}~INAL
THE IMPACT OF RECRUITMENT, SELECTION,
PROMOTION AND COMPENSATION POL.ICIES AND PRACTICES
ON THE
GLASS CEILING
submitted
to
u.s. Department of Labor
Glass ceiling Commission
prepared -.
by
Roosevelt Thomas, Ph.D.
Jeff Porterfield, M.A.
John Hutcheson, Ph.D.
Carol Pierannunzi, Ph.D.
of
The American Institute for Managing Diversity
Atlanta, Georgia
April 11, 1994
HF
55Ltq
.5
TABLE OF CONTENTS 1'\1)
TY-Ut .
EXECUTIVE S~Y 1
INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~. . . . . . . . .1
METHODOLOGY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
FORMA.L SYSTEMS 9
Recruitment and Selection 10
Social Structural Dimension 12
Selective Recruitment 12
Selective Personal Traits 15
Selective External Information 17
Social PsychologicalDimension : 19
Bias Interviewing 20
Bias Testing .22
Training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Per:ormance Appraisal 25
Rater-Ratee Race Effects 25
Rater-Ratee Gender Effects 27
Promotion/Succession Planning 28
Race and Promotion .28
Gender and Promotion 29
Compensation Practices/Policies 31
Minority-Nonminority Earning Differences 3c
Male-Female Earning Differences 33
Turnover. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Continuity in Female Employment 34
INFORMAL SYSTEMS 37
Evidence of Informal Barriers 38
Norms in Organizations 40
Mentoring 45
Group Interaction and the Glass Ceiling 52
CONCLUSIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59
Policy Implications 59
Recommendations for Future Research 62
REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
APPENDICES. . . . . . . . . . . . . . . . . . . . . . . . . .Pfdp\!ftflb.F. . . . . . . . . . . . . . . .71
MARTINP. CATHERWOOD liBRARY
NEW YORK STATE SCHOOL
IIDUSTRIAL ANn lABOR RELATIONS
C .:' "niversitv
EXECUTIVE SIlHMARY
The literature concerning policies and practices
i.pact whichrecruitment, selection, promotion, and
SQbsequently compensationthe glass - andceiling - has policy
implications for both
business/corporate
organizations and government.
this As discussed inmonograph, certain business/corporate
procedures POlicies andinherently operate to produce
and/or maintain the
effects of a glass
ceiling in given
organizations.
stem from formal These effects
organizational systems, (such as policies and
practices of recruitment and selection,
performance
promotion/succession appraisal,
planning, compensation,
:informal and turnover),systems to(such as mentoring
and norms in
organizations).
FORMAL SYSTEMS
Recruitment -.and Selection
Social Structural Dimension
:> Selecti~
social Recrui~ent - wi~instructural parameters, "mainstream" orrecruitment "traditional"
avoid efforts to
employers'
recruitment predominately
typically
sources that bring tw
direct
number h
heimtealabor pools and
of-racial/ethnic minorities disproportionate
> and/or women.Selective .Personal
personal Traits - mainstream or traditionaland
salient as attitudinal traits of the candidate are as
process. educational training in the employer's
what is Inherently problematic selection
a dominatceonsidered
to this scenario is that
mainstream or
culture which mayor matyradniottiobnealmaiinsstdreeraimved
racial/ethnic fromminorities and/or women. for
:> Selective External
differences Informationin the type of exter-nanlot only do vast
applicants vary by group, but, also tihneforemfaftoirotn put forth in
gathering required ofexternal information.
i
Social Psychological Dimension
The traditional social psychological parameters of
recruitment and selection foster bias interviewing and bias
testing due to the absence of consideration for social and
personal differences of candidates.
> Bias Interviewing - individuals from differing social groups
seem to lack common experiences and conversation patterns
which typically ease interaction in interpersonal settings.
Verbal and nonverbal cues are often absent and/or misread,
thereby exacerbating misunderstanding. This situation tends
to worsen with the interjection of class differences.
> Bias Test - employment tests, in general, have been shown
not only to be culturally biased, but weak indicators of job
performance and successful employment.
* data show that employers who do not use testingextensively, tend to rely more heavily on ~elective
recruitment and subjective impressions in the job
interview. Although tests introduce bias, subjective
means of screening may disadvantage minority and women
applicants even more. .
Performanc~ Appraisal
> Research indicates that when the rater and ratee are of the
same race, the rater tends to assign higher ratings.
> Although minority employees tend to receive lower overall
ratings than white employees, individual rater bias is not
the only causal factor. Race differences resulting in the
structural differential treatment of minorities within
organizations is also a factor.
* Treatment (or the lack of) experienced by minoritiesprompts fewer and less desirable opportunities to be
held in high esteem.
* Opportunities such as sponsorship, job procedurediscretion, supervisory support and overall acceptance
affect subsequent performance.
ii
I
j
* Non-support
networks exclUdeswhich members from
performing provide valuable crucialtheir job. informalresources
> toIn membersContrast into race
evaluations effects,
raters do not yield
same Sex
tend to be more higher rater-rateeratings.
> lenientFemale than male
However,
ratees, rater.
female
are as opposedevaluated
POsitions. more
to male
POsitively rateesin in the same job,female-oriented jOb
* Long held
particularltyraditionaljOb perCeptionsratings of related
of Sex roles,
women's jOb sex rOles,
> Women's performance.
bias the
evaluation
POsitionsperformancesare inoften typicallydue to no other Subject to beinmgale-oriented jOb
Continually reasons than discounted
level perpetuated their sex. seVerely;
processe.nvirol1JDe"t
because
and of few wom
TehnisinPhetnhoemena
sUbsequent iseValuation/decisi upper-
on-ma;'ing
promotion/succession
The presence Planningof minorities
POsition and women in any typeis low; and eVen lower to management
management nOn-existent
POsitions. in upper-level
Homosocial
surface -.and reproductionreinforce continuesthe to
are less stereotype that
qUalified women andfor minoritiesmanagement
> Racial POsitions.
where minoritiesthey are morewill sUpervise likely to beother promoted> inMinorities racial minorities. settings
order to bemust,display a higher
Considered
> for the sam
leevel of qualifications
Women have managerial in
upward mademObility, SUbstantial
POsitions.
progress
disadvantaged however,
career they
in
still areofragcaendizational
progression. with
* Women and
bidding minorities'for tendency
"managerial promotion
to rely more on formal
deprives
greater momentum" them of theirambition. - sUperior performance and
iii
i -
* Due to the number of management candidates usually
exceeding the supply of top-level managerial promotion
opportunities, superior performance and greater
ambition of women are frequently discounted as
criterion for obtaining promotion.
* The increased success of men in gaining promotion isattributed to their greater use of informal networks.
> The differences of firm type (private vs. public) indicate
that public organizations are more sensitive to political
and legal influences.
* Higher rates of entrance into management positions by
women and minorities are more likely. in public
organizations than private organizations.
Compensation
Although wages/income is only one component of compensation
(profit-sharing, stock participation programs, retirement
benefits, and vacation accumulation consisting of other forms of
compensation), it represents the bulk of the literature
.
concerning compensation differentials relative to race and
gender.
> The traditional approach to the examination of compensation
differences in male-female earnings has been that of "human-
capital" theory. This perspective argues that an
individuals earnings are a function of his/her training and
experience. In other words, "the only relevant productive
attributes of individuals is their cognitive capabilities."
* In recent years, this approach has yielded a
considerable unexplained residual in earning
differences. The residual is typically attributed to
discrimination.
> Neither human capital theory nor the declining
discrimination hypothesis can account for the minority-
nonminority wage gap.
iv
* Given the quantitative and qualitative achievements in
minority formal education, the passage of
antidiscrimination and affirmative action legislation,
the movement of minorities into occupations
characterized by lower turnover rates, and the
reduction of the social acceptability of bigotry, it
would be expected that discrimination has declined.
Subsequently, the probabilities of minority employment
and wage equality would converge. This has not
occurred.
* Explanations of minority-nonminority employment and
wage differential are more suitable in terms of the
instability of the labor market since 1970. This
market sustained employment discrimination. Therefore,
findings concerning a decline in discrimination as well
as policy recommendations focusing solely on the
augmentation of minority human capital be viewed with
skepticism.
> The disproportionate responsibilities maintained by women
outside the organization operate as a penalizing factor to
their earning potential.
* Relative to men, ongoing external constrains on females
employees have an adverse affect on productivity and
earnings. These constrains are inherent in the
individual's commitment to the family (traditional
thought to be primarily that of the women).
.
* The perception that women need autonomy and
flexibility, more so than male employees, may prompt
management to perceive women as having to prematurely
curtail career advancements in favor of this
commitment.
Turnover
Empirical evidence suggest that continuous employment is not
the norm among young women, but it is a growing trend in the
workplace. While employers may perceive women as more likely to
quit than male employees, this assumption has not been confirmed.
v
> In many instances young men and women, who are in the
process of establishing their careers, do quit their jobs in
order to advance or find a better "match" of
employer/employee needs.
> Also, younger cohorts of male and female employees do not
show differences in quitting rates.
* Only one variable, the presence of a newborn child, isshown to have any impact on female turnover.
* This impact, however, is likely to be temporary and
could be overcome by the provision of
maternity/paternity leave and/or child care.
> Women are more attached to the workplace than they were in
the past.
INFORMAL SYSTEMS
Norms in organizations
Informal systems, such as unwritten rules, norms of
behavior, organizational politics, and accepted modes of
operation have proven to be crucial in the incorporation of -.
minority group members into the workforce. "Based on cultural
audits conducted by the American Institute for Managing
Diversity, most employees feel that unwritten rules do exist and
that adherence to such rules is of paramount importance to career
advancement. These invisible barriers are rooted in traditions,
attitudes, stereotypes, and perceptions. More specifically,
norms of behavior characteristic of an individual's cultural
background, but which may be different from the norms valued by
the dominant corporate culture, are often barriers to selection
and/or advancement.
vi
Much of the literature on organizational diversity (norms in
organizations) centers around ethnicity and gender although some
attention has been given to age and functional diversity. There
is much evidence that differences still do act as barriers to
success within organizations.
> Differences ~re a deficiency for employees.
* Issues of acculturation, corporate values, and
organizational expectations are still challenges faced
by employees who are different from the organizational
norm.
* The existence of traditional stereotypes around women
and minority employees prompt the interpretation of the
same behaviors performed by whites as negatively
different.
> Employing several persons who share characteristics which
are dissimilar to that of the dominant organizational
culture, diminishes the effects of visibility, contrast, and
stereotyping.
> Individuals who are trained together are found to share more
common experiences than individuals who are trained in
segregated groups or alone.
~
Mentoring
.
Many believe that at least some of the problems related to
glass ceilings are due to placement within organizations. That
is, women, and minorities may not be "in the right place at the
right time."
> The lack of strategic placement for advancement seems to be
strongly related to the lack of mentors or sponsors at
higher levels within organizations.
* There is strong evidence that employees (the mentor as
well as the mentee) benefit from mentoring
relationships.
* Mentoring programs have been found to alleviate
turnover problems, as well as low morale; with little
to no extra funding.
vii
* Mentors act as a vehicle which provides information on
organizational norms and/or career opportunities,
delegate responsibility on key projects, and serve as
liaisons between the employee and the upper-level.
> Cultural audit data reveal that mentoring relationships are
common within the workplace.
* However, the degree to which mentors are in positions
to assist employees with career development may be
contingent upon who the mentor is; the degree of access
the mentor has to authority ~im/herself; and, whether
the mentor is capable of sponsorship.
* Racial ties were more important than. same gender
relationships when examining the dynamics of mentoring.
* Cross-race as well as cross-gender mentoring
relationships were found to be difficult to initiate
and sometimes violate organizational norms.
> Although mentoring/sponsorship is important throughout the
career of minority employees, minority employees are
frequently perceived as risky for the mentor/sponsor.
* Minority employees might even be required to prove
themselves prior to the development of a mentor/sponsor
relationship.
> Despite the fact that women are often perceived as more ~
nurturing than men, there are serious obstacles to female
mentoring networks.
.
* The differences in career-orientation between younger
and older female employees may be part of the problem.
* A strong mentor/protege relationship between two women
is sometimes threatening to men within the
organization.
* Female employees may not be able to initiate mentoring
relationships due to the fact that there are fewer
women at higher levels with whom to form relationships.
> Cross-gender mentoring presents many problems for both
mentor and protege, such as misconceptions about sexual
advances, office gossip, and other stereotypes.
viii
> Female ~?loyees tend to prefer male managers, perhaps
because ~~le managers are seen as more powerful advocates
for their subordinates.
CONCLUSIONS
POlicy Implications
Conside~ng the pitfalls of current policies and practices
which seem to .a.nfluencethe production and maintenance of a glass
ceiling, numerous policy implications surface at the
organizationa~ level.
> Startinq with the recruitment phase, organizations stand to
benefit greater if the recruitment committee is
representative (racially, culturally, and gender-wise) of
larger SOf.Ciety.
> Selection decisions are more equitable when they are not the
product o~ any particular individual.
* Selections should be committee decisions base on
prewiously established criteria.
* These committees may serve themselves and the
organization better if they consider a wider scope to
advertise job positions.
> There should be active rec~uitment on campuses and within
other institutions where female and minority candidates are
represen~ed in the selection pool.
> Organizations may consider a greater use of formalized
internship programs, particularly for women and minorities.
> In the imterview and testing phase there is room for policy
regarding the alleviation of biasness.
* Perhaps the use of outside consultation in this process
may facilitate direct assessment of individuals'
attributes. .
> The most salient problematic issue surrounding performance
appraisa2 and promotion is the organization's reliance on
informal decisionmaking networks, rather than formal
processes.
ix
* Appraisal and promotion decisions are best derived at
through inclusion of all formal committee members.
* standards for evaluation and promotion need be written
and clearly understood by all individuals involved
(committee members and candidates) - from date of hire
to present.
* To facilitate fairness in the personal dimensions of
evaluation, managers and promotion committee members
would best serve the organization if a previously
established level of training has been successfully
completed in the area of workforce diversity.
> Turnover is actually a sYmptom of the process which creates
and maintains the glass ceiling and a symptom of the glass
ceiling itself.
* Employers who wish to optimize the work potential of
persons who face a scheduling dilemma might provide
flexibility in work schedules, and at-ho~e work when
possible.
* In some instance the pressures associated with dual
roles may be alleviated by support for family concerns
such as provision of day-care, or broad family leave
policies.
* An regularly scheduled open discussion of issues that
are of concern to employees will provide employers some
insight as to how employee turnover may be minimized
and the needs of organizations and employees might best
be met.
> It is also important that employees are aware of the
personnel policies of the organizations in which they work.
* Written organizational goals for employee diversity
that outline specific programs are far superior to
word-of-mouth systems.
* Organizational leadership should outline in detail
which programs will be in force to provide for
recruitment, promotion, career planning, and mentoring
for all employees.
> The need for training was universally noted by respondents
in the cultural audit data, as well as the subject of study
for some of the literature.
x
* Training that may specifically address the needs of
specific groups, such as assertiveness training for
female managers, might also be beneficial.
> Lateral moves to provide breath of experience for all
employees may help dispel stereotypic roles for females or
minorities. .
> Formalized mentoring programs to assist employees with
organizational norms and cultures may facilitate the
integration of the nontraditional employee.
> Managers who are successful at developing the potential of
the employees should be rewarded.
* Evaluation of managers should include some aspect of
employee career development, and managers and employees
should be aware of this aspect of manager evaluation.
> The efforts of various governmental agencies are more
appropriate in encouraging and funding research to fill this
need.
* The government may even act as a clearinghouse for
research and information regarding the glass ceiling.
> Organizations that have diverse workforces would also be
likely sources for funding research on employment diversity.
These organizations have the most to gain from understanding
diversity and its related dynamics.
~
Recommendations for Future Research
Although research on differences in gender is the most
commonly addressed question in the literature noted here,
continued research on gender differences is needed. In addition,
the literature ~ddresses racial differences mostly as white/black
comparisons, ignoring other racial minorities.
> Future research should include more racially diverse groups
and make comparisons across more than one dimension.
> More specific topics which would seem to demand further
investigation include studies of diversity based on age of
employees, task differentiation (functional diversity), and
disability diversity.
xi
> In the area of compensation, forms other than wages/income
(profit-sharing, stock acquisition, intangible benefits,
etc.) need to be considered. That is the extent to which
these other forms of compensation are available and
accessible to minorities and women.
> Research is devoted to managers or those who have "made it
to the top," with little research available concerning the
"sticky floor" that retains most women and minorities at the
lower levels of responsibility and compensation.
* Research on those who fail to achieve higher levels of
responsibility may be as informative, if not more so,
as what is currently available on th~ glass ceiling.
> Many studies noted herein have small samples, use college
students as subjects, and/or are based on mail surveys.
Case studies of single organizations are sometimes available
but have little applicability to the workforce as a whole.
In few instances are testable hypotheses or even
propositions dr~wn from extant literature. other publi~hed
sources, not cited in this review, are simply the thoughts
of individuals based on personal experiences. The quick
availability of such literature may do more harm than good,
as .personal opinion may become dccepted as fact.
> Field research in this area of study is deficient.
* Field experimentation of testable hypotheses should be
conducted in a variety of organizational settings, over
time and across various dimensions of diversity.
* It is of paramount importance that applied field
research be conducted on larger samples in work
settings.
> Research on the advantages that diverse employees bring to
organizations is sorely needed.
* While there is some research on the presence of
diversity in the workforce, there is little information
available to managers that provides them with
implications of how diversity may benefit their
organizations.
xii
INTRODUCTION
Representation of minorities and women in any type
organizational management position is low (Auster 1988; Hartmann
1987; Davis and Watson 1982; Killingsworth and Reimers 1983;
Zweigenhaft 1987); and even lower to non-existent in upper-level
management positions (Dipboye 1987; Kesner 1988; Brenner,
Tomkiewicz and Schein 1989). Various explanations surface as to
why women and minorities are less likely to obtain and establish
managerial careers: statistical discrimination (Phelps 1972);
exclusion practices (Brass 1985); self-selection (McCarth~ 1986);
tokenism (Fairhurst and Snavely 1983); and differential
socialization (Noe 1988). Homosocial reproduction, or the
continuation of established social patterns (Kanter 1977),
continues to surface and reinforce the stereotype that women and
~
minorities are less qualified for management positions. The
predominance of white males in managerial positions maintain the
"traditional" or "ideal" characteristics considered to be
essential to "good management." Subsequently, employers, other
managers, and even many employees prefer to see white males
holding managerial positions (Dubono 1985; Zweigenhaft 1987;
Brenner, Tomkiewicz, and Schein 1989; Powell and Butterfield
1989). Investigations in the differentiations of firm type
(Cannings 1988; Lewis 1986) - private vs. public - indicate that
public organizations are more sensitive to political and legal
influences. Civil rights organizations, feminists organizations,
1
'"
affirmative action laws and other equal opportunity efforts make
up the bulk of such influences. Therefore, higher rates of
entrance into management positions by women and minority are more
likely in public organizations than private organizations.
There is also a widely held notion that women and minorities
are not as likely as their white, male counterpart to consciously
pursue managerial careers. There is no available research to
support these claims. However, some research (Shenhav 1991;
Bailyn 1987), indicates that women are just as likely as men to
pursue management positions.
Initiatives, such as Affirmative Action, which were/are
intended to open organizations to nontraditional employees, have
been shown to have some impact on the numbers and percentages of
women and racial minorities who are now part of the American
workforce. The most obvious gains are those made by women.
-.
Currently women constitute half of the workforce with an increase
of more that 40 percent in manag~rial positions from the 1970s to
the 1990s (Morrison, White and Van Velsor 1992:5). Despite these
optimistic statistics, being culturally different in an
organization is still a barrier to promotion. The 'glass
ceiling' or inv.isible barriers to upward mobility is seen as well
as experienced by many who do not fit traditional roles. Persons
who exhibit gender, racial, disabtlity, age, organizational
tenure, or position title differences may experience the glass
ceiling. For example, at the highest levels of corporate power
women are few, and women of color are indeed rare. Gender
2
differences are not the only variables subject to the effects of
the glass ceiling. Some estimates note that fewer than 1 percent
of the nations top managers are minority (Morrison, White, and
Van Velsor 1992:6).
The idea of a glass ceiling refers to a supplementary
constraint to the normal 'weeding' process by which individuals
are selected for promotion and increased responsibility.
Naturally there are fewer people at the uppermost levels of the
organization and selection process at higher levels eliminates
some employees. Thus the organization can be pictured as a
pyramid, narrowing at the top, holding places for only the "most.
qualified" of employees. Individuals who are different are
l~kely to face the same pressures of the inclining walls as
everyone else, but they also must contend with other, sometimes
competing, pressures: to conform; to retain their individuality;
.
to socialize with those who share interests with them; and to
deal with those who hold organi~ational power and who are
culturally different from themselves. These are systemic
pressures that apply only to specific groups within
organizations, groups which are different in some respect from
the organization as a whole.
other scholars note that there is evidence that some
individuals are systematically held down at the lowest levels of
organizations. This represents a 'sticky floor' that retains
people at low levels of pay and responsibility (Bureau of
National Affairs 1992). The sticky floor holds individuals at
3
the lower levels of employment due to job stereotypes, cultural
forces outside the organization (such as child or elderly care
being the responsibility of women), or due to the lack of career
or promotional training and opportunities.
As Figure 1 illustrates, the glass ceiling works as a
supplement to the sticky floor and the narrowing opportunities
for employment at the top, systematically to disadvantage
employees because they are nontraditional members of the
workforce, or are seeking employment above those levels that have
traditionally been reserved for them.
Exactly how much we know about this phenomenon is debatable.
Anecdotal arguments can be found, but systematic analysis is
limited. Literature on organizational diversity is available but
tends toward a narrow set of hypotheses. In general, the
literature that is synthesized in this report addresses race a.nd
gender differences. In most instances studies pertaining to
race have examined black-white d~fferences along employment
dimensions. studies examining diversity of age, function,
physical ability, religion, or culture are notably absent.
Indeed even the most popular topics for review are themselves
limited.
Cox and Nkomo (1990) find that the number of articles
dealing with race as a variable within organizational behavioral
research has actually declined since the 1970s. These authors
postulate reasons for the decline. Decisions on tenure, and
dissertation topics are often reviewed by systems which may
4
,I
FIGURE 1
THE GLASS CEILING
.If")
REGULAR EMPLOYMENT REGULAR EMPLOYMENT
PRESSURES PRESSURES
THE STICKY FLOOR
~
discount research in these areas. Journal editors may also feel
that the scope of such research is too limited or even
controversial, thereby blocking publication of such research and
limiting its continuity and impact.
Most articles in the literature deal with diversity as a
psychological concept, and psychology journals are most likely to
publish such studies. However, even among published studies, the
articles are merely comparisons of white-black performance
ratings, test score differences and organizational behavior.
Literature focuses on the higher pay levels within organizations,
failing to'address the needs or questions dealing-~lith why people
achieve or fail to achieve this level of success. Few articles
test hypotheses or even postulate them. In short, even in the
most researched areas of organizational diversity, literature is
sparse and aging. Questions are limited. Topics are non-
.
exploratory and repetitive. Quantitative analysis is rare.
This literature analysis describes the current status of
research on the recruitment, selection, promotion, training,
career development, performance appraisal, and organizational
culture of diverse groups and individuals in the workforce. In
addition to the literature, cultural audit results conducted by
The American.Institute for Kanaging Diversity (AIMD) at Morehouse
College in Atlanta, Georgia are integrated throughout the
monograph. Since 1988, AIMD has conducted cultural audits in a
number of organizations. For a more complete description of
these organizations, see Appendix A. These organizational
6
culture audits assess the degree to which individual employees
perceive that they are contributing members of the corporate
environment in which they work. Since this is intended only as
supplementary to the literature presented, comprehensive
examinations of the audits are not provided in this monograph.
Instead, data are presented that represent responses to questions
asked of employees in a variety of settings. In some cases the
type of organization (public, private, or nonprofit) is
presented, while in others, race and gender categories are
compared. These audits are used to provide empirical support
for the-literature. Telephone interviews with bcholars who have
contributed to the available literature on the glass ceiling
provide insight and are incorporated into the literature
analysis.
-.
METHODOLOGY
This research is organized around specific organizational
processes and phenomena: recruitment of new employees, selection
of employees for hiring, performance appraisal of employees,
promotion and succession planning within organizations,
compensation practices, employee turnover, mentoring within
organizations, organizational norms, group interaction and
cohesion, and corporate culture toward diversity. In many cases
articles that were reviewed cross-cut these topics and some
authors have produced multiple publications on the same general
7
topic area. Readers will note that the same names appear in many
sections of the monograph.
As the literature search progressed, specific questions
developed that are addressed. They are:
0 What are the organizational policies and practices
which either facilitate or inhibit the advancement of
diverse groups to management and decision-making
positions?
0 Under what conditions do specific groups advance or
fail to advance?
0 Is there evidence that diversity is important in-~he
advancement of employees to management and
decisionmaking positions?
0 What are the research and policy implications of the
literature analysis?
.
From these questions, an outline was drafted to guide the
synthesis of the literature. T~is outline is provided in
Appendix B.
In order to uncover as much extant literature as possible,
computer assisted searches were made over a variety of databases
using a variety' of keywords. Once scholars were identified,
searches based on the names of prominent researchers were also
conducted. Databases used for this research include Sociofile,
ABI INFORM, Lexis-Nexis, Psych Info, Infotrac, Trade and
Industry, and Manage, as well as the computerized files of
libraries at several colleges. Lengthy printouts from these
8
sources were culled to provide the literature that addresses
pertinent questions. From this list, articles were copied and
read and placed in an order corresponding to the outline.
As this synthesis is presented below, literature is explored
around two major themes: Formal systems which inhibit or
facilitate the advancement of those who are different in an
organization, and; Informal systems which may also impact on the
advancement and success of diverse groups within organization.
The section on Formal systems is subdivided to examine literature
that deals with recruitment, selection, performance appraisal,
promotion and succession planning, compensation practices and
policies, and turnover. Informal systems literature is also
subdivided, and its sections are focused on mentoring and
sponsorship, norms and unwritten rules, group cohesion, and
corporate culture. Findings from expert interviews are then .
examined, especially as they relate to the need for research in
areas where scholars find it def~cient, as well as implications
for a research agenda based on the literature synthesis.
FORMAL SYSTEMS
Formal systems are procedures that have been
institutionalized in organizations to deal with employee
relations. Such systems would include written regulations on
recruitment, hiring, promotion, compensation, appraisal, and the
like.
9
RECRUITMENT AND SELECTION
The existing literature concerning recruitment and selection
policies/practices within the U.S. labor market reflect analyses
which operate on two levels - social structural (macro) and
social psychological (micro); see Figure 2. Social structural
recruitment and selection policies/practices characteristic of
O.S. employers primarily consist of selective recruitment,
selective personal traits, and selective external information.
That is, recent investigations (Neckerman and Kirschenman 1991;
Cross et al. 1990; Culp and Dunson 1986; Bielby and Baron 198~;
Aigerner and Cain 1977; Phelps 1972; and Thurow 1975) indicate
that employers typically direct recruitment efforts ~o
predominately white labor pools (populations) and avoid
~ecruitment sources that bring them a disproportionate number of
.
racial/ethnic minorities and/or women. specific personal traits,
such as attitudinal traits, app~ar to be just as important as
educational training in the hiring decisions of many employers
iBraddock and McPartland 1987; Crain 1984; McPartland, Dawkins
.and Braddock 1986; and Rossi et al 1974). Finally, evidence
(Bishop 1986; Hollenbeck 1984; McPartland, Braddock and Dawkins
1986; Friedman and Williams 1982; Tenopyr 1981; Berg 1981; and
Kalleberg and Sorensen 1979) suggest vast differences in the
types of external information required/used of varying applicant
groups. Group characteristics influence not only the type of
10
FIGURE 2
DIMENSIONS OF LITERA TURE ON
RECRUITMENT AND SELECTION
MACRO MICRO
SOCIOLOGICAL. PSYCHOLOGICAL
RECRUITMENT SELECTION RECRUITMENT SELECTION
--<<
~
information used in the hiring decision process, but, also the
effort put forth in gathering desired information.
social structural Dimension
Selective Recruitment
An individual's chances of becoming a part of a given
candidate pool followed by employment is surely greater if he/she
is aware of job vacancies. According to Neckerman and
Kirschenman (1991) the recruitment norm for employers seems to be
a strategic selective search rather than the casting of a wide
net. Employers explain their re~ruitment strategies in terms of
practicality. Explanations range from the ease and low cost of
using personal networks to the difficulty of screening the large
numbers of applications yielded by media ads. In essence,
employers justify selective recruitment with the argument that
.
this is the most efficient way to zero in on "the best"
candidates.
A recent national survey (Braddock and McPartland 1987) of
employers (4078) reveals that the type of position strongly
influences the recruitment method. Not advertising job openings
in newspapers is one method of screening applicants. Recruiting
applicants based on the quality or location of schools also
allows employers a unique method of screening. state employment
services are perceived by employers as disproportionately
referring inappropriate, unqualified, minority candidates
(Coverdill 1990). Therefore, these services are seldom utilized.
12
Informal recruitment methods, such as small social networks and
employee referrals, are by far the most frequently used methods
for all type positions. Employers are not likely to assume a
great deal of expense for recruitment of lower-level positions.
Therefore, the most inexpensive methods are typically reserved
for lower-level positions. other recruitment methods - placing
ads in various media - are used less frequently especially when
recruitment is for higher-level positions.
Although employers' recruitment methods vary much more for
upper-level positions, the informal methods remain a major source
for cOllege-educated job ~andidates. Studies (Baker et al. 1984;
Granovetter 1974,1982; Mangum 1982) indicate that employers often
spend time and money to seek candidates from placement services,
media ads, professional organizations, and private employment
services for upper-level positions.
.
Further support for this claim is suggested in a pilot study
("A Report On The Glass Ceiling.Initiative," 1991) of nine
Fortune 500 organizations; conducted by the u.s. Department of
Labor with the intent of identifying systematic barriers to the
career advancement of minorities and women. It is noted in the
identification of attitudinal and organizational barriers that
the progress of minorities and women is affected by more than
qualifications and career choices. Internal monitoring and/or
the planning for developmental programs and policies for equal
access and opportunity, especially at the senior management
levels, is almost never considered the responsibility of the
13
organization. In addition, recruitment practices primarily
consisted of word-of-mouth and employee referral networking.
Such practices promote the filling of vacancies almost
exclusively from within. If the environment is already
homogeneous, which many are, it maintains this same "home-grown"
environment.
Executive search/referral firms are also utilized,
particularly for upper level management positions. They seem to
only add to the practice of exclusive consideration within non-
diverse candidate pools. The majority of the search/referral
firms not only failed in~he acquisition of a diverse candidate
-
pool, but in many instances were not even aware of the equal
employment and affirmative action obligations under the law.
This was evident in the fact that none had made any effort to
reach out to agencies and/or professional organizations (i.e.,
-.
National Black MBA Association, Hispanic MBAs, Who's Who Among
American Women, etc.) rich in qualified minorities and women.
Informal referrals from current employees is still the most
salient method of recruitment. Thus the use of informal social
networks is the primary method which employers use to recruit
outside individuals for job vacancies. The consequences of
informal recruitment policies/practices have proven to be severe
for minorities and women. Scientists (McCall 1972; Rossi, Berk
and Eidson 1974; Becker 1980; Braddock, Crain and McPartland
1984; Lin 1982) have long argued that minorities and women are
denied equal access to valuable informal sources of job
14
information. Minority and women job seekers typically hold
primary ties to social networks composed of other minorities and
women, who generally are not as well situated to know about
employment opportunities as the members of social networks used
by dominant group members. Consequently, an exclusionary barrier
("social network segregation") has been characteristic of the
recruitment phase.
Selective Personal Traits
Employers have a mental, if not written, list of personal
traits desired in a potential employee. The underlying conse~sus
of this practice is two-fold: (1) to select individuals with
similar .demographic backgrounds, physical features, attitudes,
values, and beliefs which creates a relatively homogeneous
environment thereby reducing the level of team conflict (Jackson,
-.
et al 1991); and (2) to screen candidates' personality and
potential behavior at the onset means that less effort is needed
in training, socialization, or monitoring them once they are in
the organization (Cohen and Pfeffer 1986). The assumptions that
personal differences are associated with cognitive differences is
widely accepted among employers (Dearborn and Simon 1958; Bass
1981; and Walsh 1988).
It is also noted (Braddock and McPartlend 1987; Committee on
Economic Development 1985; Hamilton and Roessner 1972) that
ranking among the top are those traits associated with attitude.
For instance, employers report dependability, punctuality,
15
positive attitude about work, self and authority, and the ability
to work well with others as the most important attitudinal traits
- especially when recruiting for lower to middle-level entry
positions. On the other hand, attitudinal traits are top
priority when seeking candidates for upper-level positions as
well. However, other factors come into play. According to Walsh
(1988) a high demand is placed on advanced levels of language,
computation skills, specialized knowledge, the ability to learn
quickly, and the ability to think on the spot in complex
situations. Further, formal education, sound judgement and
leadership qualities are highly valued.
On the surface these all seem like reasonable expectations
of employers. However, Hartigan and Wigdor (1989) indicate that
the average employer perceives various racial/ethnic groups and
women as typically lacking in these priority job traits.
.
Minorities and women candidates, even those with the same
educational levels of their white counterpart, are considered to
be a higher employment risk. The general consensus is that their
attitude about work and previous training in specific skills
leaves a lot to be desired.
According.to Anrig (1987) the use of negative group images,
rather than direct assessments of individuals, contributes to a
crucial exclusionary barrier. This barrier can take effect
whether the employers' perceptions are based on actual group
differences or on entirely uninformed group stereotypes. When
information about individual differences is absent, frequently
16
group identifiers such as sex and race are used to determine the
chances of selecting the "best" employee. Usually the use of
race and/or sex identifiers leads to the selection of a dominant
group member.
Selective External Information
The level of the job position influences both the type of
information used and the effort that employers may put forth in
gathering information. Employers of middle and upper-level
positions that require a college degree (or some college) also
solicit specialized knowledge characteristic of specific
disciplines. Screening information consist of the type and
reputation of the applicants' college program, the applicants'
grades, and recommendations from college officials. Even more
important are references from previous employers (Hambrick 19~1).
Hambrick further notes that when selecting individuals for lower-
level positions, employers rare1y use detailed, in-depth, or
specific information. The final screening process is often brief
and superficial. Only two sources of information are highly
valued and frequently used for lower-level positions: (1)
impressions gained from the job application/resume or during the
personal interview with the candidate, and (2) recommendations
from previous employers. As mentioned earlier, both the
emploYment application and the personal interview are extremely
important in the selection process for lower-level positions.
17
Jolly, Reynolds and Slocum (1988) state that when outside
information is used in the selection process, another
exclusionary barrier potentially forms. This barrier occurs when
employers select candidates by using specific information that
minorities and women may not be able to provide with the same
frequency or credibility. Women and minorities are concentrated
in segregated institutions of society, thereby attaching a stigma
to the information employers most frequently use to evaluate
applicants.
Lorsch (1985) argues that minorities and women are
especially disadvantaged when employers heav~ly weight a
candidate's previous employment experiences or references from
school officials. Due to the higher unemployment rates in
minority communities, minorities are less able to list consistent
work experience on job applications or describe previous
.
significant jobs during the employment interview. Additionally,
the lack of consideration of volunteer work frequently discounts
the candidates' qualifications. This is particularly true for
women who may have opted for nearby social and community
involvement due to their assuming a greater proportion of family
responsibilities.
In addition, information bias occurs when the
recommendations/references provided by minority applicants carry
less weight with employees than those provided by white
candidates. Obviously, minorities and women receive their
recommendations/references from officials of the segregated
18
institutions of which they are a part. These institutions may be
viewed in a lesser light than more mainstream institutions. In
other words, predominately white employers may be less familiar
with a predominately black school, a black church, or a black
firm that an individual may use for sponsorship of his/her
candidacy. White employers may feel more suspect of information
provided by minorities/women due to stigma or stereotypes
attached to minority sources.
Companies that wish to improve their track record on hiring
and recruiting minorities must provide education and training for
m~nagers and supervisors. Career developlnent ~ourses, networking
programs and the like can open the door for more diverse members
of the workforce. In addition, companies may wish to re~ruit
more vigorously by targeting schools where minorities are more
likely to be among potential candidates, or by hiring a
-.
minorities, women, or individuals with disabilities as
recruiters. One promising tactic is to offer internships to
students from such institutions in order for them to understand
more fully the corporate culture they may be entering, and to
allow other employees to interact with them prior to full-time
employment (see Sadano and Baler 1983:33; Rendero 1980).
social Psychological Dimension
The social psychological dimension of recruitment and
selection emphasizes the prevalence of bias interviewing and bias
testing within the interactive processes among individuals and
19
small groups of differing personal and social characteristics.
Research (Neckerman and Kirschenman 1991) shows that prejudice
and cultural misunderstandings create difficulties for minorities
and women - especially those from lower income levels - who
interview with predominately white employers. Although levels of
classic racism and sexism have declined over the years, race and
gender relations between strangers remain filled with fear,
suspicion, and moral contempt (Blauner 1989; Anderson 1990).
Individuals from differing social groups seem to lack common
experiences and conversation patterns which typically ease
interaction in interpersonal settings (Erickson 1975).' In
addition, verbal and nonverbal cues (Parsons and Liden 1984;
Hollenbeck 1984) are often absent and/or misread, thereby
exacerbating misunderstandings (Kochman 1983; Arvey 1979). This
situation tends to worsen with the interjection of class
.
differences.
Bias Interviewing
The interview is a crucial component of the hiring process.
Virtually, all employers use some form of an interview before
hiring. The pr~mary intent is to assess the candidates personal
qualities. Research (Arvey and Campion 1982; Parsons and Liden
1984; Turner, Austin, Fix and struyk 1991) indicates that the
essence of the typical interview centers around the candidates'
past work experiences; indicators of dependability and
willingness to work; and any possibility of falsification.
20
Employers often developed their own subjective "tests" of
productivity and character. For example, attention is placed on
how expressive or open a candidate appears. What is the
candidate's personal philosophy about work - his/her work ethic?
What is his/her view of life? Personal appearance is also very
important. Sitting up straight, talking expressively,
intelligently, and being articulate are positive traits. Also,
many employers mention that when a candidate comes straight out
with an answer ("you don't have to drag every word out of them")
that this is perceived as a good sign.
This IDay all have particular significance for'minority and
women candidates. Martocchio and Whitener (1992) suggest that
further .complicating the interview is the employers' general
distrust of candidates, particularly minority candidates. A
common percepti~n of employers is that candidates frequently lie
-.
about their work experience and job skills. This is mentioned
more often by employers who interview a considerable number of
minority candidates. other research (Arvey 1979; Arvey and Faley
1988; Schmitt and Noe 1986; and Hartigan and Wigdor 1989) shows
that employers complain about minority and women dressing in
shabby or inappropriate clothing or showing up late to
interviews. Employers recognize that there are definite cultural
differences operating in these interviews. Nevertheless, they
consider these differences not appropriate for the work
environment.
21
From the literature, it is obvious that job interviews are
biased in favor of individuals whose behavior and appearances
reflect a conformity to society's dominant cultural norms. In
other words, problematic interviewing goes well beyond
interpersonal skills. The root of the problem lies in differing
cultures, a lack of common understandings of "appropriate"
interaction, and conversation style (Ford, Kraiger, and
Schechtman 1986). Neckerman and Kirschenman (1991) contend that
job candidates, as well as the employer, must be sensitive to
verbal and nonverbal cues and to the hidden agenda of each
question and response. ~ue to the likelihood of minorities and
women being perceived in a negative light before they even get a
'chance to respond, they are likely to be at a disadvantage in the
interview.
-.
Bias Testing
Race and gender bias in hi~ing which results from employment
testing have been longstanding concerns (Braddock and McPartland
1987; Cohen and Pfeffer 1986; Hamilton and Roessner 1972).
Employment tests, in general, have been shown not only to be
culturally biased, but weak indicators of job performance and
successful employment (Burstein and pitchford 1990). Research
reveals that the correlation between test scores and job
performance ratings is even lower for minorities and women than
for non-minority male employees (Hartigan and Wigdor 1989).
Nevertheless, employment testing has prevailed for decades as a
22
salient measure in the hiring process. On the other hand, data
show that employers who do not use testing extensively, tend to
rely more heavily on selective recruitment and subjective
impressions in the job interview. Although tests introduce bias,
subjective means of screening may disadvantage minority and women
applicants even more - providing the candidate gets that far in .
the hiring process (Holzer 1987).
Many employers use formal skills and/or aptitude tests to
screen candidates. Nathan and Alexander (1988) suggest that this
steams from an overall distrust of the public school system and
the quality of the aVailable workforce. The use of skills
testing is much more common among clerical employers. However,
more than half of all white-collar employers use conventional
tests. Blue-collar employers also administer tests to potential
employees, but most often they are informal. In some cases
.
candidates are asked to work for several trial days.
When employers have objective means of getting information
about applicants, such as through various tests, it is expected
that less weight is placed on subjective, biased hiring
strategies. Evidence (Nathan and Alexander 1988; Hoffman, Nathan
and Holden 1991; Vance, MacCallum, Coovert, and Hedge 1988; and
Sackett, Zedeck, and Fogli 1988) shows that employers who use
tests to screen job applicants, formally or informally, tend to
employ a higher proportion of minorities and women than those
employers who do not use tests. It is noted that these findings
must be reported cautiously due to alternative explanations.
23
That is, it may be that these employers test for skills because
they attract more minority and women applicants. Or, it may also
be that their hiring criteria differ from those employers who do
not use tests. Nevertheless, the numbers hold up under several
examinations and analyses.
Training
Training is generally provided for new employees as they
settle into their new work environment. Problems may occur in
training when communication between trainers and trainees is
characterized by contradiction to the norms of the organization
or the cultural background of the trainee. Henteges, Yaney, and
Shields .(1990) propose approaches to multi-ethnic training
sessions in which goals are to increase shared experiences and
knowledge and unify values (42).
.
Training specific to the needs of individual employees or
employee groups is also recommended. Studies on the self-
awareness of managers with like career aspirations have found
that women often have lower job satisfaction levels, and that
women may benefit from assertiveness training in order to be more
effective managers (see Berryman-Fink 1985: Goh 1991:702). There
is general agreement in the literature that assertiveness is
important for managerial success and that women tend to be less
assertive than their male counterparts. The gender of the
interviewer (manager) is also of importance.
24
In some instances responses to the AIMD cultural audits
recommend human relations training for managers. It appears
clear that the expected increased diversity of the workforce will
require some accommodation by managers. It should be noted that
training to understand ethnicity and diversity is not value
driven. That is, it is not a goal in-and-of-itself, but is a
component of good management of a diverse workforce.
PERFORMANCE APPRAISAL
Rater-Ratee Race Effects
The literature relative to the effects of race and gender on
performance evaluation and promotion yields inconsistent
findings. Some investigations (Kraiger and Ford 1985; Kraiger
and Schechtman 1986 Campbell, Crooks, Mahoney, and Rock 1973) of
the effects of the evaluators' race on performance appraisals
reveal significant discrepancie~. On the other hand, other
studies (Schmidt and Johnson 1973) indicate no effects of rater-
race on performance ratings. In a similar vein, ratee-race
effects have been significant in some studies (Farr, O'Leary, and
Bartlett 1971; Landy and Farr 1973) and insignificant in others
(Fox and Lefkowitz 1974; Schmidt and Johnson 1973).
More recent investigations (Kraiger and Ford 1985; Pulakos,
Oppler, White, and Borman 1989; Landy and Farr 1980; Schmitt and
Lappin 1980), illustrate that the effects of rater-ratee race are
significant, particularly within race. That is, when the rater
25
and ratee are of the same race, the rater tends to assign higher
~atings. Several studies (Bass and Turner 1973; Casicio and
Valenzi 1978; pulakos et ale 1989; Waldman and Avolio 1991),
however, emphasize that the significance in rater-ratee race
effects on performance ratings is reduced when controls are
imposed for age, tenure, and individual ability.
Ilgen and Youtz (1986) note that although black employees
tend to receive lower overall ratings than white employees, rater
bias may not be the only causal factor. Suggested are race
differences in actual jOb performance resulting from differential
treatment minorities experience within organi~ations. Treatment
(or the lack there of) experienced by minorities prompts fewer
and less. desirable opportunities. It is argued that
opportunities such as sponsorship, job procedure discretion,
supervisory support and overall acceptance affect subsequent
.
performance.
Greenhaus (1987), for examp~e, describes that managers
{typically minorities) who are assigned routine, nonchallenging,.
meaningless tasks are likely to experience a lack of supervisory
support. Subsequently, performance feedback is based on these
.trivial tasks t~ereby hampering career aspirations over time.
Further, Greenhaus points out that non-support excludes members
zrom crucial informal networks which provide valuable resources
to members in performing their jobs. Hackman and Oldham (1976)
add that members with limited job discretion and autonomy have
fewer opportunities to exhibit decision-making skills. The
26
absence of these type opportunities promote low levels of
motivation which is reflected negatively in performance
evaluations. Lead by previous research, Greenhaus, Parasuraman
and Wormley (1990) offer the availability of sponsorship
opportunities, the level of supervisory career support, feelings
of organizational acceptance, the degree of perceived job
discretion, and participation in formal strategical career
planning as positive indicators of performance evaluation.
Rater-Ratee Gender Effects
Inconsistencies also exist in the research regarding gender
effects on performance evaluations. Although rater gender does
not seem to affec~ performance evaluations (Schmitt and Ldppin
1980; Pulakos and Wexley 1983; Peters, O'Connor, Weekly, Pooyan,
Frank, and Erenkrantz 1984), some studies (Bartol and Butterfield
.
1976; London and Poplawski 1976) have discovered that female
raters tend to be more lenient than male raters. In contrast to
race effects, same gender rater-ratee evaluations do not yield
higher ratings (Bartol and Butterfield 1976; Mobley 1982; pulakos
and Wexley 1983; Izraeli and Izraeli 1985). Some research (Rosen
and Jerdee 1973; Bartol and Butterfield 1976; Landy and Farr
1980) has indicated an interaction between sex of ratee and sex
role perceptions (stereotypes) of the job position as influential
to ratings. In other words, female ratees, as opposed to male
ratees in the same job, are evaluated more positively in female-
27
oriented jOb positions. As well, male ratees are evaluated more
positively in male-oriented job positions.
Research in the area of sex role perceptions, stereotyping,
and performance evaluation (Sackett, Dubois, and Noe 1991:
Heilman 1983: Kanter 1977) concludes that long held traditional
perceptions of sex roles, particularly job related sex roles,
bias the evaluation ratings of women's job performance. Women's
performance in typically male-oriented job positions is often
subject to being discounted severely: due to no other reason than
their sex. Kanter (1977) contends that this phenomena is
continually perpetuated because of few women in the eVQluation
environment and subsequent decision-making process. Kanter
refers to environments with le~s than 15% minority membership as
"skewed" and environments with 15%-35% minority membership as
"tilted." Performance evaluations are more likely biased in -.
skewed settings. The same bias occurs in tilted setting, but
less so.
Promotion/Succession planning
Race and Promotion
Findings (Mueller, Parcel and Tanaka 1989) indicate that
racial minorities are more likely to be promoted in settings
where they will supervise other racial minorities. Even then,
minorities must display a higher level of qualifications (than
their white counterpart) in order to be considered for the same
28
managerial positions. other literature (Wright, Costello,
Hachen, Sprague 1982; Fernendez 1981; Kerckhoff, Campbell and
Trott 1982) noting differences between racial minority and white
supervisors include findings that: 1) racial minorities have to
demonstrate that they subscribe to attitudes and values which are
presumed to be characteristic of management material in order to
be considered for advancement; and, 2) for all racial groups, but
particularly blacks, the awarding of responsibilities affecting
decisions about hiring and firing, promotion, and compensation is
subject to higher scrutiny than the awarding of more subordinates
to supervise.
The lack of decisionmaking authority in financial matters
and an assurance of a subscription to specific values and
attitudes among minority managers is consistent with Kanter's
(1977) analysis of "homosocial reproduction." This contention is
-.
threefold. First, it states that persons who hold power in
organizations are more likely to.promote others like themselves.
Second, persons of the same social background will make similar
decisions and provide continuity in management. Finally, these
choices for management will diminish the possibilities of
creating a threatening environment for current managers and/or
supervisors.
Gender and Promotion
Coinciding with the inconsistencies present in the
literature on race-genderjrater-ratee effects, empirical research
29
regarding the effects of gender on promotion is also non-
conclusive. Some studies (Eberts and Stone 1985; Cannings 1988;
Spurr 1990) suggest that there are significant gender effects on
promotion, while others (Hartmann 1987; Lewis 1986) claim no
significance. Studies that argue that there are little to no
significant gender effects on promotion frequently point to the
advancements women have made in the workforce. A safe conclusion
is that women have made substantial progress in organizational
upward mobility (Cetron, Luken, McFadden, and Weir 1987; Blau and
Ferber 1987), however, they still are faced with disadvantaged
career progression (Morrison and Von Glinow 1990). Stroh, Bret~,
and Reilly (1992) examine the career progression of male and
female members of Fortune 500 corporations who have been tagged
for management positions. After a two year period, the members
are all considered to have equal qualifications - they all
.
possess the "right stuff." Nevertheless, there are significant
differences in career progression, particularly in salary and
geographic mobility.
Cannings and Montmarquette's (1991) investigation of male-
female career progression indicates that women's tendency to rely
more on formal bidding for promotion deprives them of their
"managerial momentumft - superior performance and greater
ambition. Due to the number of management candidates usually
exceeding the supply of top-level managerial promotion
opportunities, superior performance and greater ambition are
frequently discounted as criterion for obtaining promotion. The
30
increased success of men in gaining promotion is attributed to
their greater use of informal networks. It is a less
meritocratic means than formal bidding in acquiring the attentian
of superiors. However, it appears to enable men to offset
deficits in formal performance evaluations which may frequentl.y
be lower than women's.
Snyder, Verderber, Langmeyer, and Myers (1992) emphasize
negative self-referent and organization-referent attitudes (tbat
is: salient negative attitudes which are associated with ones
self, the organization, and/or both) as potentially the most
serious barrier to wumen's upward mobility. Support is pres~t
for the claims that referent attitudes are positively related to
promoti~n outcomes (Tharenou 1979; Shamir 1986; Brockner 1988:
Tharenou and Harker 1982; Gist 1987; Romzek 1989). In essence,
positive self-referent (such as perceptions of confidence and.
competence) and positive organization-referent (such as
organizational commitment) attitudes are strongly related to
upward mobility in organizations. When a women's perceptions of
confidence and competence matches or exceeds that of their male
counterpart, her mobility may still be restricted by an
overriding noncompliance with organizational goals and values.
COMPENSATION PRACTICES/POLICIES
Although wages/income is only one component of compensation
(profit-sharing, stock participation programs, retirement
31
benefits, and vacation accumulation consisting of other forms of
compensation), it represents the bulk of the literature
concerning compensation differentials relative to race and
gender. This may reflect organizations/corporations reluctance
to publicly reveal their financial arrangements. Nevertheless,
it is safe to say that the void in research pertaining to
compensation is the absence of attention given to other forms
(other than wages/income) of compensation.
Minority-Nonminority Earning Differences
The traditional approach (Beeker 1962) to the examination of
compensation differences in male-female earnings has been that of
"human-capital" theory. This perspective argues that an
individuals earnings are a function of his/her training and
experience. In other words, "the only relevant productive
-.
attributes of individuals is their cognitive capabilities"
(Cannings 1991). In recent yearp (Becker 1985), this approach
has yielded a considerable unexplained residual in earning
differences. The residual is typically attributed to
discrimination (Butler 1982; Newnark 1987).
Shulman (1986) contends that neither human capital theory
nor the declining discrimination hypothesis can account for the
minority-nonminority wage gap. Given the quantitative and
qualitative achievements in minority formal education, the
passage of antidiscrimination and affirmative action legislation,
the movement of minorities into occupations characterized by
32
lower turnover rates, and the reduction of the social
acceptability of bigotry, it would be expected that
discrimination has declined. Subsequently, the probabilities of
minority emploYment and wage equality would converge. Shulman
(1986) finds that this has not occurred. He explains minority-
nonminority emploYment and wage differential in terms of the
instability of the labor market since 1970. This market
sustained emploYment discrimination. Therefore; Shulman concludes
that findings concerning a decline in discrimination as well as
policy recommendations focusing solely on the augmentation of
minority human capital be viewed with skepticism. The inclusion
of an unstable labor market over the past two decades must be
integrated into any analysis.
Male-Female Earning Differences
.
An examination of gender differentials indicates that the
disproportionate responsibilitias maintained by women outside the
organization (usually home) operates as a penalizing factor to
their earning potential. Relative to men, ongoing external
constrains on female employees have an adverse affect on
productivity and earnings. These constraints are inherent to the
individual's commitment to the family - that is, the division of
labor of that family due to its particular situation (Greenhalgh
1980; Chapman 1987; Fox and Hesse-Biber 1984; Pleck 1981);
whether children are a part of the family (Langer 1985; Olson and
33
Frieze 1987; Taylor 1986); and, whether it is a dual career
family (Pare 1985; Markham 1987).
TURNOVER
Continuity in Female Employment
Although it is clear that the number of working women has
risen in recent decades, it is not unanimously agreed upon as to
the level of dedication to career of the female workforce. Many
authors have expressed concern about the commitment female
workers make to the workforce. Gallagher (lY90) finds that
women's need for autonomy and flexibility may prompt them to
curtail .career advancement in favor of other commitments.
Gallagher's evidence is sketchy. However, her point is that
businesses which wish to take advantage of the talents of women
-.
must meet the human needs of this highly educated but relatively
untapped labor supply.
Light and Ureta (1990) note that the increase in female
employment levels may be due to women entering the workforce for
shorter periods of time. They also note that women who enter the
workforce seeking continuous employment and who engage in career
building should have career patterns that are similar to those of
men. Their findings suggest that continuous employment is not
the norm among young women but it is a growing trend within the
female workforce. Gender wage gaps narrow when only continuously
employed subjects are analyzed. Later research by these authors
34
(Light andUreta 1992) indicates that while employers may
perceive women as more likely to quit than male employees, this
assumption is not well founded. In many instances young men and
women, who are in the process of establishing their careers, do
quit their jobs in order to advance or find a better "match" of
employer/employee needs. Their results indicate that women are
more attached to the workplace than they were in the past and
that younger cohorts of male and female employees do not show
differences in quitting rates. Only one variable, the presence
of a newborn child, is shown to have any impact on female
t.lJ.rnover. As the authors note, this impact is likely to be
temporary and could be overcome by the provision of maternity
leave and/or chiid care.
The u.s. Bureau of Labor statistics also addresses the rates
of exit and re-entry into the workforce for men and women. Their
-.
findings (see Smith, 1984) conclude that women do have higher
rates of exit and re-entry than .men, and that men have, on
average, longer work lives than women. However, statistics from
this report indicate that activity levels for women are narrowing
this gap, and that women are developing stronger ties to the job
market.
Bureau of Labor statistics data also indicate that white
women are more likely than minority women to leave the job force
in their early thirties. They show a correspondingly strong
likelihood to return to work in the age range from thirty-nine to
forty-four. This research also indicates that periods of work
35
inactivity for white males is often later in life, while inactive
periods for minority males may be during prime work years.
Both genders show increasing propensity to reverse
retirement decisions. Race is a more important factor for male
workers than female workers, with minority males more likely to
leave and less likely to re-enter the job force than their white
counterparts. Worklife differentiation was less defined by
racial differences for women.
Some scholars have expressed concern over commitment to the
workforce among female employees. The disproportionate
responsibiliti~s maintained by women outside the organ~zation can
impact the degree to which female employees devote themselves to
their work. This may, in turn, affect rates of turnover,
compensation, and promotion. Women's perceived need for autonomy
and flexibility may prompt them to curtail career advancements in
.
favor of outside commitments to family. Rosen, Miguel and Peirce
(1989) illustrate the findings of research on the extra-
organizational demands on women. They find that women may 'bail
out' of successful careers due to these external stressors which
include difficulties with child care arrangements, short or non-
existent maternity leave policies, lack of flexibility in working
hours and other family responsibilities. It is of great
importance however to note that women responding to surveys cited
in this research listed these complicating factors as secondary
to institutionalized biases, such as lack of promotion, few
female role models, limited access to challenging and rewarding
36
task assignments, and other problems noted in other sections of
this monograph. Thus it is clear that women may note that
external stressors are salient in their decisions to leave the
workforce (either temporarily or permanently) but these factors
are not the sole determinants of their decisions.
These studies do not indicate that women are becoming
nonquitters, but instead noted that men and women quit jobs in
order to attain the same goals: career advancement in durable
employment relationships where there is a satisfactory match of
skill and demand. Taken with the idea that employers avail
themselves ~f a strong work force, these findings nidicate that
businesses have much to gain by the elimination of the gender
turnover bias.
The idea of turnover bias, that is, that women are more
likely to quit than are men, is problematic at several career
.
stages. It may serve as an impediment to the hiring, training,
and promotion of women. This, like other outdated stereotypes,
must be addressed at the organizational level.
INFORMAL SYSTEMS
Although changes have been made in some organizations
regarding the formal processes by which employees are hired,
trained, and selected for promotion, many subtle modes of
maintaining the status quo persist. Adherence to law and changes
in formal policies that may discriminate are necessary but not
37
~
sufficient steps toward the elimination of the glass ceiling.
Unwritten rules, norms of behavior, organizational politics and
accepted modes of operation may work against the incorporation of
groups into the workforce. In order to address these more subtle
barriers to diversity within the workplace, organizations need to
take proactive steps. In the sections that follow, evidence of
such barriers is presented as well as tactics that have been
successfully utilized to break down such barriers.
Evidence of Informal Barriers
Cultural audit data (see Appendix A for a summary of
companies included in the aUdits) indicate that most employees
feel that unwritten rules do exist and that adherence to such
rules is of paramount importance to career advancement.
Traditions, attitudes, stereotypes, and perceptions are some of
.
these invisible barriers. Norms of behavior based on cultural
background that are different from the norms valued by the
dominant corporate culture may also provide barriers that are not
addressed in corporate literature and standard practices.
Cultural audit data reflect these ideas. When asked whether
unwritten rules do exist, 68 percent of all respondents indicated
that they perceived unwritten rules. This finding was consistent
across race and gender groups (see Table 1). Unwritten rules
were identified by these respondents as adhering to corporate
dress, willingness and availability to stay overtime,
participation in office politics, social interaction with
38
TABLE .1
CULTURAL AUDIT DATA
ARE THERE UNWRITT~N RULES?
RACE AND GENDER CATEGORY
-----------------------------------
WHITE WHITE
RESPONSES MINORITY MINORITYMALES TOTALSFEMALES MALES FEMALES PERCENT
-----------------------------------------------------------------
YES 592 269 148 85
69.3% 109468.8% 66.0% 61.2% 68.0%
NO 195 94 55 37 381
22.8% 24.0% 24.6% 26.6% 23.7%
DO NOT KNOW 67 28 21 17 133
7.8% 7.2% 9.4% 12.2% 8.3%
TOTAL CASES 854 391 224 139 1608
53.1% 24.3% 13.9% 8.6%
----------------------------------------------------------------- 100.0%
NO TWO GROUPS ARE SIGNIFICANTLY DIFFE~ENT AT THE .05 LEVEL.
-.
39
colleagues and 'keeping your mouth shut.' Moreover, when asked
how useful unwritten rules were in understanding the corporate
environment, 41.2 percent of respondents indicated that they were
very useful and 22.6 percent indicated that they were mostly
useful (see Table 2). As was noted earlier in the section on
performance appraisal, employees often find informal feedback
more important than formal performance appraisals.
Thus it appears that informal barriers do exist and may be
more crucial for career development than formal barriers. As is
noted in the sections following, barriers to informal feedback,
and other informal decioionmaking processes may exclude people
who violate corporate norms. Such exclusion may provide
additional barriers to career development.
Horms in Organizations
.
Much has been made of the consequences of being different in
an organizational setting. Much of the literature on
organizational diversity centers around ethnicity and gender
although some attention has been given to age and functional
diversity. Scholars have noted that increasing diversity has
caused conflict. in finding a standard workforce role model (Fine,
Johnson and Ryan 1987). Stereotypes of successful members of the
organization have traditionally found differences to be
deficiencies for employees (Fant 1982; Segal 1962; Kanter 1977;
Washington 1987; Fine, Fern and Ryan 1990) as well as for
organizations (Sadano and Baler 1983:30). Later studies focused
40
TABLE 2
CULTURAL AUDIT DATA
HOW USEFUL ARE UNWRITTEN RULES?
RACE AND GENDER CATEGORY
-----------------------------------
WHITE WHITE MINORITY MINORITY TOTALS
RESPONSES MALES FEMALES MALES FEMALES PERCENT
-----------------------------------------------------------------
NOT USEFUL 20 16 5 8 49
4.7% 9.2% 5.1% 11.6% 6.4%
BARELY USEFUL 27 8 9 3 47
6.4% 4.6% 9.1% 4.3% 6.2%
SOMEWHAT USEFUL 114 20 24 22 180
27.0% 11.5% 24.2% 31.9% 23.6%
MOSTLY USEFUL 104 40 19 10 173
24.6% 23.0% 19.1% 14.5% 22.6%
VERY USEFUL 157 90 42 26 315
37.2% 51.7% 42.4% 37.7% 41.2%
TOTAL CASES 422 174 99 69 764
55.2% 22.8% 13.0% 9.0% 100.0%
~~--
DIFFERENCES BETWEEN RACE AND GENDER CATEGORIES ARE STATISTICALLY
SIGNIFICANT AT THE .05 LEVEL.
4.
on differences as advantageous for the organization and
innovation within organizations (Baird and Bradley 1979; Foeman
and Pressley 1987). More recently, scholars call for a move away
from the ideas toward differences as merely differences and calls
upon managers to use these differences to produce a better
workforce as well as a better place in which to work (Fine,
Johnson and Ryan 1990).
However, there is much evidence in the literature that
differences still do act as barriers to success within
organizations. Washington (1987) finds that issues of
acculturation, corporate values, and organizational expectations
are still challenges faced by employees who are different from
organizational stereotypes. Indeed the same behaviors performed
by white and minority employees may be interpreted differently by
managers. Literature in anthropology (see Okamura 1981) is .
helpful in determining why this is so. Being different offers
role constraints in which actors may feel that certain behaviors
are outside the range of what is acceptable for him/her. In some
cases this is a highly salient factor limiting behavior, while in
some instances the role of being different is less keenly felt.
In any case, the behavior of someone who is different in an
organization is open to interpretation by other organizational
members who may interpret behaviors as different, regardless of
whether they actually are, simply because actors are different.
Persons who are different, especially persons who are one-
of-a-kind in an organization face three issues: visibility,
42
contrast and stereotyping (Sodano and Baler 1983). Being highly
visible may draw attention to the employee when he/she makes
errors, causing him/her to behave more cautiously and be more
self-conscious than others in order to avoid such attention.
Such cautious behavior may not reflect a person's highest
potential. Since those who are different in organizations often
have weaker or no support systems (as is noted in more detail in
the mentoring sections that follow) maintenance behavior becomes
more important, since negative attention will have more serious
drawbacks without such support.
Co-workers as well as manage~s may scrutinize the employee
who is different more than others who conform. High visibility
for a person who represents the first of a type of employee, such
as the first woman or first black, may feel pressure to perform
well in order to make way for others (Morrison, White, and Van
-.
Velsor 1992: Sadano and Baler 1982: Sarason 1973).
Contrast is also important.for those who are different. A
dominant organizational culture may actually develop more
strongly to distance itself from diversity. Thus, people who are
different are isolated, not only by their differences, but by the
accentuation of those differences by members of the majority.
Radin (1980) found that people who feel this contrast may attempt
to rejoin the majority on the fringes, as might be the case when
a female attempts to become a "good old boy".
Finally, stereotyping is problematic for those who are
different. Women who are expected to serve in traditional roles,
43
who are addressed differently from men, or mistaken for executive
wives, and minorities who are assumed to be lower level employees
are examples of such stereotypes. Individuals find themselves
locked into these stereotypes resulting in overcompensation,
reduction of self-esteem and difficulty in rapport with the
dominate corporate culture (Sadano and Baler 1983).
In a cognitive sense, being different is also a barrier to
acculturation within society (Okamura 1981:454) or within an
organization (Washington 1987). Limited knowledge of corporate
complexities, limitations on the breadth of education experience,
and in some instances .diffe~ing religious beliefs, and language,
while not "inferior" may cause problems in an organizational
context. Actors may lack understanding of corporate signs or
norms, as well as the meaning that may be ascribed to these
norms.
-.
There is some evidence that employing several persons who
share characteristics which are.dissimilar to that of the
dominant corporate culture, diminishes the effects of visibility,
contrast, and stereotyping. Moreover, individuals who are
trained together are found to share more common experiences than
individuals who are trained in segregated groups or alone. Thus
one remedy to the problems of isolationism might be to hire
several persons who share characteristics and train them in
groups. An alternative approach provides experiences within
which persons of differing backgrounds can interact, such as
training programs, in order to establish a common experience (See
44
Sadano and Baler 1983:29-30). People are more likely to be
accepted if they are not viewed as tokens, but are instead
perceived to be active members of a group within the
organization. In theory, problems of self-esteem would be
minimized under this ideal.
As indicated earlier, studies show that female employees
During
prefer formal regulations to informal decisionmaking.
critical situations in which decisions are made more informally,
women have a higher tendency to be left out of discussions (Doll
et al 1982; Henning and Jardim 1976; Radin 1980). Thus
differences have worked within organization~ to maintain the
status quo on diversity.
Obviously the organization seeking to maximize contributions
from female employees in times of crises could maximize these
contributions by adhering to formal procedures which include
-.
female employees in discussions and decision making.
One such program that may Qe institutionalized into a formal
system is mentoring. While the evidence below indicates that
mentoring often occurs informally, there are differences between
informal mentoring networks based on race and gender.
Hentoring
Many believe that at least some of the problems related to
glass ceilings are due to placement within organizations. That
is, women, and minorities may not be "in the right place at the
right time." If this is in fact true, then the logical next
45
question is "Why not?" The lack of strategic placement for
advancement may be related to the lack of mentors at a higher
level within organizations, especially in early career stages.
There is strong evidence that employees benefit from mentoring
relationships with more senior colleagues and/or managers (Thomas
1990:479; see also Gabaro 1987; Thomas and Kram 1987; Blackwell
1989; Collins 1983). Some scholars emphasize that sponsorship or
mentoring is one way that organizations separate those who will
be considered for upward mobility from those who will not (see
especially Turner 1965). Mentors provide information on
organizational norms or career opportunities, delegate
responsibility on key projects, and serve as liaisons with
employees and the levels above. Mentors may also develop the
potential within employees that they take under their wings.
Howard and Munch (1991:13) outline the benefits for the employee
-.
as well as the mentor noting that mentors benefit from the
relationship by developing thei~ listening and managerial skills,
while employees are likely to be able to set more realistic
career paths, and highlight their skills. Mentors provide
encouragement, feedback, and information concerning office
politics to their proteges (Kalbfleisch and Davies 1991).
Sponsorship, while akin to mentoring as defined above may be
thought of as a more formal process within organization. For
example sponsors might indicate which employees would be most
appropriate for increased responsibility or promotion. The
assumption therefore is that sponsors themselves have more access
46
to organizational power than other mentors who might be at the
same level of authority as the person being mentored (see Thomas
1990:480; Turner 1965).
Cultural audit data reveal that approximately half of all
respondents indicate that they have mentors at work (see Table
3 ) . However, the degree to which these mentors are in positions
to assist employees with career development issues may be
contingent upon who the mentor is, what access. the mentor has to
authority him/herself, and whether the mentor is capable of
sponsorship.
Thomas' (~990) study of a public utility company-found some
patterns of mentoring and protege selection by mentors based on
race and gender. His findings indicate that cross-race mentoring
does occur, in that black employees are often mentored by white
males. However he also noted that there were high numbers of
.
minority employees who sought mentoring relationships with
persons of their own race. Thi& suggests that employees derive
different benefits from the two types of mentoring relationships.
In one case, employees mentored by white males seek out these
relationships in response to organizational culture. This is
supported by the fact that these relationships were often with
managers and within the employees specialty field. On the other
hand, same-race mentoring relationships were more likely to be
cross-departmental, or with peers. Thus, the cross-race
relationships were clearly more career focused while the same-
race relationships were more social. If theory holds, white
47
TABLE 3 "
CULTURAL AUDIT DATA
DO YOU HAVE A MENTOR?
ORGANIZATIONAL TYPE PRIVATE PUBLIC NON PROFIT
White White Minority Minority White White Minority Minority White White Minority Minority
Male Female Male Female Male Female Male Female Male Female Male Female
YES 310 156 126 62 293 157 81 57 7 18
(47.9) 4(51.6) 16(50.4) (46.4) (44) (44) (42) (44) (50) (45) (28.5) (51.6)
NO 336 146 124 71 362 193 108 70 7 22 10
(52.0) (48.3) 15(49.5) (53.3) (55) (55) (57) (55) (50) (55) (71.2) (48.3)
TOTAL 646 302 250 133 655 co350 189 129 14 40 14 31 -:t
managers and black proteges are less comfortable with each otmer
than white managers and white proteges. The problem that these
results imply for minority, and perhaps women employees, is tbat
whites in positions of power within organizations may feel 1ess
comfortable working with minorities and therefore select whi~
for important and responsible tasks.
Kalbfleisch and Davies (1991) found that racial ties were
more important than same gender relationships when examining
mentoring among black professionals. Despite the low number ~f
subjects in this analysis, their finding are of some note.
Cross-race as well as cross-gender mentoring relationships are,
difficult to initiate and violate organizational norms.
other scholars have noted that minority employees may b~
perceived as risky for the sponsor/mentor. Alvarez (1979)
suggested that minority employees might be required to prove
~
themselves prior to a sponsor/mentor relationship would develop-
Thomas (1990) did not find support for this conclusion in th~ one
organization on which he focused, but earlier studies (Davis and
Watson 1982; Dickens and Dickens 1982) did find that sponsorship
and mentoring was important for a greater share of the career af
minority employees, even after their careers were more firmly
established.
Despite the fact that women are often seen as more nurturing
than men, studies (Parker and Kram 1993; Ragins and Cotton 19.91l
have found that there are serious obstacles to female mentoring
networks. Scholars note that the differences in career-
49
oriBntation between younger and older female employees may be
part of the problem. In addition, a strong mentor/protege
relationship between two women may be opposed by men within
orga'11izations. Male employees may see such a relationship as
threatening to their own power base. In such an instance there
may be an organizational climate that discourages women from
initiating cross-hierarchical relationships (Parker and Kram
1993: 47).
other studies note that female employees may not be able to
initiate mentoring relationships due to the fact that there are
fewer women at higher levels with whom to form re1ationshlps
(Brown 1986; Couric 1987). Cross-gender mentoring presents many
problems for both mentor and protege, such as misconceptions
aboat sexual advances, office gossip, and other stereotypes (see
Ragins and Cotton 1991:940).
-.
Goh (1991) finds that women who are supervised by men are
less satisfied than their male cpunterparts and are less likely
to J'DJementored. Female managers tend to be perceived by their
employees as more emotional (and that perception reinforces
ste:n:-eotypes,see Devanna 1987). On the other hand female
employees tend to prefer male managers, perhaps because male
ma~agers are seen as more powerful advocates for their
sub'ordinates (see Goh 1991:703; Liden 1985).
Male employees are also more likely to form mentoring
relationships in informal ways, such as through sports activities
or ~ut-of-work socialization from which women are isolated.
50
Ragins and Cotton (1991) provide empirical evidence that men face
fewer barriers to mentoring, and are more likely to have
experience in mentoring relationships that makes new
relationships easier. Their evidence also indicates that
employees who have longer tenure and higher levels of
responsibility within the organization are more likely to have
mentoring relationships. This further indicates that female and
minority employees, who are most likely to hold lower level
positions, are less likely to be mentored.
steps may be taken to change the corporate environment to
facilitate these relationships. These may include a discussion
among female employees at different hierarchical levels
concerning goals, and career choices, the initiation of multiple
mentoring relationships, rewarding managers for the development
of junior employees, education of managers as to the needs of
~
employees, and programs to increase self-awareness.
These studies indicate that sponsorship/mentoring may be
more important for minorities and women than for white males and
yet less available to them. Moreover, the relationships that do
exist may be more for social and psychological rewards rather
than for career benefit. other studies examine exactly how
mentoring relationships affect career outcomes.
Mentoring programs were found to alleviate turnover
problems, as well as low morale. After a mentoring program was
adopted within a subunit of the Department of Defense, Howard and
51
Munch (1991:14) found that the program operated with little to no
extra funding, and contributed to the quality of leadership.
In some cases, formal mentoring programs have proven less than
successful (see Morrison and Von Glinow 1990:205) and training in
how to be a mentor has been more beneficial. All employees would
benefit from such training as it promotes awareness of the
barriers that exist within organizations while allowing managers
and employees to seek out those persons with whom they are
comfortable.
The desire to work in an environment and with colleagues
with whom you are comfortable is hardly a startling concept.
Social interactions are often the basis for business decisions
and social patterns indicate that people tend towar~ those with
similar traits as their own. In the workplace, there are
productive, financial, as well as social consequences for such
-.
patterns.
Group Interactions and the Glass Ceiling
As an organization itself is a microcosm of society, so is a
workgroup within an organization a microcosm. The interactions
of groups members may be thought of in a variety of ways: as a
"melting pot," where a new culture is derived from the
incorporation of two or more distinct cultures (A+B=C); as a
cooptation of new groups into an existing framework, (A+B=A); or
as a blending of groups without the loss of identity of any or
either group (A+B=AB). These patterns are complicated by the
52
dominance of one group or identity over others, and may
contribute to the degree to which members of a group are
permitted to take part in decisionmaking, and/or gain or retain
control of the group's attention. Workgroups, perhaps best
defined as two or more interdependent persons who relate to a
larger organization collectively and who are given responsibility
for a task (Hackman 1990:45), have become more popular in the
American workforce in recent years (Jackson 1992:142-3). Clearly
.
the ability of minority and female employees, as well as
employees who differ on other dimensions to work within groups is
of paraTh0unt importance to their acceptance within the
organization as a whole, their ability to be recognized for their
accomplishments, and their ability to develop successful careers.
As is noted throughout this literature analysis, acceptance
within groups runs counter to the ideas of homosocial behavior
(Kanter 1975). Such behavior has been linked to the decision by
managers to hire persons of like.social attributes. It can
therefore be reasoned that similar decisionmaking patterns would
play into the decisions of group members to accept diversity
within the group.
Despite this potential reluctance to work in socially
diverse groups, there is ample evidence that diverse work groups
are productive. Wood (1987) found that mixed gender groups were
more productive than were single-gender groups, and Magjuka and
Baldwin (1991) have shown that in groups where there was
diversity of job function and occupation, team-based employee
53
involvement programs were more effective. Milliken and Vollrath
(1991) also concluded that functional diversity as well as
hierarchical diversity within groups enhances group performance.
One study using college students as subjects, found that
racially mixed groups outperformed racially segregated groups
(Rube and Eatman 1977). Other experimental studies suggest that
the inclusion of divergent viewpoint produced a higher work
product (see for example, Nemeth and Kwan 1987). studies of
management teams confirm the hypothesis that diverse groups are
more productive (see Murray 1989: Bantel and Jackson 1989).
Thus despite the idea that newcomers to an organization
should adjust to the norms and values of that organization group
(as in the melting pot idea above), this does not appear to be
in the best interest of the organization or group. Recent
literature notes that women and minorities are not likely to
.
perceive organizational life the same ways that white males do
(Fine, Johnson and Ryan 1990). This is at least one potential
explanation for the fact that group heterogeneity is associated
with higher levels of turnover and with a lack of group cohesion
(Jackson, et al 1991). How then to accept and even encourage
diversity without contributing to turnover and lack of cohesion?
Clearly the solution to this paradox is in the creation of an
organizational environment in which emotional and personal costs
of diversity are minimized and in which organizational benefits
of diversity are maximized. Organizational norms which encourage
conformity and reward assimilation, managers who implicitly or
54
explicitly impose their own values on groups, and the avoidance
of open communication are barriers to group productivity.
How Can People of Diverse Backgrounds Best Work Together?
Three tables of findings from the cultural audits are
presented here to illustrate how employees perceive problems
within organizations. In each case employees were asked what
remained to be done to remove barriers to women, minority
employees, and white male employees. As Tables 4-6 indicate,
different barriers were found to exist for different types of
e.iuployees. When asking about female employees, respondents noted
that education and training would be important for female
employees' caree~ development. Some comments were also rioted on
the elimination of sexism, however comments by some private
sector employees indicated that women should be kept from
-.
performing some jobs (requiring lifting and physical labor).
This idea in practice would vio~ate the civil rights of female
employees excluded from such jobs. Responses also included
family and flexibility needs, which was not noted as strongly for
male employees.
Training appears to be an across-the-board concern for
employees. Some respondents indicated that training for white
males include human relations training to deal with diversity.
In all three tables, respondents indicated that employees should
be treated fairly and equally, although perceptions of who was
being disadvantaged were disparate. White male employees seem to
55
":
TABLE 4
CULTURAL AUDIT DATA
WHAT NEEDS TO BE DONE TO HELP FEMALE EMPLOYEES
OROAffllA TIONA!. TIn! f'IllVAre "PUllUC NONftlOf1T
.,... "Ue MI-IJ MI-IJ
Male ....aIe N.Ic FeI8Uc ""ke MI-,OJN.Ic Fe.aIe "'1-''1 ""'ke "'iooriOJ MI...dlJ
~Io" """ ...... FeI8Uc N"".I"c r..aIe N.Ic FeI8Uc
n..... 19 j.I 19 n n 1)
(17.91 67 )(11.11 (H) UQI.I) (1S.4) I(17.1) 4(U.I) (11.7) (1U) (6.1) (11.01 (11.9)
1ft..
hlrt, .. )7 19
(U) 91 J4 )1(IS.') I' 17 )('.7) (11.4) 1(11.11 (1U) I 1(19.0) (IS.)) (1U) (LI) (10.11
,... e..... ('11
OaaIIIIo4
)O 1 I
(U) 1(0.1) (L)
J4 14 16
(1.9) 10(10.1) 0(6.1) (U) 1 1(9.0) 1(0.0) (4.) (10.11 (H)
_CI
01.. as.--,
61 .. 41 11
(IUI SO 116 11 1)ClU) (19.) (10.4) (9.1) 0(1904) (1U) 1 1 I(10.7) (0.0) ( 4.) (10.1) (11.91
~DiO.......
EI.~ Sr.t...
II J6 II
(11.11 "II 11 6J )1(IS.II II(1'.)1 (10.7) (14.1) I(16.1) I 0
,
(19.0) (16.1) ,(U) (10.') (0.0) (11.91
Kn,..
em...- 'I-fD)
.. ) IS
(U) I 0(0.1) 0(6.9) 0(1.11) 0(0.0) 0(0.0) 0 0(0.0) (0.0) 0(0.0) (0.01 (0.01 10.01
ea.... o.-.e!op-
Mn_,
14 II 1) ) JO )I(LI) 1(4.'1 (10.11 (L9)
U
(U) 1 I(UI (4.) 0 0(11\.1) (IU) ( LI) (0.0) (0.01
"-1- f_ii"
flea"''', ""'. )1
"II 11 9 J!
(..41 II I~ U
)
(1".'1 (10.11 (
0
1.1\ (6.1) (I 1.1) 0 0(911 111.71 (0.0) (')\ (0.01 (001
Itn.°P'.108
)
II
(0.01 (0.1) ( I1.11 (1.9) 0 1(LII (0.1) 1(1.1) I(0-91 (0.0) (4.;) (10.11 (11.91
""'.81' U IS 10 1 ..
(UO) (6.)) I 10(406\ ( I ~)1.9) I(9.D) (1.1) .(6.1) (0.91 (J I.J)
(4'-') (11.71 (14.)1
00 ,... ~.. II 10 1 J4 )
(UI (U) . ,(406) ( 71.9) 4(10.11 ( LI) 4(4.3) ( L7) II(lJ.o) (1L7) (11.1) (14..1)
TotUc 61 %J7 11. \0) 1)1 111 '6) \\I 16 47 It 41
TABLE 5
CULTURAL AUDIT DATA
WHA T NEEDS TO BE DONE TO HELP MINORITY EMPLOYEES
ORGANIZATIONAL TYPE PRIVATE PUDLIC NON PROFIT
White White Minority Minorily While White Minority Minority White While Minorily Minorily
Male Female Male Female Male Female Male Female Male Female: Male Female
Educat ion!rraining 87 475 20 38 153 90 55 33 5 I3 3 6
(30.3) (23) (20) (31.6) (23.5) (28.2) (26.6) (27.9) (31.2) (27.0) (20.0) (27.2)
Treat Fairly 44 30 17 11 155 50 49 26 1 11 4 8
(15.3) (150 (17) (8.1) (23.8) (15.6) (23.7) (22.0) . (6.2) (22.9) (26.6) (36.3)
Deiter Qualifications 26 17 9 10 102 55 13 10 0 4 2 0
(9.1) (8.5) (9) (8.1) (15.6) (17.2) (6.3) (8.4) (0.0) (8.3) (13.3) (0.0)
I'romolC/( live: 33 36 21 26 41 35 30 21 2 1 1
ResponsilJil ilies I(11.4) (18) (21) (20) (6.3) (10.9) ' (14.5) (17.7) (12.5) (2.0) (6.6) (4.5) .......
If)
Respeci Differences 24 41 23 26 69 49 46 21 4 6 2
Elirninale Predjudices 3(8.3) (20.5) (23) (20) (10.6) (15.3) (22.3) (17.7) (25.0) (12.5) (13.3) (13.6)
Recru itmcnt/J 1iri ng 5 8 2 7 18 8 4 6 I 4 3 1
(1.7) (4) (2) (5.6) (2.7) (2.5) (1.9) (5.0) (6.2) (8.3) (20) (4.5)
Nothing 68 21 8 6 112 32 9 1 3 9 0 3
(23.1) (105) (8) (4) (17.2) (10.0) (4.3) (0.1) (18.7) (18.7) «(J.{I) (13.6)
- - - -
'1ol;'1s 2H7 ::!IIII 1110 124 0511 319 206 liS 16 48 I:' )')
TABLE 6
CULTURAL AUDIT DATA
WHAT NEEDS TO BE DONE TO HELP WHITE MALE EMPLOYEES
OIlOAHTlATIONAL me raN An PUllUC NON PllOAT
WhIle WIUIe 1041_"
Mole loIIood" WhIle«,..k N.Je r-&Ie "u.
loIIood" loIIood" WhIle
104... f<aaIc M&Ie «oaak ""ilc MJood" MJ_"Mole ....... Mole f<aaIc
''''.11<,-
Tn"". )1 IA )\( 106.1) ( U) .0 So )1 11('1.1) ('.J) (6.1) 0(16.7) 0(10.4) 0 0
(1'-') (CI.D) (0.0) (0.0) (0.0)
nu.
fckIJ .. II
(1.9) (1I.'$) ('.S) ().7) (I"
So :16 , ) ,
)CU) (16.7) (11.\) 1(10.7) .(10.01 (1J.01 (14.1) (10.0)
II..
80...
Q.o.IIIIo4
.. 1 I
(U) I 0(O~ 0(UI 0 )(O.t) 0(0.0) (0.0) 1 4(0.0) (0.0) 0(lUI (11.11 (1"') (0.0)
Pr-
010. ~k,
\I , ) 1) , )
(1"0.1) (0) JII(S.t) (1.1) (U) (C) I 0(S.t) (H) . \ I( 6. 7) (0.0) (7.1) ('0.0)
"".n.. "'......io.
01
Vi.."."..'.O...M.......
10 7 4
(U) 1( Y7 II 10 I
c..o
(~ 1.7) ( 1.9) (11.)1 0(11.7) 0(6.S) 0 0
if)
('.0) (001 (0.0) (0.0) (0.0)
T......, .. 1J 11 10
(IU) 114 10 1J n )(U) )
("') ('.J) (17.') (11.4) (16.4) (10) < \(10.0) (11.1) (1"') (10.0)
c...c Dncl--.'
Mn-.
7 ) I 0
(1.1) :16(1.1) ( .111.1) I 0(0.01 (4.1) (14) I 0 I(0.6) (0.0) 0(6.7) (0.0) (7.1) (0.0)
Addrrv r'''liI)1
Flnibll.)' ,.....
4 0 0 1< ,(07) I UO) I1011 ( 0 0 0100' 1.:1 ' 0I 1.!I (191 ( ~.I 1001 (001 (0.01 (0.01
Nuth.... 291 II. 1:0 00 19 40 })
(4'.0) 111.1) (11.1) U(16.1) I II(14.01 (,,,)) 4(11.7) 0(11.1) (10.01 (41.7)
(1'-') (0.0)
Du Ka-
""" 41 1) U 17
(6.7) :16 J6(lJ.O) .. II(10.') (11.1) I(4.1) (11.1) (U) 0 0(17.1) (I J.J) (11.1) (0.0) (0.0)
r.... - 1.1) DO 107 'U 1).1 111 .. 11 )6 14 10
perceive that they are not being treated fairly due to -reverse
discrimination," a view that was also expressed by others,
notably white females in public sector employment.
Of the formal processes addressed in this monograpb, speci~
attention was paid to promotion in the responses of the employees
surveyed. Not surprisingly, females were most likely to menti~
that females needed to be promoted, minorities were most 1ikely
to note that minorities needed to be promoted and white males
were the most likely to respond that white males needed to be
promoted.
CONCLUSIONS
Policy Implications
Based on this extensive literature review, there are three
~
major phases justifying policy recommendations: 1) employee
acquisition, 2) employee development, and 3) laborforce
maintenance. The employee acquisition phase involve equitable
pOlicies/practices in the recruitment and selection of a diverse
workforce.
> Recruitment committees should be representative (racially"
culturally, and gender-wise) of larger society.
> Selection decisions are more equitable when they are not the
product of any particular individual. In other words,
selections should be committee decisions base on previously
established criteria.
> These committees may serve themselves and the organization
better if they venture beyond environments that are familiar
59
and comfortable (not just because committee members attended
school there); beyond those environments that yield
individuals that are similar to the current makeup; and
beyond those environments that are used because they are
cheaper.
> There should be active recruitment on campuses and within
other institutions where female and minority candidates are
represented in the selection pool.
> Organizations may consider a greater use of formalized
internship programs, particularly for women and minorities.
The benefits of internships are twofold: 1) the exposure
socializes the individual to the organizational environment
and appropriate protocol: and, 2) the organization is
allowed the opportunity to train and season potential
employees.
> In the interview and testing stage there is room for policy
regarding the alleviation of biasness. The use of outside
~onsultation in this process may facilitate direct
assessment of individuals' attributes. It appears that
individuals inside the organization are so attached that
they may not be able to evaluate candidates ina non-bias
manner. Rather than consider what the organization is in
need of, individuals tend to access candidates based on
their personal values, attitudes and beliefs. Outside
involvement as well as committee decisions will operate to
diminish such bias.
.
* The employment developmen~ phase consist of
pOlicies/practices pertaining to the development of employees
once they are a part of the organization/corporation~ Keep in
mind the efforts of the overall development process is to elevate
each employee.to his/her maximum potential.
> The most salient problematic issue surrounding performance
appraisal and promotion is the organization's reliance on
informal decisionmaking networks, rather than formal
processes. Appraisal and promotion decisions are best
derived at through inclusion of all formal committee
members. Standards for evaluation and promotion need be
written and clearly understood by all individuals involved
60
(committee members and candidates) - from date of hire to
present.
> To facilitate fairness in the personal dimensions of
evaluation, managers and promotion committee members would
best serve the organization if a previously established
level of training has been successfully completed in the
area of workforce diversity. This training should not be
limited to descriptions and composition profiles of the
current workforce. However, this training should stimulate
sensitivity to personal, social, and cultural differences.
Communication skills (how to listen as well as how to
provide feedback) relative to these differences are
essential.
> Training is a necessary component for a successful
organization. Training programs provide employers
opportunities to optimize employee potential and alleviate
problems associated with the glass ceiling. For example,
on-the-job training grou~s with diverse members provide
opportunities to form more social mentoring relationships.
In addition, group members are then provided with a common
experience, despite their diversity in other areas.
Training that may specifically address the needs of specific
groups, such as assertiveness training for female managers,
would be beneficial.
> Lateral moves to provide breath of experience for all
employees may help dispel stereotypic roles for females or
minorities.
> Formalized mentoring programs to assist employees with
organizational norms and cultures may facilitate the
integration of the nontraditional employee.
> Managers who are successful at developing the potential of
the employees should be rewarded. Evaluation of managers
should include some aspect of employee career development,
and managers and employees should be aware of this aspect of
manager evaluation.
* The laborforce maintenance phase involves policies and
practices which pertain to maintaining a workforce as well as the
environment which is operating at its greatest potential.
61
> Turnover is actually a symptom of the process which creates
and maintains the glass ceiling and a symptom of the glass
ceiling itself. Turnover may be a response to the pressures
of work and other roles in the home. Employers who wish to
optimize the work potential of persons who face this
scheduling dilemma might provide flexibility in work
schedules, and at-home work when possible.
> In some instances the pressures associated with dual roles
may be alleviated by support for family concerns such as
provision of day-care, or broad family leave policies.
> Open discussions of issues that are of concern to employees
will provide employers some insight as to how employee
turnover may be minimized and the needs of organizations and
employees might best be met. It is important to note that
these ideas will maximize the efficiency of the
organization, as well as provide strong employee loyalty.
> Innovative policies on family leave, child care, flex-time,
at-home work, job sharing, maternity/paternity leave, travel
requirements and relocation will facilitate an overall
reduction in turnover.
> Employees must be aware of the personnel policies of the
organizations in which they work. Written organizational
goals for employee diversity that outline specific programs
are far superior to word-of-mouth systems. Organizational
leadership should outline in detail which programs will be
in force to provide for recruitment, promotion, career.
planning, and mentoring for all employees.
> Written policies outlining-performance appraisal should be
available to all employees.
Recommendations for Future Research
Although r~search on differences in gender is the most
commonly addressed question in the literature noted here,
continued research on gender differences is needed. In addition,
the literature addresses racial differences mostly as white/black
comparisons, ignoring other racial minorities.
> Future research should include more racially diverse groups
and make comparisons across more than one dimension. More
62
specific topics which would seem to demand further
investigation include studies of diversity based on age of
employees, task differentiation (functional diversity), and
disability diversity. Additionally, in the area of
compensation, forms other than wages/income need to be
considered. That is the extent to which these other forms
of compensation are available and accessible to minorities
and women.
> Research is devoted to managers or those who have "made it
to the top," with little research available concerning the
"sticky floor" that retains most women and minorities at the
lower levels of responsibility and compensation. Research
on those who fail to achieve higher levels of responsibility
may be as informative, if not more so, as what is currently
available on the glass ceiling. Many studies noted herein
have small samples, use college students as subjects, and/or
are based on mail surveys. Case studies of single
organizations are sometimes available but have little
applicability to the workforce as a whole. In few instances
are testable hYPo'~heses or even propositions drawn from
extant literature. other published sources, not cited in
this review, are simply the thoughts of individuals based on
personal experiences. The quick availability of such
literature may do more harm than good, as personal opinion
may become accepted as fact.
> Field research in this area of study is deficient. Field
experimentation of testable hypotheses should be conducted
in a variety of organizational settings, over time and.
across various dimensions of diversity. It is of paramount
importance that applied field research be conducted on
larger samples in work settings. Furthermore, research on
the advantages that diverse employees bring to organizations
is sorely needed. While there is some research on the
presence of diversity in the workforce, there is little
information available to managers that provides them with
implications of how diversity may benefit their
organizations.
> A crucial role currently stands vacant for government (at
all levels). Given the void in recent literature, the
efforts of various governmental agencies are more
appropriate in encouraging and funding research to fill this
need. The government may act as a clearinghouse for
research and information regarding the glass ceiling.
> It must become acceptable within the academic community to
value and pursue research surrounding these issues. Journal
editors, dissertation committees, promotion and tenure
committees, and the academy as a whole must encourage such
efforts. Evidence that this is not currently the case is
63
reflected in the dated literature and the decline in th&
number of published articles associated with issues of
equity in employment.
> Organizations that have diverse workforces should be funded
for research on employment diversity. These organizations
have the most to gain and offer from understanding diversity
and its related dynamics.
.
64
REFERENCES
Alvarez, R. (1979). Discrimination in Orqanizations. Jossey
Bass.
Arvey, R.D. (1979). Unfair discrimination in the employment
interview: legal and psychological aspects. Psychological
BUlletin, 86:4, 736-765.
Arvey, R.D. and Campion, J.E. (1982). The employment interview: A
summary and review of recent research. Personnel Psycholoqy,
35, 281-322.
Baldwin, M. and Johnson, W.G. (1992). Estimating the effects of
wage discrimination. Review of Economics and Statistics,
446-453.
Bantel, K.A. & Jackson, S.E. (1989). Top Management and
Innovation in banking: Does the composition of the top team
make a dilference? Strateqic Management Journal, ~O, 107-
124.
Berryman-Fink, C. (1985). Male and female managers' views of the
communication skills and training needs of women. Public
Personnel Management, 14, 307-313.
Blackwell, J.E. (1989). Mentoring: An action strategy for
increasing minority faculty. Academe, 75, 8-14.
-.
Braddock, J.H. and McPartland. (1987). How minorities continue to
be excluded from equal enployment opportunities: Research on
labor market and institutional barriers. Journal of Social
Issues, 43:1, 5-39.
Brown, D.A. (1986). The role of mentoring in the professional
lives of university faculty women. Dissertation Abstracts
International, 47:160.
Bureau of National Affairs (1992). sticky floor holds down
majority of women illgove~nment jobs in New York. Government
Employees Relations Report, 30:1490, 1526.
Bureau of Labor statistics (1986). Worklife Estimates: Effects of
Race and Education. U.s. Department of Labor, Bulletin 2254.
Butler, R.J. (1982). Estimating Wage Discrimination in the Labor
Market. Journal of Human Resources, 17, 606-621.
Cannings, K. (1991). An Interdisciplinary approach to analyzing
the managerial gender gap. Human Relations, 44:7, 679-695.
65
Cohen, Y. and Pfeffer, J. (1986).Organizational hiring standards.
Administrative Studies auarterly, 1-24.
Collins, N.W. (1983). Professional Women and Their Mentors,
Prentice-Hall, Englewood Cliffs.
Couric, E. (1989). Women in law firms: The high price of
admission? National Law Journal, Dec.11, 2-12.
Cox, T.,Jr.,and Nkomo, S.M. (1990).Invisible men and women: A
status report on race as a variable in organization behavior
research. Journal of Organizational Behavior, 11, 419-431,
Davis, G. and Watson, G. (1982). Black Life in Corporate
America, Random House, New York.
Devanna, M.A. (1987). Women in management: Progress and promise.
Human Resource Management, 26, 469-481.
Dickens, F. and Dickens, L. (1982). The Black Manager, New York:
Amacom Publishing.
Doll, W. et ale (1982). Fast tracks to Success for Women in
Personnel. Personnel 59(3), 12-22.
Duleer, H.O. and Sanders, S. (1992). Discrimination at the top:
American-born asian and white men. Industrial Relations,
31:3, 416-432.
Fant, O. (1982). Racial diversity in organizations and its .
implications for management. Personnel 59(5), 60-68.
Fine, M.G., JOhnson, F.L., and Ryan, M.S. (1990). Cultural
diversity in the workplace. Public Personnel Management, 19,
305-319.
Gabbaro, J.J. (1987). The development of working relationships.
In: Lorsch, J. (Ed.) Handbook of Organizational Behavior,
Prentice Hall, Englewood Cliffs, New Jersey, 172-189.
Gallagher, M. (~990). Women: The glass-ceiling excuse is so
much nonsense. Business Month, October, 42-44.
Goh, S. C. (1991). Sex differences in perceptions of
interpersonal work style, career emphasis, supervisory
mentoring behavior, and job satisfaction. Sex Roles, 24,
701-709.
Greenhaus, J.H., Parasuraman, S., and Wormley, W.M. (1990).
Effects of race on Organizational experience, job
performance evaluations and career outcomes. Academy of
Manaqement Journal, 33:1, 64-86.
66
~
Hackman, J.R. (1990). Work teams in organizations: An Orienting
framework, In J.R. Hackman (Ed.) Groups That Work (And Those
That Don't): Creating the Conditions for Effective Teamwork,
San Francisco: Jossey-Bass.
Hamilton, G.S. and Roessner, J.D. (1972). How employers screen
disadvantaged job applicants. Monthlv Labor Review, 14-21.
Henning, J. and Jardim, A. (1976) The Manaaerial Woman.
Doubleday, New York.
Howard, R.E., and Munch, J.S. (1991). Mentoring: A federal
women's program initiative. The Bureaucrat, Fall, 13-14.
Jackson, S.E. (1992). Team composition in organizational
settings: Issues in managing an increasingly diverse
workforce. In S. Worchel, W. Wood and J. Simpson (Eds.),
Group process and productivity. Newbury Park, CA: Sage (pp.
138-173).
Jackson, S.E., Brett, J.F., Sessa, V.I., Cooper, D.M. Julin,
J.A., and Peyronnin, K. (1991). Some differences make a
difference: Individual dissimilarity and group heterogeneity
as correlates of recruitment, promotions, and turnover.
Journal of Apulied Psychology, 76:5, 675-689.
Kanter, R.M. (1977) Men and Women of the Coruoration. Basic
Books, New York.
.
Kalbfleisch,P.J. and Davies, A.B. (1991). Minorities and
mentoring: Managing the multicultural institution.
Communication Education 40, 266-271.
Light, A. and Ureta, M. (1990). Gender differences in wages and
job turnover among continuously employed workers. American
Economic Review, 80 (2) 293-297.
Light, A. and Ureta, M. (1992). Panel estimates of male and
female job turnover behavior: Can female nonquitters be
identified? Journal of Labor Economics, 10 (2), 156-181.
Magjuka, R.J. and Baldwin, T.T. (1991). Team-based employee
involvement programs: Effects of design and administration.
Personnel Psychology, 44, 793-812.
Martin, L. (1991). A Report on the Glass Ceiling Initiative.
u.S. Department of Labor.
Martin, L. (1992). Pipelines of Progress: An Update on the Glass
Ceiling Initiative. u.S. Department of Labor.
67
Martocchio, J.J. and Whitener, E.M. (1992). Fairness in personnel
selection: A meta-analysis and policy implications. Human
Relations, 45:5, 489-506.
Milliken, F.J. and Vollrath, D.A. (1991). strategic
decisionmaking tasks and group effectiveness: Insights from
theory and research on small group performance. Human
Relations, 44, 1229-1253.
Morrison, A. M., and Von Glinow, M. A." (1990). Women and
Minorities in Management. American Psychologist, 45 (2),
200-208.
Morrison, A.M., White R.P. and Van Velsor, E. (1992). Breaking
the Glass Ceiling. Addison-Wesley, Reading, Mass.
Mueller, C.W., Parcel, T.L., and Tanaka, K. (1989).
Particularism in authority outcomes of black and white
supervisors. Social Science Research, 18, 1-20.
Murray, A.I. (1989). Top management g~oup heterogeneity and firm
performance. Statistical Management Journal, 10, 125-141.
Neckerman, K.M. and Kirschenman, J. (1991). Hiring strategies,
racial bias, and inner-city workers. Social Problems, 38:4,
433-447.
Nemeth, C.J. and Kwan, J.L. (1987). Minority influence, divergent
thinking and detection of correct solutions. Journal of
ApDlied Psychology, 17, 788-799. .
Neumark, D. (1988). Employers' Discriminatory Behavior and the
Estimation of Wage Discrimination. Journal of Human
Resources, 23, 279-295.
Parker, V. A., and Kram, K. E. (1993). Women mentoring women:
Creating conditions for connection. Business Horizons,
March-April, 42-51.
Parsons, C.K. and Liden, R.C. (1984). Interviewer perceptions of
applicant qualifications: A multivariate field study of
demographic characteristics and nonverbal cues. Journal of
Applied Psychology, 69:4, 557-568.
Pazy, A. (1992). Sex-linked bias in promotion decisions.
Psvchology of Women Ouarterly, 16, 209-228.
Pulakos, E.D., Oppler, S.H. White, L.A. and Borman W.C. (1989).
Examination of race and sex effect on performance ratings.
Journal of Applied Psychology, 74:5, 770-780.
68
Okamura, J.Y. (1981). situational ethnicity. Ethnic and Racial
studies, 4:4, 452-465.
Radin, B. (1980) Leadership training for women in state and local
government. Public Personnel Management 9(2), 52-60.
Ragins, B.R. and Cotton, J.L. (1991). Easier said than done:
Gender differences in perceived barriers to qaining a
mentor. Academy of Management Journal, 34:4, 939-951.
Rendero, T. (1980) Consensus. Personnel 57 (3), 4-10.
Rosen B., M. Miguel, and E. Peirce (1989). Steminq the Exodus of
Women Managers. Human Resource Management, 28, 475-491.
Ruhe, J., and Eatman, J. (1977). Effects of racial composition
on small work groups. Small Group Behavior, 8, 479-486.
Saal, F.E. and Moore, S.C. (1993). Perceptions of promotion
fairness and promotion candidates' qualifications. Journal
of Applied Psychology, 78:1, l05-110.
Sackett, P.R., DuBois, C.!. and Noe, A.W. (1991). Tokenism in
performance evaluation: The effects of work group
representation on male-female and white-black differences in
performance ratings. Journal of Applied Psvcholoay, 76:2,
263-26.
Sadano, A.G. and Baler, S.G. (1983) Accommodation to Contrast: .
Being Different in the Organization. in Advances in
Occupational Mental Health S.L. White, Ed., Jossey-Bass, San
Francisco, 25-36.
Sarason, S. (1973) Jewishness, blackness, and the nature-nurture
controversy. American Psychologist 28, 962-971.
Segal, B. (1962) Male nurses: A study in status contradiction and
prestige loss. Social Forces 41, 31-38.
Shenhav, Y. (1992). Entrance of blacks and women into managerial
position in scientific and engineering occupations: A
longitudinal analysis. Academy of Management Journal. 35:4,
889-901.
Shulman, S. (1987). Discrimination, Human Capital, and Black-
White Unemployment. Journal of Human Resources, 22, 361-
376.
Snyder, R.A., Verderber, K.S. Langmeyer, L. and Myers, M. (1992)
A reconsideration of self- and organizatio-referent
atttitudes as "causes" of the glass ceiling effect. Group
and organization Management 17:3, 260-278.
69
Stroh, L.K. (1992). All the right stuff: A comparison of female
and male managers' career progression. Journal of Applied
Psycholoqy, 77:3, 251-260.
Thomas, D.A. (1990). The impact of race on managers'
experiences of developmental relationships (mentoring and
sponsorship): An intra-organizational study. Journal of
Orqanizational Behavior, 2, 479-492.
Thomas, D. and Kram, K. (1987). Promoting career-enhancing
relationships: The role of the human resource professional.
In: London, M. and Mone, E. (Eds) EmDloyee Career
Development and the Human Resource Professional, praeger
Press, Greenwood.
Tom, V.R. (1971). The role of personality and organizational
images in the recruiting process. Organizational Behavior
. and Human Performance, 6, 573-592.
Turner, R. (1965). Modes of Social ascent through education:
Sponsored and contest mobility. American Suciological
Review, 25, 855-867.
Waldman, D.A. and Avolio, B.J. (1991). Race effects in
performance evaluations: Controlling for ability, education,
and experience. Journal of Applied Psycholoqv, 76:6, 897-
901.
Walsh, J.P. (1988). Selectivity and selective perception: An
investigation of managers' belief structures and information
processing. Academy of Management Journal, 31:4, 873-896.
Washington, C.W. (1987). Acculturation of minorities in large
organizations. The Bureaucrat, 29-34.
Wood, W. (1987). Meta-analytic review of sex differences in
group performance. Psvchological Bulletin, 102, 53-71.
70
.
APPENDIX A
EXPLANATIONS OF THE CULTURAL AUDIT ORGANIZATIONS
-.
71
This section provides explanations of the organizations
included in the cultural audit data presented in the monograph.
Not all organizations are included in each table.
Since 1988, the American Institute for Managing Diversity,
Inc. Has conducted cultural audits in a number of organizations.
These audits include surveys of samples of employees in sufficient
numbers to represent the organizations' workforces. While these
surveys are adapted to specific organizational contexts, there are
a number of survey items which are comparable. Presented here are
some of those items. In some cases wording may differ from
organization to organization, but interorganizational comparisons
can be drawn.
Organizations which have conducted cultural audits are
described below. Not all organizations noted below are represented
in each table, as some questionnaires did not include the questions
illustrated in this monograph.
Organization A: This organization is a local facility owned by an
national manufacturing fir. It is located in rural southeastern
community. One hundred and twenty interviews were conducted at
this site. The majority of respondents were blue-collar workela.
Responsibilities ranged from line supervision to maintenance.
Clerical workers were also included.
Organization B: A national manufacturing firm, this organizationis
employees were interviewed at three sites in the Midwest. Ninety
employees were interviewed including managers, clerical, technical,
and research and development personnel.
Organization C: One hundred and sixty-six employees of this
national food products firm were included in this audit. All
interviews were conducted at the national headquarters in the
Midwest. Management, technical,- clerical employees were included.
Organization D: Over 580 employees of this large transportation
firm were interviewed in the firm's headquarters in the midwest.
Managers, clerical personnel, client-contact personnel, technicians
and maintenance employees were included.
Organization E: This is the local division of a national non-
profit service-organization in a southeastern metropolitan area.
One hundred managers, technicians, and clerical workers were
included.
Organization F: The local division of a national non-profit
organization were interviewed in a midwestern metropolitan area.
Management, clerical and programmatic staff were included as well
as volunteers. In all, 318 responses were achieved.
Organization G: Located in a southeastern metropolitan area, this
organization is a division of a health related public agency.
Sixty-seven employees were included in this analysis.
72
Organization H: Over 460 employees of this public research and
development organization were included in the audit. This
organization is located on the west coast and is affiliated with a
large educational institution. Scientists, managers, technicians
and clerical personnel were included.
organization I: This organization is an agency of the federal
government and employees were interviewed at three sites.
Scientists, managers, clerical workers were included in this audit
of over 450 persons.
.
\
73
APPENDIX B
ANNOTATED RESEARCH OUTLINE
-.
74
THE IMPACT OF RECRUITMENT, SELECTION, PROMOTION
AND COMPENSATION POLICIES AND PRACTICES
ON THE GLASS CEILING
Research Outline
I Introduction
A. Discussion of the u.S. work force
- Although women and minorities experience more
opportunities today than twenty years ago, many
obstacles still exist which inhibit employment and
upward mobility within the labor force. Evidence
indicates that over two-thirds of the new entrants
into the workforce between now and the year 2000
will be women and minorities (U.S. Bureau of Labor
Statistics, 1989). If successful management of
such a diverse work force is to occur, greater
understandings of recruitment, retention, and
promotion must be ascertained.
B. Research questions
- The specific questions considered more salient
to this research effort are as follows:
1) what are the organizational policies and
practices which facilitate or inhibit the
advancement of women and minorities to
management and decision-making positiorls
in business?:
2) under what conditions do women and
minorities advance or fail to advance?:
3) is there evidence that individual
characteristics such as age, tenure within
an organization, gender, and
race/ethnicity are important in the
advancement of employees to management and
decision-making positions?:
4) are there organizational characteristics
(i.e., size, service/manufacturing,
profit/nonprofit, private/public,
governmental, industry type, and/or
geographic area) which influence
advancement?:
75
5) what is the significance of formal and
informal systems (within organizational
structures) in terms of facilitating or
inhibiting the advancement of women and
minorities?; and,
6) what are the research and policy
iDplications of the literature analysis?
- In sum, these questions address the more general
issues framing this inquiry - what works?, for
whom?, and under what circumstances?
II Methodology
A. Data Sources
This inquiry examines the content composing four
primary data sources:
1) Literature Review
* A computer-aided literature search insures
the thorough and comprehensive obtainment
of relevant research. CD-ROM reference
databases such as ABI/INFORM, DISSERTATION
ABSTRACTS, ERIC, INFOTRAC, NEWSPAPER
ABSTRACTS ONDISC, PSYCLIT, SOCIOFILE, and
are employed.
i. ABI/INFORM - (last five years), -.
indexes over 800 journals covering all
areas of business and management
including: corporate structure, business
conditions and trends, management
techniques, products, industry, and
analysis of individual companies and
their competitors.
ii. DISSERTATION ABSTRACTS - (1861-
current quarter), indexes doctoral
dissertations completed at u.S.
accredited institutions, in all areas of
studies.
iii. ERIC - (1966-current quarter),
indexes journals, magazines, and
documents in the area of education and
training.
iv. INFOTRAC - (1989-current month),
indexes over 1,100 periodicals dealing
with business, management, social
76
science, humanities, as well as genera~-
interest topics.
v. NEWSPAPER ABSTRACTS ONDISC - (1985-
current month), indexes nine major u.s-
newspapers: Atlanta Constitution/Atlan~
Journal, Boston Globe, chicago Tribune~
Christian Science Monitor, Los Angeles
Times, New York Times, USA Today, Wa11
Street Journal, and Washington Post.
vi. PSYCLIT - (1974-current quarter),
indexes over 1,300 journal titles fron
over 50 countries. Also included are
summaries of English language books aDd
book chapters from 1987 to date. Its
subject matter covers all areas of
psychology, statistics, methodology,
social processes and social issues,
intelligent systems, and learning
mdthods.
vii. SOCIOFILE - (1974-current three
months), indexes in Sociological
Abstracts and Social Planning/Policy amd
Development Abstracts (SOPODA). AIso
included are relevant dissertations fr~
1986 to date. The subject matter cover.$
sociology, group interaction, social
culture and social structure, .
organizational culture, methodology,
poverty and social welfare, family
structure, and feminist/gender studies-
* By crosstabulating terms which are
associated with the research issues,
titles, authors, journals, books, and
abstracts are acquired. Upon noting the
relevant sources by reading the abstracts"
the source is physically retrieved. If the
source is unpublished, the author is
contacted in order to receive a physical
copy.
2) Expert Interviews
* As the literature is reviewed, leading
scholars, bureaucrats, and business
persons are identified. Later these
individuals are contacted and
interviewed by phone. The purpose of the
interview is to cross-
check the currency
77
of the existing literature. As well,
there is solicitation of their perceptions
of needed research and policy.
~) Cultural Audit Results (AIMD)
* The American Institute for Managing
Diversity, Inc. has been fortunate to work
with numerous organizations over the past
several years on various issues of
diversity management. One of the
conditions of a working relationship with
the Institute is that we are allowed to
utilize the data from these projects for
research purposes. ordinarily, cultural
audit results are confidential
information. In other words, it is not
likely that one would co.e across this
type data in any body of data.
Nevertheless, the Institutes's Cultural
Audit results are reviewed in order to
support or suggest additional human
resource practices and trends.
~) Focus Group Discussion
* Collaboration among select members and
consultants of the Institute insures
validity in the synthesis and analysis of
the data sources.
B. Resource Synthesis strategy -.
- ~he following selection protocol guides the
synthesis of pertinent data:
.
.a.)Academic and Commercial/Popular
* Priority is given to the literature that is
academic in nature. That is, special
attention is place on the source,
author(s), and research method. Quality
works from commercial or popular
magazines/journals are included as support
material.
~) Women and Minorities
* Literature is retrieved as it pertains toany population experiencing the effects of
a "glass ceiling." However, a greater
focus is awarded to women and minorities in
this search.
3) Age of Individual, Tenure with organization,
Size of Organization, Type of Organization,
Industry Type and Geographic Area
78
--
* In addition to gender and race/ethnicity as
factors influencing the advancement of
employees within the workforce, other
variables are considered - separately and
in combination with one another.
4) Formal and Informal Systems
* Traditionally, the recruitment, retention,
and promotion of employees takes place
within formal organizational systems. These
systems are usually written, public, and
considered the official path to successful
emploYment. However, frequently informal
systems within an organization superceed
formal systems. When thi$ occurs, those
individuals following the official path to
advancement typically face numerous
barriers. Both systems are taken into
consideration in this research.
III Synthesis and Analysis
An analysis is done of the total body of data produced by
the four primary data sources. Patterns and trends are
then noted within the confines of the following dimensions
of organizational structure:
A. Recrui tment
* How are women, minorities and other groups
excluded from the recruitment process both.
within the organization for upper level
positions and from outside the organization?
What types of ~ecruitment strategies work best
to facilitate the hiring/promotion of women and
minorities?
B. Selection
* What criteria are used to determine who among a
group of candidates will be selected for
advancement or for hiring? How do these
criteria remove women, minorities and other
diverse groups from consideration?
c. Performance Appraisal/Promotion/Succession Planning
* Are there criteria currently being used for
formal evaluation which disadvantage some groups
of employees, without regard for the quality of
their work or their contributions to the
organization?
79
D. Compensation Practices and Policies
* Are the current industry standards in terms of
practices and policies equitable? Are efforts
being .ade to insure that all persons who
contribute equally to well being of an
organization compensated equally? What
practices promote such equality? How can
organizations identify whether their employees
are compensated for equal work?
E. Turnover
* Given that turnover is costly to the
organization and its productivity, are there
particular groups of employees who have higher
rates of turnover? How can such turnover be
avoided? Are women and minorities (as well as
other diverse groups) systematically encouraged
to leave organizations? What are the roots of
turnover for women/minority employees?
F. Mentoring/Sponsorshi~
* How do mentoring relationships (formal and
informal) contribute to the reduction of
barriers for specific groups of employees? Does
mentoring provide employees with advantages as
they seek to advance at an organization?
G. Unwritten Rules
* Are the unwritten rules in the cultures of
organizations (informal systems) which.
systematically disadvantage groups of employees?
How can this be avoided? How can employees of
diverse backgrounds best work together within a
single organization?
rv Summary and Conclusions
A. Policy Implications
B. Recommendations for Future Research
80