Morgan, & V. K. Oppenheimer (Eds.), Mal
Change. pp. 291-329. New York: Academic Press.

Pfeffer, J. (1983). Organizational Demography. In L. L. Cummings & B. M.

Staw (Eds) Researchjnflrganizationalmharior. Vol 5. PP 299-
357. Greenwich. CT: JAI Press.

Pfeffer, J. (1985). Organizational Demography: Implications for Management.
CaliforniaJlmagemmLRexim Vol. 28. pp 67- 81

Pfeffer, J., & 0' Reilly, C. A. III. (1987). Hospital Demography and Turnover
Among Nurses. lndnstriaifiolatinns, Vol. 26,pp. 158-173.

Piper, William E. , Marache, Myriam, Lacroix Renee, Richardsen, Astrid M.,
& Jones, Barry D. (1983). Cohesion as a Basic Bond in Groups.
HumanBolations. V0126. PP- 93-108-

Porras, J. 1., (1991, August). Racism in a Research University: Some
Observation and Comments. Paper Presented at the Iaskfinmjhe

WW. Annual
Meetings of the National Academy of Management Meetings. Miami,

Florida.

Porter, L. W., Steers, R. M. (1973). Organizational, Work, and Personal
Factors 1n Employee Turnover and Absenteeism. W
mum. Vol. 80, pp. 151- 176.

Porter, L. W, Steers, R. M, Mowday, R. T, & Boulian, P. V. (1974).
Organizational Commitment, Job Satisfaction, and Turnover Among

Psychiatric Technicians. WWW, Vol. 59, pp.
603-609.

Powell. G. (1988). flomenandejnManagemont. Newbury Park. CA:
Sage.

151

Premack, S. L., & Wanous, J. P. (1985). A Meta-Analysis of Realistic Job

Preview Experiments. WW, Vol. 70, p. 706-
719.

Price, James L. (1977). W. Ames, 10: Iowa State

University Press.

Rabinowitz, S, & Hall, D. T. (1977). Organizational Research on Job
Involvement. WW. Vol. 84. pp. 265-288.

Reynolds, Larry. (1992). Companies Will Work Together on Workforce
Diversity. HRM. December.

Riger, S. & Galligan, P. (1980). Women 1n Management: An Exploration of
Competing Paradigms. Americanfisyohoingist. Vol. 35, pp. 902-910.

Rhodes, S. R., & Steers, R. M. (1990). WWW“.
Reading: Addison-Wesley.

Roberts, Karlene H., & O'Reilly, Charles A. III. (1979). Some Correlates of

Communication Roles in Organizations. Aoadgnnmfluanagemgnt
.Innrnal, Vol. 22, pp. 45-57.

Rosenbaum, M. (1986). The Repulsion Hypothesis: On the Non-Development

of Relationships. lonrnaloLBrrsonalinLandSocialPslchology Vol
51, pp. 1156-1166.

Ruhe, J. , & Batman, J. (1977). Effects of Racial Composition on Small Work
Groups. WM, Vol. 8, pp. 479-486.

Runyon, K. E., (1973). Some interactions Between Personality Variables and

Management Styles. WW, Vol. 57, pp.
288-294.

Rynes S., Rosen, B. (1994). What makes Diversity Programs Work.
Magazine. Vol. 39, No. 10. pp. 67-73.

Schneider, B. (1983). An Interactionist Perspective on Organizational
Effectiveness. In L. L. Cummings & B. M. Staw (Eds). Win

QrganizationaLBrharinr, Vol. 5, pp. 1-.31 Greenwich. CT. JAI Press.

152

Schneider, B. (1987). The People Make the Place. hrsonnolmohoiogy. Vol.
40. pp. 437- 453.

Schwab, D. P. (1980). Construct Validity in Organizational Behavior. In B. M.

Staw, &L L. Cummings(Eds..) WW
Marion Vol. 2, pp. 3-43, Greenwich, CT. JAI Press.

Schwartz, Felice (1989). Management Women and the New Facts of Life.
WWW, January/February, pp. 65-76.

ShaW. M. B. (1981).Grono_Dynamics..IheBs1choloaLof.SmalLGmuP
Bohaxior. (3rd ed.) New York: McGraw-Hill.

Shipper, Frances, C., & Shipper, Frank. M. (1987). Beyond EEO: Toward
Pluralism. W. Vol. 30, Iss. 3. pp. 53-61.

Smith, Ken G., Smith, Ken A., Olian, Judy D., Sims, Jr., Henry R, O'Bannon,
Douglas P., & Scully, Judith A. (1994). Top management Team
Demography and Process: The Role of Social Integration and

Communication. WW, Vol. 39, pp. 412-
438.

Society for Human Resource Management/Commerce Clearing House, Inc.

(1993).Managing_nirers1m8umx May 26

Solomon, Charlene M. ( 1995). Affirmative Action: What You Need to Know.
BersonneLJonrnal. August. pp. 56457.

South, Scott J. Bonjean, Charles M, Markham, William T, & Corder, Judy.
(1982). Social Structure and Intergroup Interaction: Men and Women

of the Federal Bureaucracy. AmonmSogologioaifioxim, Vol. 47,
pp. 587- 599.

Staw, B. M. (1984). Organizational Behavior: A Review and Reformulation of

the Field' 3 Outcome Variables. AnnnaLRoziolofiksychology, Vol.
35, pp. 627-666.

Stone, E. F., & Hollenbeck, J. R. (1984). Some Issues With the Use of

Moderated Regression QrganizationaLBoharioLandflnman
Performance, Vol. 34, pp. 195-213.

153

Tang, S. F., & Kirkbride, P. S. (1986). Developing Conflict Management Skills
in Hong Kong: An Analysis of Some Cross-Cultural Implications.

ManagemenLEdncatioLandmmonment. Vol 17 PP 287- 301

Terborg, J. R., & Ilgen, D. R. (1975). A Theoretical approach to Sex
Discrimination in Traditional Masculine Occupations. Organizational
BeharioLandfiumanPerformance Vol 13 PP 351 --376

Thomas, D. A., & Alderfer, C. p., (1989). The Influence of Race on Career
Dynamics: Theoretical Research on Minority Career Experience. In

m. B. Arthur, D. T. Hall, & B. S. Laws (Eds.). W
M. pp. 133-158. New york: Cambridge University Press.

Thomas, E. J., & Fink, C. F. (1963). Effects of Group Size. Psychological
Bulletin. Vol. 60. PP. 371-384.

Thomas J r Roosevelt R. (1990). From Affirmative Action to Affirming

Diversity. KW, Iss. 2, March/April, pp. 107-
119.

 

York: AMACOM.

Thomas, Victor. (1994). The Downside of Diversity. Irainingj;
Wont. Vol.48, No. 1, pp. 60—66.

Tom, V. R., (1971). The Role of Personality and Organizational Images 1n the

Recruiting Process OrganizationaLBeharioLandHnman
Performance, Vol. 6, pp. 573 -592.

Triandis, H. C. (1960). Cognitive Similarity and Communication in a Dyad.
W, V01. 13, 175-183.

Triandis, H. C., Hall, E. R., & Ewen, R. B. (1965). Member Heterogeneity
and Dyadic Creativity. Kumanfiolations, Vol 18, pp. 33-35.

Tsui, A. S., & O'Reilly, C. A. III. (1989). Beyond Simple Demographic
Effects: The Importance of Relational Demography in Superior-

Subordinate Dyads. AcademLoLManagflnentJournal. Vol. 32. PP-
402423.

154

Tsui, A, Eagan, T, 0' Reilly, Charles A. (1991). Being Different: Relational

Demography and Organizational Attachment. Moors.
. M“ U-‘IIJ. M1am1

 

Florida, August, 183 187.

USA Today. December 2. (1987). W.

U S Department of Labor. BureanflfLaboLStansnmEmnloxment.
W Washington
DC, GPO, March, 1991.

Vaid-Raizada, V. K. (1985). Management of Interethnic Conflict in an Indian

Manufacturing Organization. PsychologicaLReports, Vol. 56, pp. 731-
738.

Voydanoff, P. (1988). Work and Family: A Review and Expanded

Coneeralization JonrnaloLSoclaImhazioLandPersonality Vol
3, Iss. 4, pp. 1-22.

Vroom, V. R. (1966). Organizational Choice: A Study of Pre-and Post-

Decision Processes QrganizationalPeharioLandfluman
Performance, Vol. 1, pp. 212- 226.

Wagner, W. Gary, Pfeffer, Jeffrey, & 0' Reilly, Charles A. III. (1984).
Organizational Demography and Turnover 1n Top-Management Groups.

WW, Vol. 29, pp. 74-92.

Wanous. J P (1980) QrgmizationallntuLRecnnnnenLSelectionand
WW. Reading, MA: Addison-Wesley.

Ward, Russell A., La Gory, Mark, & Sherman, Susan R. (1985). Neighborhood
and Network Age Concentration. Does Age Homogeneity Matter?.

SocialPslchologicalQnarterly Vol. 48 PP 138-149

Watson, Warren E., Kumar, Kamlesh, & Michaelson, Larry K. (1993).
Cultural Diversity's Impact on Interaction Process and Performance:
Comparing Homogenous and Diverse Task Groups. _Academy_of

Mmagementlonml. Vol. 36. No. 3. PP- 590-602-

Webb, Eugene J., Campbell, Donald T., Schwartz, Richard D. & Sechrest,
Lee.(1966). W
W Chicago: Rand McNally.

155

Weick. K. E.. (1979). IheSocialPsxchologLofflrganizing. (2nd ed-).
Reading, MA: Prentice-Hall.

Weiss, D, Dawis, R., England, G. & Lofquist, L. (1967). MannaLfor
W, (No. 22)., Minneapolis.
University of Minnesota, Minnesota Studies in Vocational Rehabilitation.

Werner, C. & Parmlee, P. (1979). Similarity of activity Preferences Among
Friends: Those who Play Together Stay Together. SociaLPsxchology
Quarterly. Vol. 42, pp. 62-66.

Wiley, D. L. (1987). The Relationship Between Work/Nonwork Role Conflict
and Job Related Outcomes: Some Unanticipated Findings. ,lnnrnaLof
Management. Vol. 13, pp. 467-472.

Youngblood, Stewart A., & Chambers-Cook, Kimberly. (1984). Child Care

Assistance Can Improve Employee Attitudes and Behavior. Personnel
Administrator. February. PP- 93-95-

Zenger, Todd R. , & Lawrence, Barbara S. (1989). Organizational Demography:
The Differential Effects of Age and Tenure Distributions on Technical

Communication. WWW. Vol. 32, pp.
353-376.

APPENDICES

APPENDIX A

156
APPENDIX A

ATTITUDES TOWARDS DIVERSITY PROGRAMS IN THE BANKING
INDUSTRY

QUESTIONNAIRE
INTRODUCTION

The purpose of this survey is to collect input from employees about the effects
of diversity programs in the banking industry. Diversity programs are a reLof
artirities that support a heterogenous work environment and promote knowledge and
acceptance of cultural, gender, and racial differences that prevent people from
working together and facilitate all employees to reach their full potential by
eliminating subtle barriers that prevent people from being more productive. Your
input and suggestions are very important to the whole process. Please answer all of
the questions. If you wish to comment on any question or qualify your answers, please
feel free to use the space in the margins and at the end of the questionnaire, or use
additional pages. Your comments will be carefully read and taken into account
seriously.

Participation in this survey is entirely minimal. You may Mariam
Wu in the survey at any time. You indicate your voluntary agreement to
participate by completing and returning this questionnaire. All responses to the
questionnaires will be kept strictly confidential. Further, you W
yourself on the questionnaire. No one in the Bank will see the completed surveys.
Only group statistics and aggregate results will be disclosed as feedback to your bank
and all interested employees. Questionnaires will be processed by the researcher
alone. Moreover, questionnaires will be destroyed once the analysis is complete.

Please take the time to complete this survey and return it. It should take
approximately 30 minutes of your time to read and complete the survey. It will be
appreciated if you could respond by August 31,1994. You can eitheLmail the
completed survey in the W provided for this purpose OR returnjhe
completeisnmlmjhemarkedmx, kept 1n a centralized location at your bank.

If you have any questions, please call the researcher Shobha Ramanand at:
(51 7) 355-2909.

157
QUESTIONNAIRE OVERVIEW

The questionnaire that follows is divided into six sections: (1) Job involvement,
(2) Company commitment, (3) Job satisfaction, (4) Knowledge and effectiveness of
diversity programs, (5) Attitudes towards diversity programs, and (6) Personal
background information. The first three sections deal with questions about your
worklife in general. Sections four and five deal with your knowledge and effectiveness
of diversity programs, and your attitudes towards them. The last section focuses on
background questions about you, because these factors are directly related to the
various diversity programs that have been implemented in your organization.

THANK YOU IN ADVANCE FOR YOUR HELP AND PARTICIPATION.
33mm: JOB mvonvmmn'r

INSTRUCTIONS: The following statements are comments people have made or
might make about their work. Please consider each statement and circle
the number which best represents the extent to which you agree or
disagree with the statement. The rating scheme for these statements is
listed below. Do not respond in the way you think others believe you
should behave. YOUR RESPONSES KILL BE KEPT CONFIDENTIAL AND YOUR
IDENTITY KILL_KOI BE REVEALED TO ANYONE WITHIN OR OUTSIDE THE BANK.

1 = Strongly Agree 5 3 Slightly Disagree
2 a Agree 4 = Neither Agree 6 a Disagree
3 = Slightly Agree nor Disagree 7 = Strongly Disagree

1. I'll stay overtime to finish a job, even if I'm not paid for it.
1 2 3 4 5 6 7

2. You can measure a person pretty well by how good a job he or she
does.

1 2 3 4 S 6 7
3. The major satisfaction in my life comes from my job.

1 2 3 4 S 6 7
4. For me, mornings at work really fly by.

1 2 3 4 5 6 7

(AMP

10.

11.

12.

13.

14.

15.

158

- Strongly Agree 5 . Slightly Disagree
. Agree 4 a Neither Agree 6 - Disagree
. Slightly Agree nor Disagree 7 . Strongly Disagree
I usually show up for work a little early, to get things ready.
1 2 3 4 S 6 7
The most important things that happen to me involve my work.
1 2 3 4 5 6 7
Sometimes I lie awake at night thinking ahead to the next day's work.
1 2 3 4 S 6 7
I'm really a perfectionist about my work.
1 2 3 4 5 6 7
I feel depressed when I fail at something connected with my job.
1 2 3 4 5 6 7
I have other activities more important than my work.
1 2 3 4 5 6 7
I live, eat, and breathe my job.
1 2 3 4 5 6 7
I would probably keep working even if I didn't need the money.
1 2 3 4 S 6 7
Quite often I feel like staying home from work instead of coming
in.
l 2 3 4 5 6 7
To me, my work is only a small part of who I am.
1 2 3 4 5 6 7
I am very much involved personally in my work.

1 2 3 4 5 6 7

159

1 = Strongly Agree 5 2 Slightly Disagree
2 = Agree 4 = Neither Agree 6 = Disagree
3 a Slightly Agree nor Disagree 7 = Strongly Disagree

16. I avoid taking on extra duties and responsibilities in my work.
1 2 3 4 5 6 7

17. I used to be more ambitious about my work than I am now.
1 2 3 4 5 6 7

18. Most things in life are more important than work.
1 2 3 4 5 6 7

19. I used to care more about my work, but now other things are more
important to me.

1 2 3 4 S 6 7

20. Sometimes I'd like to kick myself for the mistakes I make in my
work.

1 2 3 4 5 6 7
QIQIIQH_II: COMPANY COMMITMENT

INSTRUCTIONS: Listed below are a number of statements that represent
possible feelings that people might have about the company for which
they work. Think about how you feel about your BANK now. Consider each
statement and circle the number which best represents the extent to
which you agree or disagree with the statement. The rating scheme for
these statements is listed below. Do not respond in the way you think
others believe you should behave. YOUR RESPONSES KILL BE KEPT
CONFIDENTIAL AND YOUR IDENTITY HILL_NQI BE REVEALED TO ANYONE WITHIN OR
OUTSIDE THE BANK.

1 = Strongly Agree 5 = Slightly disagree
2 = Agree 4 = Neither Agree 6 = Disagree
3 = Slightly Agree nor Disagree 7 = Strongly Disagree

1. I am willing to put in a great deal of effort beyond that normally
expected in order to help this bank be successful.

1 2 3 4 5 6 7

2. I talk up this bank to my friends as a great organization to work
for.

160

1 = Strongly Agree 5 = Slightly Disagree
2 = Agree 4 = Neither Agree 6 = Disagree
3 = Slightly Agree nor Disagree 7 = Strongly Disagree
3. I feel very little loyalty to this bank.
1 2 3 4 5 6 7
4. I would accept almost any type of job assignment in order to keep
working for this bank.
1 2 3 4 5 6 7

5. I find that my values and this bank's values are very similar.
1 2 3 4 S 6 7

6. I am proud to tell others that I am part of this bank.
1 2 3 4 5 6 7

7. I could just as well be working for a different bank as long as the
type of work was similar.

1 2 3 4 5 6 7

8. This bank really inspires the very best in me in the way of job

performance.
1 2 3 4 S 6 7
9. It would take very little change in my present circumstances to

cause me to leave this bank.
1 2 3 4 5 6 7

10. I am extremely glad that I chose this bank to work for over others
I was considering at the time I joined.

1 2 3 4 5 6 7

11. There's not too much to be gained by sticking with this bank
indefinitely.

l 2 3 4 5 6 7

12. Often, I find it difficult to agree with this bank's policies on
important matters relating to its employees.

1 2 3 4 5 6 7

161

1 - Strongly Agree 5 = Slightly Disagree
2 a Agree 4 = Neither Agree 6 = Disagree
3 = Slightly Agree nor Disagree 7 = Strongly Disagree

13. I really care about the fate of this bank.
1 2 3 4 5 6 7

14. For me this is the best of all possible companies for which to work.
1 2 3 4 5 6 7

15. Deciding to work for this bank was a definite mistake on my part.
1 2 3 4 S 6 7

16. I never say anything bad about myself.
1 2 3 4 5 6 7

17. I am always honest in everything I do.
1 2 3 4 S 6 7

18. There have been occasions on which I felt jealous of other people.
1 2 3 4 5 6 7

19. I can remember times when I was embarrassed.

l 2 3 4 5 6 7

162

SEQIIQN_III: JOB SATISFACTION

INSTRUCTIONS: The purpose of these statements is to give you a
chance to tell how you feel about your present job, what things
you are satisfied with and what things you are not satisfied
with. Read each statement carefully. Circle the number which
best represents how you feel. Be honest. YOUR RESPONSES KILL BE
KEPT CONFIDENTIAL AND YOUR IDENTITY HILL_NQI BE REVEALED TO
ANYONE WITHIN OR OUTSIDE THE BANK.

obWNH

Extremely Satisfied S = Slightly Dissatisfied
- Satisfied 6 = Dissatisfied

Slightly Satisfied 7 = Extremely Dissatisfied

I can't decide whether I am satisfied

or not

On my present job, this is how I feel about:

1.

Being able to keep busy all the time.

1 2 3 4 5 6 7

. The chance to work alone (independently) on the job.

1 2 3 4 5 6 7

. The chance to do different things from time to time.

1 2 3 4 S 6 7

. The chance to be "somebody" in the community.

1 2 3 4 5 6 7

. The way my immediate supervisor handles employees.

1 2 3 4 5 6 7

. The competence of my supervisor in making decisions.

1 2 3 4 5 6 7

. Being able to do things that don't go against my conscience.

l 2 3 4 5 6 7

. The way my job provides for steady employment.

1 2 3 4 5 6 7

163

1 = Extremely Satisfied 5 = Slightly Dissatisfied
2 a Satisfied 6 = Dissatisfied
3 . Slightly Satisfied 7 = Extremely Dissatisfied
4 = I can't decide whether I am satisfied

or not

9. The chance to do things for other people.

1 2 3 4 S 6 7
10. The chance to tell people what to do.

1 2 3 4 5 6 7
11. The chance to do something that makes use of my abilities.

1 2 3 4 5 6 7
12. The way company policies are put into practice.

1 2 3 4 5 6 7
13. My pay and the amount of work I do.

1 2 3 4 5 6 7
14. The chances for advancement on this job.

1 2 3 4 S 6 7
15. The freedom to use my own judgement.

1 2 3 4 5 6 7
16. The chance to try my own methods of doing the job.

1 2 3 4 S 6 7
17. The working conditions.

1 2 3 4 5 6 7
18. The way co-workers get along with each other.

1 2 3 4 5 6 7

164

1 3 Extremely Satisfied 5 = Slightly Dissatisfied
2 a Satisfied 6 = Dissatisfied
3 a Slightly Satisfied 7 = Extremely Dissatisfied
4 = I can't decide whether I am satisfied

or not

19. The praise I get for doing a good job.
1 2 3 4 5 6 7
20. The feeling of accomplishment I get from the job.

1 2 3 4 5 6 7

165

SIQIIQE_I!: KNOWLEDGE, AND EFFECTIVENESS OE DIVERSITY PROGRAMS AND
PRACTICES.

1. These questions relate to your knowledge of diversity programs and
practices within your bank, and how effective each of these programs
have been. PLEASE CIRCLE EITHER I]fi_QB_flQ TO YOUR KNOWLEDGE WHETHER
EACH OP THESE DIVERSITY PROGRAMS EXISTS IN YOUR BANK. IE YOU HAVE
CIRCLED 133 TO THE QUESTION “DOES THE PROGRAM EXIST“, PLEASE CIRCLE HOW
EFFECTIVE YOU THINK EACH PROGRAM IS. IF YOU CIRCLED 89 TO THE QUESTION

"DOES THE PROGRAN EXIST". PLEASE flEI2_TQ_IEE_NEZT_QHEEIIQN IN THIS
SECTION. Thank You.

DIYEBflIII_EBQQBAM&_AND_BBAQIIQE§ DOC. th. 30' 31fO¢t1VO II Th0

 

Program, Program
Exist
1. Sexual Harassment Policy Yes/No Excellent Good Fair Poor
2. Physical access for employees Yes/No Excellent Good Fair Poor
with disabilities
3. Flexible work schedules Yes/No Excellent Good Fair Poor
4. Allowing greater flexibility Yes/No Excellent Good Fair Poor
with respect to days off for
religious holidays (beyond
legal requirements)
5. Parental leave policy Yes/No Excellent Good Fair Poor
6. Job redesign to permit Yes/No Excellent Good Fair Poor
greater employment
opportunities to
handicapper employees
7. Support for employees who Yes/No Excellent Good Fair Poor
are caregivers to elderly
or ill relatives
8. Interpersonal skills Yes/No Excellent Good Fair Poor
training
9. Subsidized day care benefits Yes/No Excellent Good Fair Poor
10. Cultural awareness training Yes/No Excellent Good Fair Poor

11. Alternative career paths for

12.

13.

14.

15.

16.

17.

18.

2.

employees willing to sacrifice

rapid advancement for more

family time.

and networking

and networking

and networking
single parents

Support groups

assistance for minorities.
Support groups

assistance for women.
Support groups

assistance for

Mentoring programs for women
Mentoring programs for
minorities

Fast track career programs
for women

Fast track career programs
for minorities

Excellent

Overall, How would you rate your Bank'-

Good

166

Does the
Program
Exist

Yes/No

Yes/No

Yes/No

Yes/No

Yes/No

Yes/No

Yes/No

Yes/No

Fair

How Effective

Program

Excellent Good

Excellent Good
Excellent Good
Excellent Good
Excellent Good
Excellent Good
Excellent Good
Excellent Good

diversity programs.

Poor

Is The

Fair

Fair

Fair

Fair

Fair

Fair

Fair

Fair

Poor

Poor

Poor

Poor

Poor

Poor

Poor

Poor

167
EEQIIQH_I: ATTITUDE TOWARDS DIVERSITY PROGRAMS IN THE BANKING INDUSTRY

INSTRUCTIONS: The purpose of these statements is to give you a chance
to tell how you feel about diversity programs in your bank. Read each
statement carefully. Please consider each statement and circle the
number which best represents the extent to which you agree or disagree
with the statement. The rating scheme for these statements is listed
below. Do not respond in the way you think others believe you should
behave. Be honest. YOUR RESPONSES flILL BE KEPT CONFIDENTIAL AND YOUR
IDENTITY EILL_NQI BE REVEALED TO ANYONE WITHIN OR OUTSIDE THE BANK.

1 = Strongly Agree 5 = Slightly Disagree
2 = Agree 4 = Neither Agree 6 a Disagree
3 = Slightly Agree nor Disagree 7 a Strongly Disagree
1. A more diverse workforce will better enable my bank to serve its
clients and customers.
1 2 3 4 S 6 7
2. In the banking industry, diverse organizations will be more
successful and innovative.
1 2 3 4 S 6 7
3. Valuing diversity is just another word for affirmative action.
1 2 3 4 S 6 7
4. Diversity programs are necessary to serve the needs of diverse
employees.
1 2 3 4 S 6 7
5. Diversity programs tend to pit employees against each other.
1 2 3 4 5 6 7
6. Diversity programs help to attract and retain valuable employees.
1 2 3 4 5 6 7
7. The special treatment of some employees only serves to increase

negative attitudes about these employees.

1 2 3 4 5 6 7

N

10.

11.

12.

13.

14.

15.

16.

17.

Strongly Agree
Agree
Slightly Agree

168

S a Slightly Disagree
4 = Neither Agree 6 . Disagree
nor Disagree 7 = Strongly Disagree

Diversity programs help to reduce racism and sexism in the

workplace.

1 2

3

4 5 6 7

Diversity programs help to facilitate work with people who are

different

from each other.

1 2

4 5 6 7

The existence of a diverse workforce makes it more difficult to
uphold performance standards.

1 2

3

4 5 6 7

Diversity programs help employees to develop their own special
talents and abilities.

1 2

Diversity programs

1 2

Diversity programs

1 2

Diversity programs
career development.

1 2

Overall, the costs
potential benefits.

1 2

3

3

3

3

4 5 6 7
are just a passing fad.

4 5 6 7
discriminate against some employees.

4 5 6 7

contribute to an employee's personal growth and

4 5 6 7

of a more diverse workforce outweigh the

4 5 6 7

Diversity programs help to distribute organizational rewards and
resources more fairly and equitably.

1 2

3

4 S 6 7

Diversity programs help to reduce conflict in the workplace.

1 2

3

4 5 6 7

169

SIQIIQH_!I: BACKGROUND INFORMATION

Please respond to the following questions. Ihnx_3;g_ng;_in;gndgfl_§g
r1I1al_x9nr_idantiIxI__Ihax_ara_nnrnlx_hnins_nakad_t9_haln_in_tha

analxliz_gf_thn_snnnt19nnl YOUR RESPONSES HILL BE KEPT CONFIDENTIAL AND
roan mm”! mm as ummm: 'ro wrmm on 001er:. m am.

1. What is your sex? Please circle. Female Male

2. How old were you (years) on your last birthday.

18 years -25 years____ 40 years -45 years____
25 years -30 years____ 45 years -50 years_____
30 years -35 Years____ 50 years -55 years____
35 Years -40 Years____ 55 years and above____

3. What is your race? Please circle the appropriate response.

Black/African American Caucasian Hispanic Asian American
Indian

Other

(Please State)

 

4. What is the highest level of education that you have completed.
Please circle the appropriate response.

1. Completed high school. 3. Completed a college program.
2. Some college education. 4. Have been to graduate school.
5. Are you presently married? Please circle. Yes No.

6. Please state number of children living at home.

 

7. Are you the primary wage earner in your family? Please circle.
Yes No

8. Are you a part-time employee? Please circle. Yes No.

9. How long have you been employed in this company? Years

 

10. Which department/or branch are you currently
working?

 

170

11. How long have you been working in this department/or
branch? Years

 

12. What is your job title?

 

ADDITIONAL comm-rs -----------------------------------------------------

 

THANK YOU FOR YOUR PARTICIPATIONIIIIIIIII

APPENDIX B

171
APPENDIX B

An Example Showing the Computation
of a Single Individual Dissimilarity Variable

tau! 1' .W.H-.Il| .n it'll .- t I U'Il n.’ .H'

In this equation, 'n' is the number of group members, 's' is the individual's value on
the attribute, and sj is the jth member's '3’ value on the attribute. In other words, this
measures the square root of the mean squared distance for each individual attribute,
such as age, tenure, or gender etc. , of individual 'i' from all other group members.

For example there are three individuals in one group. They have a single score each on
one variable.

I l"! l-' S

1 5
2 7
3 8

Individual 1, Individual Dissimilarity = (5-7)2/ 2 + (5-8)2/2
=2+45
= 1.4 + 2.1 = 3.5
Individual 2, Individual Dissimilarity = (7-5)2/2 + (7-8)2/2
= 1.4 + .7 = 2.1
Individual 3, Individual Dissimilarity = (8-5)2/2+ (8-7)2/2

=2.1+.7=2.8

APPENDIX C

172
APPENDIX C

WWWM

Heterogeneity = (1-Zpiz)

For e.g., a group has 50% male and 50% female.

pm =.5 pm2 =.25
pf =.5 pf =.25
sz =.5

Heterogeneity = 1-.5 = .5

For e.g., a group has 90% male and 10% female.

pm =.9 pm2 =.81
pf =.1pf2 :01
2132 =.82

Heterogeneity = 1-.82 =.18

"lllllllllllllll“