Ely y Thomas - 2001 - Cultural Diversity at Work The Effects of Diversi

Administrative Science
Quarterly http://asq.sagepub.com/
Cultural Diversity at Work: The Effects of Diversity Perspectives on Work Group

Processes and Outcomes
Robin J. Ely and David A. Thomas
Administrative Science Quarterly 2001 46: 229
DOI: 10.2307/2667087
The online version of this article can be found at:

http://asq.sagepub.com/content/46/2/229
Published by:
http://www.sagepublications.com
On behalf of:
Johnson at Cornell University
Additional services and information for Administrative Science Quarterly can be found at:
Email Alerts: http://asq.sagepub.com/cgi/alerts
Subscriptions: http://asq.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations: http://asq.sagepub.com/content/46/2/229.refs.html
>> Version of Record - Jun 1, 2001
What is This?
Downloaded from asq.sagepub.com at Scientific library of Moscow State University on January 19, 2014
CulturalDiversityat This paper develops theory about the conditions under
Work:The Effects of which cultural diversity enhances or detracts from work
group functioning. From qualitative research in three cul-
DiversityPerspectives turally diverse organizations, we identified three different
on WorkGroup perspectives on workforce diversity: the integration-and-
Processes and learning perspective, the access-and-legitimacy perspec-
Outcomes tive, and the discrimination-and-fairness perspective. The
perspective on diversity a work group held influenced
how people expressed and managed tensions related to
Robin J. Ely and diversity, whether those who had been traditionally
David A. Thomas underrepresented in the organization felt respected and
HarvardUniversity valued by their colleagues, and how people interpreted
the meaning of their racial identity at work. These, in
turn, had implications for how well the work group and
its members functioned. All three perspectives on diversi-
ty had been successful in motivating managers to diversi-
fy their staffs, but only the integration-and-learning per-
spective provided the rationale and guidance needed to
achieve sustained benefits from diversity. By identifying
the conditions that intervene between the demographic
composition of a work group and its functioning, our
research helps to explain mixed results on the relation-
ship between cultural diversity and work group out-
comes.0
Americanmanagement literature,both popular(e.g., Thomas,
1991; Morrison,1992) and scholarly(e.g., Jackson et al.,
1992; Cox, 1993), is rife with advice that managers should
increase workforcediversityto enhance work group effec-
tiveness. Empiricalresearch on whether and how diversityis
actuallyrelatedto work groupfunctioningis limited,however,
and the evidence is mixed, depending in parton what kinds
of differences constitute the "diversity"in question (see Mil-
likenand Martins,1996; Pelled, 1996, for reviews).
Researchers have examined the impactof diversityin identity
group memberships, such as race and sex (e.g., Cox, 1993;
Jackson and Ruderman,1995); organizationalgroup member-
ships, such as hierarchicalposition or organizationalfunction
(e.g., Banteland Jackson, 1989; Ancona and Caldwell,1992);
and individualcharacteristics,such as idiosyncraticattitudes,
values, and preferences (e.g., Hoffman,1959; Meglino,
Ravlin,and Adkins, 1989; Bochnerand Hesketh, 1994).
? 2001 by Cornell University.
Althoughcertaintypes of diversityappearto be beneficial,
0001 -8392/01/4602-0229/$3.00. studies focused on race and gender have demonstrated both
positive and negative outcomes (see Williamsand O'Reilly,
1998, for review), suggesting that certainconditions may
The order of authorship bears no relation
to the authors' relative contribution to the
moderate these outcomes. To date, however, most scholars
ideas in this paper; they produced the have only speculated as to what these conditions might be.
ideas in full collaboration. This research As a result, consultants and managers interested in diversity
was funded by grants from the Ford
Foundation and the HarvardBusiness have had to rely largelyon some combinationof common
School Division of Research. We are sense and good faith for the rationalesthey advance about
grateful to our colleagues in the FSC
Research Group-Elaine Backman,
why and how companies should address the issue.
Herminia Ibarra,Maureen Scully, and
Kathleen Valley-for helpful comments on We set out to develop theory, grounded in people's experi-
earlier drafts of this paper. We also thank ences in culturallydiverse work groups, about the conditions
Pamela Ellis, Russell Peace, Dalia Radley- underwhich diversityenhances or detracts from work group
Kingsley, and Rose Miller for their
assistance with data collection. Finally, functioning.Fromour research, we identifiedthree different
we thank Linda Johanson, Rod Kramer, perspectives on workforcediversitythat people embrace,
Joe Porac, and the anonymous ASQ
reviewers for their helpful comments and each with differentimplicationsfor a work group'sabilityto
suggestions. realizethe benefits of its culturaldiversity.We use these
229/AdministrativeScience Quarterly,46 (2001): 229-273
observations here to examine criticallysome of the themes
and basic assumptions of previous research and to propose
new directionsfor both researchers and practitionersinter-
ested in diversity.
Diversityis a characteristicof groups of two or more people
and typicallyrefers to demographicdifferences of one sort or
anotheramong group members (McGrath,Berdahl,and
Arrow,1995). Researchers have generated numerous dimen-
sions for classifyingdemographicdifferences, often positing
differentoutcomes for people and work groups, depending
on the degree and natureof those differences. Pelled (1996)
made one set of predictionsabout the impactof racialdiver-
sity among group members and anotherabout the impactof
functionalbackgrounddiversity,based on the visibilityof race
and the job-relatednessof functionalbackground.Others
have distinguishedamong the effects of diversitydepending
on whether differences are cultural(Cox, 1993; Larkey,
1996), physical(Strangoret al., 1992), inherentand
immutable(Maznevski,1994), or role-related(Maznevski,
1994; Pelled, 1996).
Perhaps more importantly,researchers' predictionsabout any
one diversityvariabledifferdepending on which of its dimen-
sions they see as criticalto determiningits impact. Pelled
(1996) predictedthat racialdiversity,as a source of visible dif-
ferences, would incite intergroupbias and lead to negative
outcomes for work groups, while Cox, Lobel,and McLeod
(1991) predictedthat racialdiversity,as a source of cultural
differences, would enhance creative problemsolving and
lead to positive outcomes for work groups. Maznevski(1994)
suggested that racialdiversity,as a source of inherentand
immutabledifferences would providegroups with different
kinds of informationfrom which they could potentiallybene-
fit, but such differences would often be difficultfor parties to
understandand accept. As these examples illustrate,both
the types and dimensions of demographicvariablesin which
one is interested shape one's inquiry.
In this research, the demographicvariablesin which we were
interested include race, ethnicity,sex, social class, religion,
nationality,and sexual identity,all of which contributeto cul-
turalidentity.Accordingto Cox (1993), culturalidentities stem
from membership in groups that are socioculturallydistinct.
They are often associated with particularphysical(e.g., skin
color),biological(e.g., genitalia),or stylistic (e.g., dress) fea-
tures, though these may be more or less identifiable,
depending in parton people's choices about whether and
how they wish to be identifiedby others. Members of a cul-
turalidentitygrouptend to share certainworldviews (Alderfer
and Smith, 1982), norms, values, goal priorities,and sociocul-
turalheritage (Cox, 1993). The culturalmarkersof such
groups can be communicatedthroughcommunicationstyle,
rules, shared meaning, and even dialects or languages, which
others may or may not recognize as culturallylinked(Larkey,
1996). The degree to which one personallyidentifies with
one's culturalidentitiesand the value one places on them
varyacross culturalgroups and across members within cul-
turalgroups (Cox, 1993; Thomas, 1993; Ely,1995; Ragins,
1997). Moreover,a person may vary in the degree to which
230/ASQ, June 2001
Cultural Diversity
he or she identifieswith, values, or expresses a particular

culturalidentityat any given time, depending on the salience
and meaning of that identityin the context within which he
or she is operating(Ely,1995; Larkey,1996). Hence, cultural
identity,as we understandit, is socially constructed, com-
plex, and dynamic.
In addition,culturalidentitiesare associated in the largersoci-
ety with certainpower positions, such that some cultural
identitygroups have greater power, prestige, and status than
others (e.g., Ridgewayand Berger,1986; Nkomo, 1992;
Ragins, 1997). InWestern society, men as a groupare more
powerful-have higherstatus and hold more positions of for-
mal organizationaland politicalpower-than women as a
group;similarly,whites are more powerfulthan people of
color;Christiansare more powerfulthan Jews; presumed
heterosexuals are more powerfulthan gays, lesbians, and
bisexuals;and the middle, upper-middle,and upperclasses
are more powerfulthan the workingand lower classes.
There is much theoreticaland empiricalsupportfor the
notion that payingattentionto differences in power and sta-
tus is criticalfor understandingdiversityin organizations.In
Alderfer's(1987) theory of intergrouprelations,for example,
the distributionof power among culturalidentitygroups, both
inside the organizationand in the largersociety, is key to how
people think,feel, and behave at work. Similarly,proponents
of status characteristicstheory (Ridgeway,1988; 1991) argue
that much of what we thinkof as the effects of membership
in particularidentitygroups, such as race or sex, are in fact
producedby the status value our society ascribes to those
groups. Inorganizations,status differentialsare reinforced
when higher-statusidentitygroups are disproportionatelyrep-
resented in positions of organizationalauthorityand are chal-
lenged when they are not (Alderfer,1987; Lauand
Murnighan,1998). Perceptionsof one's relativestatus in the
organization,in turn, influence one's expectations and behav-
iors. Empiricalevidence showing differentialimpacts of race
and sex as a functionof the social status accorded different
race and sex groups supports the general position these the-
ories advance that to understandthe impactof culturaldiver-
sity in work groups, one must consider the relativepower
positions of culturalgroups both in and outside of the organi-
zation (e.g., Ruhe and Eatman,1977; Zimmer,1988; Tsui,
Egan,and O'Reilly,1992).
By casting the demographicvariablesof interest in this study
as aspects of culturalidentity,the meaning and conse-
quences of which are sociallyconstructed and dynamic,we
were well positioned to consider the role that differentwork
group conditions might play in shaping whether and how cul-
turaldiversityinfluences work groupfunctioning.This
approach,together with attentionto organizationaland soci-
etal power differences between culturalidentitygroups,
structuredour conceptual framingof diversity.
DIVERSITY AND WORKGROUPFUNCTIONING
Researchers interested in the impactof demographyon indi-
vidualand group behaviorin organizationshave taken several
differentapproaches,two of which are especially relevantto
231/ASQ, June 2001
our work. The first involves research on how the proportional
representationof certaindemographicgroups influences
those traditionallyin the minority.The second involves
research on the effects of group composition on outcomes
relatedto work group effectiveness.
Effects of proportional representation. Much of the litera-
ture on proportionalrepresentationhas focused on the
question of whether increasingthe numberof traditionally
underrepresentedgroups, such as white women and people
of color, has a positive or negative impacton members of
those groups. Some theorists have argued that increased
numbers of women, for example, should lead to greater con-
tact between men and women (Blau,1977), less stereotyped
perceptions of women (Kanter,1977), and less spilloverfrom
sex roles to work roles (Gutek,1985); hence, discrimination
against women should subside as their numbers increase.
This line of reasoningsuggests that increasingthe numbers
of people in traditionallyunderrepresentedgroups in organiza-
tions will ultimatelyenhance a work group's effectiveness by
removingthe barriersassociated with minoritystatus and
thereby enablingall people to be maximallyproductive(Cox,
1993; Larkey,1996). Blalock(1957) has argued, alternatively,
that numericincreases in the representationof groups tradi-
tionallyin the minoritythreaten the majority.Hence, men, for
example, should react to increasingnumbers of women in
the workplacewith heightened levels of discriminatory
behavior,to limitwomen's power gains. Yoder(1991)
described this response as "backlash"from the majority.Pro-
ponents of this view have argued that balancingnumbers as
a strategy to end discriminationis by itself insufficient;it is
also necessary to attend to the ongoing relationships
between groups, particularly to intergroupstatus and power
differentialsthat would otherwise remainintact (Zimmer,
1988; Alderfer,1992).
Empiricalevidence exists to support both claims (forreviews,
see Martin,1985; Konrad,Winter,and Gutek, 1992). Some
studies have shown that when they are in the numerical
minorityin a group,women and people of color experience
negative outcomes (e.g., Taylorand Fiske, 1976; Spangler,
Gordon,and Pipkin,1978; lzraeli,1983; Dworkin,Chafetz,
and Dworkin,1983); others have shown that women and
people of color experience more positive outcomes when in
the numericalminority(e.g., Harlanand Weiss, 1981; South
et al., 1982; Deaux and Ullman,1983; Torenand Kraus,
1987). Proponentson both sides of the debate tend to agree
that increasingthe numbers of traditionallyunderrepresented
groups without alteringpower relationsbetween dominants
and subdominantsis unlikelyto improvethe position of those
groups substantially(South et al., 1982; Konrad,Winter,and
Gutek, 1992). Conclusionsas to whether numberbalancingis
sufficient to alter power relationsremainequivocalat best,
however, and the conditions, if any, underwhich such efforts
might enhance work group effectiveness have yet to be
determined.
Effects of group composition. The second approachto
understandinghow demographicdiversitymight influence
work groups is predicatedon the notion that demographic
232/ASQ, June 2001
Cultural Diversity
diversityincreases the availablepool of resources-net-

works, perspectives, styles, knowledge, and insights-that
people can bringto bear on complex problems. Some have
speculated as to what those new resources might be, focus-
ing on the potentialcontributionsthat traditionallyunderrepre-
sented people, such as women and people of color, may
have to offer work groups. Others have examined empirically
the linkbetween group diversityand group outcomes, focus-
ing on the potentialcontributionsthat diverse groups have to
offer relativeto those that are more homogeneous.
Those interested in the contributionsof traditionallyunderrep-
resented groups have argued that the culturalstyles and per-
spectives of these people, althoughtypicallyignoredor deval-
ued, are in fact valuableassets to work groups. The most
vocal proponentsof this point of view are those who contend
that women's difference from men, particularly their relation-
ship orientation,which has traditionallymarkedthem as ill-
suited for the hard-driving,task orientationof the workplace,
in fact constitutes an effective and much-needed manage-
ment style. Hence, they argue, gender diversityin managerial
rankswould serve the group's needs better than most cur-
rent arrangements,in which men are numericallydominantat
those levels (Helgesen, 1990; Rosener, 1990).
Debates about the merits of these arguments rage across
the disciplineson both empiricaland politicalgrounds (see
Harding,1986; Di Stefano, 1990). Althoughsome have pro-
vided compellingqualitativeaccounts of "women's differ-
ence" (Gilligan,1982; Belenkyet al., 1986), Eaglyand John-
son (1990) concluded from their meta-analysisof quantitative
research on sex differences in leadershipstyle that such dif-
ferences are minimalat best. Based on the lackof quantita-
tive empiricalsupport (e.g., Epstein, 1988; Mednick,1989)
and on arguments that the case for the feminizationof man-
agement maintainsthe power imbalancebetween men and
women (Calasand Smircich,1993), some scholars have
urged social scientists to abandonnotions about women's
uniquequalitiesand contributions(e.g., Flax,1990; Valian,
1998; Fletcher,1999; Elyand Meyerson, 2000).
The parallelcase for racialdiversityin organizationsis less
well developed. It is based on researchthat documents cul-
turaldifferences between whites and blacks in communica-
tion styles. Some have used this research to suggest that
blackculturalvalues, such as assertiveness and forthright-
ness, and language patterns, such as verbalinventiveness,
may be beneficialin workplaceinteractionsand represent
positive attributesratherthan deficiencies in need of remedi-
ation (Foemanand Pressley, 1987), but we know of no
empiricalwork that examines this hypothesis directly.
The skepticism as well as mixed results concerninginter-
group differences in organizationalbehaviordiminishthe
potentialvalue of this line of research for elucidatingthe rela-
tionship between culturaldiversityand work group effective-
ness. Women and people of color may well bringdifferent
perspectives and styles to the workplace,but research has
yet to demonstrate whether, underwhat conditions,and with
what consequences they actuallyexpress them.
233/ASQ, June 2001
Others interested in group compositionaleffects have taken
a differenttack, focusing on the impact of diversityin the
work group, ratherthan on the merits of newcomers who
make the work group diverse. Here again, the argumentfor
diversityis based on the notion that members of heteroge-
neous groups have differentpoints of view, but instead of
identifyingwhat those points of view are and who holds
them, these scholars contend that what is importantis the
diversityitself: heterogeneous groups are more likelyto gen-
erate a diverse set of recommended approachesto tasks or
solutions to problems;this in turnstimulates effective group
discussion, which leads ultimatelyto high qualitydecisions
(Wanousand Youtz,1986). Forgroups that are heteroge-
neous on the culturalidentityvariablesin which we are inter-
ested, the evidence for this hypothesis is mixed. Mixed-sex
groups have performedboth better (Hoffmanand Maier,
1961; Ruhe, 1978; Wood, 1987) and worse (Zillerand Exline,
1958; Kentand McGrath,1969; Clement and Schiereck,
1973; Murnighanand Conlon, 1991) than single-sex groups.
Similarly,groups that are racially,ethnically,and/ornationally
diverse have demonstrated both positive outcomes (Fiedler,
1966; Ruhe and Eatman,1977; Watson, Kumar,and
Michaelsen, 1993; Cox, Lobel,and McLeod, 1991) and nega-
tive outcomes (Fiedler,Meuwese, and Oonk, 1961; Shaw,
1983; Tsui, Egan,and O'Reilly,1992) relativeto groups that
are homogenous on these dimensions.
Recent studies of factors that moderate the relationship
between culturaldiversityand work group effectiveness have
begun to make some sense of these findings, suggesting
that when group members share common goals and values,
culturaldiversityleads to more beneficialoutcomes (Chatman
et al., 1998; Jehn, Northcraft,and Neale, 1999). We elaborate
this moderatorstrategy in our paper by suggesting that the
impactof culturaldiversityon groupfunctioningis influenced
by what we call the group's "diversityperspective":group
members' normativebeliefs and expectations about cultural
diversityand its role in their work group.The characteristics
of diversityperspectives includethe rationalethat guides
people's efforts to create and respond to culturaldiversityin
a work group;normativebeliefs about the value of cultural
identityat work;expectations about the kindof impact, if
any, culturaldifferences can and should have on the group
and its work;and beliefs about what constitutes progress
toward the ideal multiculturalwork group. A diversity
perspective can be both explicit,as in verbalor written state-
ments or policies, and implicit,as in the unstated assump-
tions that underliethe way a person manages his or her sub-
ordinatesor the way a group structures its work. Following
Thomas and Ely(1996), we argue that diversityperspectives
are classifiableinto three types: integrationand learning,
access and legitimacy,and discriminationand fairness. Inthe
present study, we present evidence for each of these per-
spectives drawnfrom (1) the rhetoricparticipantsespoused
when we asked them directlyabout the impact of cultural
diversityat work, (2) the implicitand explicitassumptions in
participants'descriptionsof organizationalevents and their
own organizationalbehaviorand experiences, and (3) the
implicitand explicitassumptions underlyingtheir work
234/ASQ, June 2001
Cultural Diversity
group's policies and practices.We also show how these per-

spectives influence intermediatework group outcomes that
are importantfor maintainingthe integrityand properfunc-
tioningof the group.
Overview of the Present Study
With theory-generationin mind, we set out to investigate
underwhat conditionsculturaldiversityin a work group
enhances or detracts from the group'sfunctioning.This ques-
tion requiredthat we develop an approachto conceptualizing
and assessing the work group'sfunctioning.As Cox (1993)
pointed out, to assess the impactof diversityon a firm'sbot-
tom-line performanceis problematic,since it is difficultto
isolate the specific causes of outcomes like profitability,and
culturaldiversityis likelyto be a relativelydistal factor.There-
fore, we identifiedseveral kindsof intermediateoutcomes
that ought to be more proximallyrelatedto the culturalcom-
position of the work group, includingboth achievement and
affective outcomes (Cox, 1993). These includedgroup
processes and individualexperiences that seemed to follow
from diversityperspectives: (1) the natureof race relationsin
people's immediatework environment,includingthe nature
of conflictand conflict resolution;(2) the extent to which par-
ticipantsfelt valued and respected by coworkers and supervi-
sors; and (3) the meaning and significanceparticipants
attached to their own racialidentityat work, including
whether and how they personallyvalued and expressed
themselves as members of their racialidentitygroup.We
also documented aspects of individualand group functioning
that we could reasonablyattributeor relate to these process-
es and experiences. These variedacross work groups and
includedparticipants'statements about their own self-effica-
cy and abilityto work effectively and contributeproductively
to work group or organizationalgoals, the qualityof services
they produced,their abilityto reach desired markets,and the
efficacy of their work group's practices.We sought concrete
examples or incidents participantsdescribed that might illus-
trate how a diversityperspective shaped group processes
and individualexperiences and how these, in turn, influenced
individualor groupfunctioning.Figure1 summarizesthe con-
ceptual model that we develop in the remainderof this paper
to systematize our observations.
We studied three professionalservices firms, each of which
had significantsuccess in recruitingand retaininga culturally
diverse workforce.Two had reputationsfor being high-func-
tioning, multiculturalfirms;the thirdwas experiencingcon-
flicts and had concerns about the qualityof its performance.
This variabilitygave us an opportunityto investigate in the
field what conditionsfoster more positive work relationships
and outcomes in some instances and less positive outcomes
in others. Althoughwe were interested in examiningdiversity
across a range of culturaldifferences, we focus our analysis
in this paper primarilyon race, because, even though the
organizationsin our study were all culturallydiverse, different
kindsof culturaldifferences were salient in each. Inone,
salient culturaldifferences includedrace, social class, and
sexual orientation;in another,they were race, gender, and
social class; and in the third,they were race, gender, religion,
235/ASQ, June 2001
Figure 1. Relationship between cultural identity diversity and work
group functioning.
DIVERSITYIN WORK
CULTURAL GROUP
IDENTITIES FUNCTIONING
INTERMEDIATEGROUP OUTCOMES
* Quality of Intergroup Relations
* Degree of Feeling Valued and Respected
* Meaning and Significance of Cultural
Identity at Work
WORK GROUP
DIVERSITY
PERSPECTIVE
and nationality.We focused on race because it was the

aspect of diversitythat was salient in all three and would
allow'us to make work group comparisons across firms.
Althoughdifferentculturalidentitygroups are associated with
differentsocioculturalpatternsand intergrouprelations,
because they share many of the basic features we outlined
above, we should be able to generalize much of what we
learnfrom our analysis of race to diversityon other aspects
of culturalidentity.
Ouremphasis on culturalidentityhelped to frame our
research in two additionalways. First,because the distribu-
tion of power inside the organizationcan either reinforceor
challenge the racialimbalanceof power in the largersociety,
with significantconsequences for work groups and their
members (Alderferand Smith, 1982; Alderfer,1987; Ridge-
way, 1988), we wanted to controlfor power differentials
between whites and people of color in the organization.It
was important,therefore, that in all three organizationsin our
sample, people of color held significantpositions of both for-
mal and informalauthority.Althoughmany have hypothesized
that this should bode well for a work group's abilityto man-
age its diversityeffectively (e.g., Cox, 1993; Larkey,1996),
people's experiences in these organizationswere mixed. This
accords with inconsistent findings in the literatureabout the
impactof increased minorityrepresentation.Our research
design gave us the opportunityto explore the potentiallydif-
ferent ways in which people managed the contradiction
between the racialimbalanceof power in the largersociety
and the more balancedsituationinside these organizations.
Such differences, we speculated, might help to explainwhy
increasingminorityrepresentationsometimes leads to
positive and sometimes to negative outcomes. Second, con-
ceiving of culturalidentityas socially constructed led us to
investigate the meanings people attributedto their own and
others' culturalidentities, how they expressed their cultural
identities at work, and with what consequences. We were
2361A5Q,June 2001
CulturalDiversity
especially attentive to how context might shape people's

thoughts, feelings, and behaviorsin this regardand how
these, in turn, might influencethe role of culturaldiversityin
the work group'sfunctioning.
METHOD
We studied a consultingfirm,a financialservices firm,and a
law firm.We based the research in all three sites on Alderfer
and Smith's (1982; Alderfer,1987) embedded intergroupthe-
ory,which delineates a method for researchingintergroup
relationsin organizations.The method involves a three-phase
process of entering the organizationand negotiatingthe
terms of the inquiry,collecting data, and providingfeedback
(see Alderfer,1980, for details). Each phase is designed to
maximizeunderstandingof how cultural-identity-group mem-
berships influence people, their relationships,and their work.
The Law Firm
The law firm is a small, nonprofitpublic-interestlaw firm
whose mission is to protect and advance the rightsand well-
being of economicallydisadvantagedwomen. Foundedabout
20 years earlier,the firm had undergone a transitionover the
previousten years from a professionalstaff composed entire-
ly of whites to one that includeda programstaff that was at
least half people of color.Althoughthe senior management
positions of the firmwere still held by whites, we included
the firm in our study because people of color held positions
of significantauthorityin the firm.This firm had a reputation
for being a high-functioning,multiculturalorganization.It had
12 employees at the time of our study; six were white, six
were people of color,and all participatedin this research.
This includedthe executive and associate directorsof the
firm (bothwhite), the managingattorney (white),five pro-
gram/professionalstaff (two white, two Latinas,and one
Asian American),and four supportstaff (one white, two Lati-
nas, and one AfricanCaribbean).We also interviewedthree
former members of the programstaff. One, a Latina,had
been the first woman of color to join the professionalstaff.
Another,a white woman, had witnessed the demographic
change from an all-whiteprofessionalstaff to a multicultural
one. The thirdwas an AfricanAmericanwoman who had
recently left the professionalstaff after six years.
The Financial Services Firm
The financialservices firm is a for-profitcompany whose mis-
sion is to develop and revitalizethe economy of the largely
poor,AfricanAmericanurbancommunityin which it is situat-
ed. Inthe course of the firm's20-year historyit had changed
from a predominantlywhite professionaland managerialstaff
to one that includedabout 40 percent people of color, mostly
AfricanAmericans. Likethe law firm,this firm had a reputa-
tion for being a high functioning,multiculturalorganization.
We interviewed29 employees or about 24 percent of the
firm.We began by interviewingall seven members of the
management committee (fourwhites and three African
Americans)and two senior human resources managers (one
white and one AfricanAmerican)and then focused the
remainderof our data collection in the loan departmentand
237/ASQ, June 2001
in the two departments of the Sales Division.Accordingto
the management committee, these departmentstogether
represented a range of the firm'sdiversity-relatedexperi-
ences. We interviewedall members of the loan department
and the smaller departmentin the Sales Division(seven
whites and five AfricanAmericans)and eight members, or
about two-thirds,of the largerdepartmentin the Sales Divi-
sion (allAfricanAmericans).
The Consulting Firm
The consultingfirm is a nonprofit,internationalplanningand
consultingcompany that focuses on foreign and domestic
urbaneconomic development. Havingoperated for many
years as a predominantlywhite organization,over the 15-year
periodpriorto our data collection, it had implemented an
aggressive affirmativeaction plan designed to increase the
numberof white women and people of color in the organiza-
tion, especially in professionalpositions. At the time of our
study, 40 percent of the firm'sprofessionaland managerial
staff were people of color. Unlikethe other two firms, this
one was strugglingto sustain its diversityin the face of a
series of conflicts and performanceconcerns. We inter-
viewed 37 employees or about 30 percent of the firm.This
includednine members of the management committee (six
white and three AfricanAmerican),16 projectleaders/middle
managers (nine white, five AfricanAmerican,and two Latina),
and 12 supportstaff (five white and seven African
American).1This interviewee groupwas proportionatelyrep-
resentative of the four work groups that constituted the
firm'sstructure:AdministrativeSupport,Research and
Development, NorthAmericanOperations,and International
Development. The lattertwo groups were the largest and
accounted for over 90 percent of the firm'sfee-for-service
work.
Data Collection
We collected data primarilythroughinterviewswith partici-
pants and by observing between two and six staff meetings
in each organization.We tape-recordedand transcribedthe
interviews, which lasted between one and two hours each,
took detailed notes duringstaff meetings, and made field
notes after each site visit. The composition of the data collec-
I tion teams variedacross research sites, depending on the
We often categorize the nonwhite mem-
bers of our sample as a single group,
size and race and sex composition of the firm.A team of two
which we call "people of color." Although people, one AfricanAmericanand one white, collected the
the particular racial and ethnic identities data in the law firm;a team of four, includingtwo African
of these members varied, they too
referred to themselves, in the law and Americansand two whites, collected the data in the consult-
financial services firms, as members of ing firm;and a team of three, includingtwo AfricanAmeri-
the larger group, "people of color," and,
in the case of the consulting firm, as
cans and one white, collected the data in the financialser-
members of the larger group, "third-world vices firm. One or both of the authorswere on each data
people." Therefore, despite the many dif- collection team. Formost interviews, interviewerand inter-
ferences among the racial and ethnic
groups represented in this study, partici- viewee were matched on race and sex, since there is some
pants themselves seemed comfortable evidence to suggest that such matchingincreases the validity
identifying with a larger category, such as of the data, especially on emotionallycharged topics such as
the one we use here. Following our par-
ticipants' lead, we also use the labels race relations(Alderferet al., 1980). There were some cross-
"African American" and "black" inter- race/cross-sex interviewer-intervieweepairsas well, howev-
changeably, though we sometimes use
"black" to refer more generally to people er, and several interviews were conducted jointlyby cross-
of African descent. race/cross-sex interviewerteams.
238/ASQ, June 2001
Cultural Diversity
Interviewscentered on fourtypes of questions. First,we

asked participantsdirectlyabout theirobservations,beliefs,
and attitudesconcerningculturaldiversity,its value, and its
impact,if any, on the group'sworkand work processes. Sec-
ond, we asked whether culturaldiversityhad posed any partic-
ularchallenges or opportunities.Third,we asked people about
the salience of theirown culturalidentitygroups and the
impactof these groupmemberships,if any, on theirown work
and experiences in the organization;we were especiallyinter-
ested in people's perceptionsof how theirculturalidentity
groupmembershipsinfluencedtheirabilityto work effectively
and exert influencein theirwork group. Finally,we asked what
intergrouprelationswere likeand whether intergrouprelations
had influencedtheirwork positively,negatively,or not at all.
We consistentlyprobedfor examples and incidentsthat would
supportand illustrateparticipants'views. Table1 describes the
participantsfrom each firmwho were involvedin the interview
phase of data collection;in both the consultingfirmand the
financialservices firm,we also administereda firm-widesur-
vey followingthe interviews.We focus on the interviewdata
in this paperbecause they were of greatest value in our efforts
to generate theory.2
Data Analysis
2
We completed our work in each organiza- The authors independentlyread all of the transcriptsand field
tion with feedback sessions in which notes from each organizationto identifythemes that might
members of the data collection team pre-
sented their findings to the organization, explainsimilaritiesand differences within and across firms'
to provide organization members with a experiences of their diversity,in particular,how and under
picture of their organization, as the team what conditionsdiversityenhanced or detracted from their
saw it, and to give them an opportunity to
react to and discuss the team's findings. effective functioning.We then met to discuss our observa-
Although our preparation for these ses- tions and discovered that we had seized on the same insight:
sions was a first step in our process of
learning about the organizations in our
there seemed to be three differentperspectives that gov-
sample, we did not analyze the data to erned how members of work groups created and responded
address our research question directly to diversity,and these perspectives seemed to have impor-
until we had completed data collection
and feedback in all three. For this reason, tant implicationsfor how well the groups functioned (Thomas
we were not able to design the surveys and Ely,1996). This then became our workinghypothesis,
to test our emerging hypotheses directly. which framed and guided the remainderof our data analysis.
To the extent that they contained mea-
sures of relevant constructs, however,
they confirmed findings from our inter- Ouranalysis revealed considerablewithin-firmvariability,over
view data. time and across work groups, in both perspective and out-
Table 1
Racial Composition of Firms and Participants
Support Staff Middle Managers/Professionals Senior Managers
Percent Number Percent Number Percent Number

Firm* in firm in sample in firm in sample in firm in sample
Consulting firm
People of color 63% 7 42% 7 31% 3
White 37% 5 58% 9 69% 6
Financial services firm
People of color 91% 3 41% 10 40% 5
White 9% 1 59% 4 60% 6
Law firmt
People of color 75% 3 60% 3 - -
White 25% 1 40% 2 100% 3
* The total number of employees in the consulting firm was 119; the total number in the financial services firm was
121; the total number in the law firm was 12.
t This sample also included three former program staff members, two people of color and one white.
239/ASQ, June 2001

comes. This variabilityis consistent with the argumentthat
the most appropriateunitof analysis for linkingdiversityper-
spectives with outcomes of interest is the work group.The
variabilityin perspectives we found across work groups with-
in firms providedus with the opportunityto investigate our
developing hypotheses about how work group diversityper-
spectives influencework group functioning.
WORKGROUPPERSPECTIVES ON DIVERSITY
Ouranalysis supportedour argumentthat the perspective
that governed work groups' orientationtoward diversitywas
associated with differentlevels of individualand groupfunc-
tioning.We identifiedthree diversityperspectives that
appearedto have differentimplicationsfor how well people
functioned in their work groups and, therefore, how likely
their work groups were to realizethe benefits of their diversi-
ty: the integration-and-learningperspective, the access-and-
legitimacyperspective, and the discrimination-and-fairness
perspective. Each providesa rationalefor why the work
group should increase its culturaldiversity,yet only the first
was associated with what appearedto be sustainable perfor-
mance gains attributableto diversity.Retrospectivedata from
participantssuggested that work groups' perspectives could
develop and change over time, but, at the time of our data
collection, a single, dominantperspective on diversitypre-
vailed in each group we studied. If there were dissenting
views withinthe group,they came from a small minoritywho
expressed concerns privatelythat certainaspects of the
group's perspective on diversitywere problematic.The pre-
vailingperspective in the group nevertheless shaped mem-
bers' experiences in predictableways. Althoughthere was
within-firmvariabilityin the diversityperspectives work
groups held, each perspective seemed to be best illustrated
in one of the three firms.
Integration-and-LearningPerspective
Accordingto the integration-and-learning perspective on
diversity,the insights, skills, and experiences employees
have developed as members of variousculturalidentity
groups are potentiallyvaluableresources that the work group
can use to rethinkits primarytasks and redefine its markets,
products, strategies, and business practices in ways that will
advance its mission. This perspective linksdiversityto work
processes-the way people do and experience the work-in
a mannerthat makes diversitya resource for learningand
adaptivechange. The integration-and-learning perspective and
the outcomes associated with it were evident in the program
functionof the law firm,which includedthe attorneys and
policyanalysts in the firm,and in the management commit-
tee of the financialservices firm.We focus our descriptionon
the programfunction in the law firm, however, because peo-
ple there were especially articulateabout how and with what
consequences this perspective evolved over the course of
their efforts to diversifytheir workforce, in particular,their
programstaff. Where this perspective was evident in the
financialservices firm, it was associated with the same kinds
of processes and outcomes we observed in the law firm.
240/ASQ, June 2001
Cultural Diversity
The law firm had developed a successful practicein its first

ten years, representinga largelywhite female clientele in
employment-relateddisputes. Nevertheless, in lightof their
mandate to protect and advance the economic rightsand
interests of all low-incomewomen, the firm'sattorneys
viewed their inabilityto attractwomen clients of color as a
significantshortcoming.To address this problem,they decid-
ed to diversifytheir all-whiteprogramstaff. They began by
hiringa Latinaattorneyto head what they called the
"women-of-colorproject."The project'spurpose was to
expand their work into the Latinacommunityand demon-
strate their commitment to advocacy on behalf of all low-
income women. By virtuallyall accounts, however, this
change in staff composition moved them far beyond that
originalgoal. Over the next ten years, they underwenta tran-
sition from a staff composed entirelyof whites to one that
includeda programstaff that was at least half people of
color. More importantly,however, this change in the demo-
graphiccomposition of the programstaff entirelyreshaped
the characterand prioritiesof the firm'swork in unanticipated
ways as members learnedfrom their diversityand integrated
what they had learnedinto the core work of the organization.
Several staff members, both currentand former,described
the change as follows:
Our mission is still the same-the economic empowerment of
women. But our strategies or how we define them have radically
changed from a fairly straight feminist approach. We're still talking
about sexual harassment, comparable worth [TitleVIIcases], those
are the same. But our diversity made us look at the organization's
program and how we had to change the work that we do-the sub-
stantive legal stuff that we do. So now we're looking at minimum
wage, manufacturers' liability.... That's not traditional sex discrimi-
nation, but these are primarilywomen workers who are affected by
these things.
At first, we were like, "[industry name] workers? That's men and

women. Where's the gender discrimination?" And [the Latina attor-
ney] was beating us over the heads with a stick and saying, "Hey,
most of these folks in this industry are women; most of them are
women of color; most of them are non-English-speaking women.
What better place for us to be?" And eventually the staff said,
"Right, you're right, that does make sense. That is a way for us to
go. "
Associated with this transformation in the firm's work was a
shift in its perspective on its program staff's diversity. No
longer was its diversity confined to a particularproject: "Our
women-of-color project became integrated in such a way that
it was no longer this special little program off to the side,"
one programstaff member explained. "Itnow just perme-
ates the whole picture,"added another.Theirnew perspec-
tive on diversity-an integration-and-learning
perspective-
was groundedin the notionthat culturalidentityshapes how
people experience, see, and know the world. Hence, cultural
differences can be a source of insightand skillthat can be
broughtto bear on the organization'score tasks. This discov-
ery enabled staff members to see their diversitynot only as a
resource throughwhich they could gain entree into previous-
ly inaccessible niche markets but, more importantly,as a
241 /ASQ, June 2001
resource from which they could learnnew ways of recon-
ceiving and reconfiguringtheir work as well. As one white
woman attorneyexplained, "[Diversity]means differences in
terms of how you see the issues, who you can work with,
how effective you are, how much you understandwhat's
going on.... There's not a sense of 'you'rejust like me."'
And althoughseveral people spoke to the discomfortthat
often comes with such differences, they also emphasized
the need to look "beyond feeling comfortable ... to the dif-
ferent types of skills people bring."
This perspective on culturaldifferences requiredthat program
staff members place a high value on process-on time spent
exploringtheir differentpoints of view and deliberatingabout
whether and how they should informthe work. Describing
herself as "the process queen," the executive director
stressed the importanceof "learninghow to not be afraidof
the differences, learningabout conflict,and learningto be
willingto go toward it and tryingto talk about hardthings."
Similarly,a former programstaff member explainedthat
"process is critical":
[Therehas to be] a kindof group process of makingsure that

there's the time and a safe situationand that people are gonna be
encouragedto say what they're worriedabout, even if it's not politi-
callycorrect.... Youneed to provide,to whatever degree possible,
permissionfor people to say what's on their mindand struggle
throughthe consequences and innerpersonaldynamicsof saying
those things.... People have to be willingto take risks.Youhave to
be willingto be wrong. It's not something lawyers do easily. I'm not
sure anybodydoes. But lawyers especiallyjust hate to be wrong.
And a bunch of white liberalwomen lawyers hate to be politically
incorrect.
Recognizingthat people from differentculturalbackgrounds
might bringdifferentsets of experiences and skills to work
did not dictate a cultural-identity-based
divisionof labor
among the programstaff. Several people felt strongly,for
example, that one need not be "gay to raise gay issues" nor
"a person of color to raise issues of concern to women of
color."A white attorneyexplainedthat althoughshe could
not be the founderof a Latinaorganizationbegun in her
office, she would work with the group eventually.She talked
about diversityas a learningexperience: "I've learneda lot
about things that just weren't in my background.I don't
mean about salsa or whatever, but about . . . what life experi-
ences are like in other places." As this woman suggests, the
programstaff's diversitywas to serve as a resource on which
all members could draw to expand their knowledge base as
well as their networks. This meant a deep commitment to
educating and learningfrom each other and reflects a central
premise of the integration-and-learning perspective on diver-
sity: while there may be certainactivities at certaintimes that
are best performedby particularpeople because of their cul-
turalidentities,the competitive advantage of a multicultural
workforce lies in the capacityof its members to learnfrom
each other and develop within each other a range of cultural
competencies that they can all then bringto bear on their
work.
242/ASQ, June 2001
Cultural Diversity
As a result, white members of the programstaff had to learn

to take up, on their own, the issues and concerns that might
initiallyhave been raised by their colleagues of color so that
certaintasks did not always fall to one group or another.As
one white employee put it, "It'simportantthat people of
color coming into the organizationdon't see themselves as
coming in and just educating a bunch of white folks; you
have to demonstrate in a realway that you've been educated
when you come back."Virtuallyeveryone, both white and of
color,commented on the personaland professionalgrowth
the staff's diversityhad affordedthem. As one white attorney
reflected, "Ithinkabout things differently.Things I've taken
for granted I can no longertake for granted. My sensitivities
are just different."
To facilitatethis kindof learning,the programstaff had to
organizetheirwork differently.Whereas traditionallya case
would have been staffed by a single attorney,it now would
be staffed by at least two. This enabled people to engage
more easily in the kindof cross-culturallearningand expo-
sure that had become so centralto the way they operated
and, more importantly,demonstrated how, with this perspec-
tive on diversity,their work processes, as well as their work,
were open to change.
Accordingto this perspective, one measures progress in
efforts to diversifyby the degree to which newly represented
groups have the power to change the organizationand tradi-
tionallyrepresented groups are willingto change. The execu-
tive directorof the law firmdescribed her litmustest of how
well an organizationis managingits diversityas how much
change there is in the power structure:
Is the organizationtryingto assimilate people into what already
exists? Or do they want to create something that's differentfrom
what was there before-and maybe not know what that means? If
you want people to be partof an organizationand have ownership
in the organizationthen they have to have power and some control.
I thinkthe way that we successfully did it here was in terms of the
program.The power and who is in controlof our programhas really
changed.... Youcan't assume that what's traditionallybeen done is
the rightway to go.
Access-and-Legitimacy Perspective
An access-and-legitimacyperspective on diversityis based in
a recognitionthat the organization'smarketsand constituen-
cies are culturallydiverse. It therefore behooves the organiza-
tion to match that diversityin parts of its own workforceas a
way of gainingaccess to and legitimacywith those markets
and constituent groups. Workgroups in which this perspec-
tive prevailsuse their diversityonly at the margins,to con-
nect with a more diverse market;they do not incorporatethe
culturalcompetencies of their diverse workforces into their
core functions. This perspective constitutes the rationale
behindthe now popularlytouted business case for diversity
(Coxand Blake, 1991). The access-and-legitimacyperspective
guided the law firm's initialefforts to diversifyits program
staff and continuedto providethe rationalefor the cultural
composition of its administrativeand management staff. It
was most vivid,however, in parts of the financialservices
243/ASQ, June 2001
firm, which we focus on here for our description. In each
instance it was associated with similar kinds of outcomes.
In the financial services firm, the access-and-legitimacy per-
spective was especially evident in the diversification that
occurred in two departments of the Sales Division-Retail
Operations and External Deposits. Retail Operations was
responsible for servicing the banking needs of a predomi-
nantly black, working-class, urban clientele to whom the firm
marketed its services locally, in the surrounding neighbor-
hood. External Deposits was responsible for servicing the
banking needs of a predominantly white, affluent clientele to
whom the firm marketed its services nationally. Mirroringthe
racial and class composition of these markets were the pre-
dominantly black, working-class employees who staffed
Retail Operations and the predominantly white, middle- and
upper-middle-class employees who staffed External
Deposits. This staffing pattern characterized these depart-
ments from the lowest- to the highest-ranking employees.
Members of both Retail Operations and External Deposits
readily acknowledged the importance of their racial make-up
as a way of gaining access to and legitimacy with their
respective clientele. Explaining the role of the black staff in
Retail Operations, the white manager of External Deposits
explained:
If [the firm]were all white, our relationshipswith the community

would be extremely strained.And our retaildeposit base would be
very much threatened. [Thecommunity]would be saying, "What
are these white people doing runninga bankin the middleof our
community?"And they'd be right.We've operated in blackcommu-
nities for 20 years. Ifwe aren'tfullyintegratedourselves, it's pretty
hypocritical.
This manager's black counterpart in Retail Operations com-
mented similarly:
Formanagementto come into a blackneighborhoodand undertake

[this mission],they would be remiss not to thinkwe have to get
some differentcolor people in here to help us do this. It would give
the communitya level of comfortthat there are people in the orga-
nizationwho actuallyknow how to relateto . .. the people that are
in the neighborhood,and what they actuallyfeel, and, you know,
how they actuallycommunicatewith one another,and those kinds
of things.... I mean, we are in the heartof the blackcommunity.
This perspective provided a similar though less elaborate
rationale for the predominantly white staff in External
Deposits. Several people commented that External Deposits'
white clientele were probably "more comfortable" with the
white staff who served them. One staff member summa-
rized the importance of having both white and black staff:
I thinkif we were all black,we'd have a lot of obstacles. We

wouldn'thave access to a lot of the resources that we do. Minority-
owned banksthat are almost exclusively minorityhave reallystrug-
gled because they're not as connected to those [white-controlled]
resources. I thinkit could still be done, but it would be a harder
task. If we were all white, I thinkwe'd be in as bad or worse shape
[as if we were all black],just because of the discomfortwith the
community,or not being able to relateto the borrowersor stand in
their shoes so to speak.
244/ASQ, June 2001
Cultural Diversity
Despite this apparentsymmetry, however, the access-and-

legitimacyperspective in fact defined a much more circum-
scribed role for blacks than for whites, limitingthe contribu-
tions of blacks to just that-access and legitimacy-whereas
the contributionsof whites were more widely evident. For
example, a white employee in ExternalDeposits described
the overallcultureof the firmas much more consistent with
the cultureof her predominantlywhite departmentthan with
the cultureof RetailOperations,which was predominantly
black:
. . . if you perform and exceed expectations, regardless of color, you

are acknowledged and recognized.... The problem is that what is
expected of senior management here has a cultural bias towards
whites. And ... if you're in that cultural modus, you don't under-
stand why it's exclusionary.... Everyone is expected to work a lot
of hours. There is this emphasis on perfectionism, this emphasis on
sort of intellectual discussion and debate. People are very, very mis-
sion-driven. And that's not to say that African Americans aren't also
able to do all that. But because of historical racial issues they have
been limited.... So there aren't a lot of people from the neighbor-
hood that would be senior management level, and there are an
awful lot who would be in those low-paying, pretty routine, white-
collar jobs.
Hence, althoughculturalidentityin these two departments
was clearlya legitimate resource to be used in service of the
Sales Division'swork, the access-and-legitimacyperspective
provideda relativelynarrowdefinitionof the value blackcul-
turalidentityhad to offer, relativeto white culturalidentity.
Blacks in RetailOperationswere invitedto use their cultural
identity,but only at the boundariesbetween the organization
and its blackmarket.By contrast, there was a perception
among employees in these departmentsthat whites' cultural
identityshaped how the Sales Divisionfunctioned more
broadly,with middle-and upper-middle-classwhite culturein
particulardictatingthe work norms and standardsmost
valued.
Withthe access-and-legitimacyperspective, one measures
progress in diversificationefforts by whether there is suffi-
cient representationeither in those boundarypositions or in
visible positions that would enhance the legitimacyof the
organizationfrom the perspective of its outside markets.
Althoughthis raised the question of how many whites would
be too many, as well as the converse, how many blacks
would be enough, this perspective providedno clear
answers. Rather,as one participantsurmised,
It may be a function of the inner workings of the manager's mind
that it's time for me to hire a minority or something. And that's legit-
imate in this organization. While it seems unfair that maybe the
most qualified person or the best person for the job might not get
that position, maybe the best qualified person isn't the right person
for the organization, and maybe it's time to hire a minority.
Discrimination-and-Fairness Perspective
The discrimination-and-fairness perspective is characterized
by a belief in a culturallydiverse workforceas a moralimper-
ative to ensure justice and the fairtreatment of all members
of society. It focuses diversificationefforts on providingequal
245/ASQ, June 2001
opportunitiesin hiringand promotion,suppressing prejudicial
attitudes, and eliminatingdiscrimination.A culturallydiverse
work group,therefore, is meant to be evidence of just and
fairtreatment of employees. In contrast to the previoustwo
perspectives, in the discrimination-and-fairness perspective
there is no instrumentallinkbetween diversityand the
group'swork. Workgroups in the consultingfirm provided
the best illustrationof this perspective and the processes and
outcomes associated with it. Infact, there was very littleevi-
dence of any other perspective in the consultingfirm,and
this perspective was largelyabsent in the other two firms we
studied.
Consultingfirmemployees expressed this perspective most
clearlyin their statements about why the firm'saffirmative
action programwas important.One white manager
explained, "Thefirmcreated a communitythat is diverse
based on a very clear sense that there should be equalityand
justice." Similarly,an AfricanAmericanmanagerdescribed
the firm'sphilosophyas "everyone being equal or justice for
all, being fairin regardsto hiring,treatingstaff the same." A
white managerelaboratedas follows:
I think[the firm],from my vantage point, has made tremendous

progress in its commitmentto buildboth a just society inside, as
well as a just society outside the organization..... I thinkthe organi-
zation has committed itself to restructuringits population,its per-
sonnel makeup,in orderto rightsome of the wrongs caused by
racism and sexism in our society.... And the cost has been to turn
down a lot of good, qualifiedwhite people for jobs, which we've
had to do in orderto make this programwork. There'ssimply no
way around it.... The other side of it is that the people of color in
this organizationhave added immensely to it, I believe.... They
have enrichedthe organization;they have helped us live up to our
ideals of equalityand justice.
Accordingto this perspective, culturaldiversity,as an end in
itself, was not to influencethe organization'swork in any fun-
damentalway. Althoughthe firm established two commit-
tees whose mandate was to "infuse the firm'sactivities"
with a "feminist"and "racial"perspective, respectively,in
practice,these committees had virtuallyno impacton the
firm'swork. Instead, consistent with their discrimination-and-
fairness perspective, they served a policingand advocacy
function, scrutinizingthe firm'streatment of women and peo-
ple of color for evidence of sexism and racism and advocat-
ing on behalf of those groups when they deemed necessary.
To the extent that these committees did influencethe firm's
program-relatedwork, many employees were critical:"These
committees tend to sometimes have more leverage, more
power than perhapsthey ought to have in decision making,"
one white manager lamented. "Theyare sometimes allowed
to make interventionsand judgments of certain programs
based on their [politicalclout] ratherthan on their knowledge
and information."Anotherrepudiatedany attempts the com-
mittees might make to influence programmaticdecisions or
directions "on racialgrounds,"arguingthat they should have
no role in the "normaldecision-makingprocess of the organi-
zation."
246/ASQ, June 2001
Cultural Diversity
Many members of the organization,both white and of color,

pridedthemselves on being blindto culturaldifferences.
Althougheach group questioned the other's abilityto uphold
this virtue, members of both groups equated the organiza-
tion's philosophyof justice with its commitment to the notion
that "everyone is the same," "everyone is just a human
being here; it doesn't matterwhat color he is." As one
AfricanAmericanclaimed, "Idon't see people in color, I treat
them all the same." Consistent with this insistence that
everyone is the same, there were at least two norms that
operated to suppress any differences that did exist. The first
was to avoid conflictwherever possible. Many reportedhav-
ing received a clear and consistent message from manage-
ment that to express conflictwas "potentiallydangerous,"as
it "mightdo more damage than good." The second was a
norm requiringassimilationto a white culturalstandard.As
one white managerexplained,while the goal was to be
"entirelyrace blind"in personnel decisions, the "expectation
is still that people will speak in normalEnglishand write the
way white people write." Althoughsome people complained
about management's enforcement of these norms, they saw
no inconsistency between their commitment to "color-blind-
ness" and their concerns that these norms were oppressive.
Similarly,the small minorityof professionalstaff in the pro-
gram areas who felt that incorporatingrelevant,race-based
insights into their work was importantnevertheless tended
to espouse many of the norms and values associated with
the discrimination-and-fairness perspective, which mitigated
against their being able to do so.
Accordingto the discrimination-and-fairnessperspective, one
measures progress in diversityby how well a work group
achieves its recruitmentand retentiongoals. As one African
Americanexecutive explained, "a systematic monitoringof
numbers"was a key indicatorof whether or not "thingsare
going along smoothly."A Latinamanagerexpressed a similar
sentiment about the importanceof numbers: "A significant
numberof people of color is a sign of something good about
the organization."
Table2 summarizes the characteristicsof the three work
group perspectives on diversity.Eachof the three different
sets of expectations and beliefs that people held about cul-
turaldiversityand its role at work shaped individualexperi-
ences and group processes in differentways, which had
implicationsfor individualand groupfunctioning.
INTERMEDIATE GROUPOUTCOMES
Quality of Intergroup Relations
The integration-and-learning perspective is predicatedon the
notionthat a diverse group of people comes together for the
express purpose of learningfrom one another how best to
achieve the work group's mission, but that often meant ten-
sion-filleddiscussions in which people struggled to hear each
other's points of view before resolvinghow to proceed with
the work. As one white programstaff member in the law
firmexplained, "Cross-racediscussions occur with some fre-
quency and sometimes with some tension, because it's hard.
There are real differences here. And that stuff is being dis-
247/ASQ, June 2001
Table2
Summary of WorkGroup Diversity Perspectives
Characterizationof Discrimination-
perspective Integration-and-learning Access-and-legitimacy and-fairness
Rationalefor diversifying To informand enhance To gain access to and To ensure justice and
core work and work legitimacywith diverse equalityand eliminate
processes marketsand clients discrimination
Valueof culturalidentity High;a resourcefor Moderate;a resourceonly Low;it is a basis for

learning,change, and at the interfacebetween unjustdiscrimination;
renewal;should integrate organizationand mar- should assimilateto
culturaldifferences into kets/clients;should differ- dominantwhite culture
core work and work entiate to gain access and
processes as appropriate legitimacy;otherwise,
assimilateto dominant
white culture
Connectionbetween Direct;incorporated Indirect;race-based Limited;norms againsta
culturaldiversityand work throughoutthe work divisionof laborto connection
enhance access and legiti-
macy
Indicatorsof progress Increasedrepresentation Increasedrepresentation Increasedrepresentation
of traditionally
underrepre- of traditionally
under- of traditionally
under-
sented groupsthat have representedgroups, representedgroups
power to change organi- especially in boundaryor
zation;process and prod- visible positions
uct innovation;shared
sense that culturaldiversi-
ty is resourcefor learning
cussed. It's not hidden undera rock."One former attorneyof

color described her particularexperience of workingthrough
differences in point of view with the executive director:
I would take on the executive director, and she and I would go at it.
But . .. we'd really hear each other, and I think we learned a lot
from one another. And you can come at her. And she can come
back at you with reason, using the history of the organization, why
that won't work.... And I'd remind her that the point of the organi-
zation was to let go of that history and only hold on to it where it
makes sense.... I would ... just hang in there until I was sure that
she was really rejecting an idea or my client on its merits. Not
because it was new or unsettling. And sometimes she'd really con-
vince me that the rejection was based on merit. And sometimes,
there were some things I should have let go earlier I'm sure.
Certainkindsof problemswere inevitable,and they seemed
to result from the fact that the programstaff were not
immune to the way race relationswere structuredin the larg-
er culture.Two kindsof tensions in particulararose in the
programstaff's race relationsas a result, and, althoughwe
viewed each as stemming from the difficultyof livingup to
the vision of diversityset forth, the kindsof relationshipsand
processes the vision encouraged were precisely the mecha-
nisms that eased those tensions and helped people work
toward resolution.Hence, the perspective seemed to contain
a self-correctingmechanism that both reinforcedthe vision
and maintainedits usefulness to the organization.
The first tension concerned the twin problems of burnoutfor
the attorneys of color,who sometimes felt called upon to do
of
more than their fairshare of the work, and marginalization
248/ASQ, June 2001
Cultural Diversity
white attorneys, who sometimes felt less centralto the

firm'swork as a result. People attributedboth of these prob-
lems to the "realityof the world out there," yet they seemed
manageable largelybecause people were able to discuss
them. As one white attorneyexplained, "we're prettyopen
about talkingabout those things here, so it's not like this
unspoken thing." She elaborated:
Likesometimes people are puttingtogether panels and for good

reasons they want a diverse panel. So I'llbe the last one they'd ask,
even if I'mthe person who's done the most work in the area,
because they'd preferto have [one of the women of color].And
then we would talkabout how it would be strange that organiza-
tions that I work with would call up [a blackattorney]and ask her to
be on a panel. So that both put a burdenon her and kindof made
me feel strange about being excluded. But it was something we
understoodbecause we thought the role model and the diversity
aspect of the panel was an importantthing to do.
The second kindof tension was the disappointmentevery-
one felt when people's failureto use their own or to seek
others' cross-culturalknowledge threatened to compromise
the programstaff's effectiveness. One such incident
occurredduringa staff meeting we observed, which the
executive directorafterwardtold us was "a very good view
of what goes on here-people engaging in what is not
always the easiest conversationand being reallywillingto
take the time to challenge each other and to be educated by
each other."A local Latinocommunitygroup had invitedthe
firmto join in a fund-raisingevent involvinga Latinotheater
group.The executive and associate directors,unawareof the
importanceof the group in the community,decided to
decline the invitation,without consultingprogramstaff, on
the grounds that it would interferewith a largerfund-raising
event alreadyscheduled. When one of the Latinaprogram
staff was informedof the decision, she felt that the directors'
lack of culturalknowledge had led them to a hasty and costly
decision, and she placed it on the agenda for the next staff
meeting. At that staff meeting, the Latinas,across hierarchi-
cal lines, expressed unifieddisagreement with the decision,
describingthe event as "an importantvehicle for us to do our
work with this community."The staff seemed to have diffi-
culty resolvingthe conflict untileveryone was able to see the
decision as more properlyprogram-relatedthan administra-
tive. The administrativefunction in the firm had yet to devel-
op an integration-and-learning perspective on diversity.With
no clear sense of how racialdiversitymight enhance that
function, managers had not sought and were initiallyresistant
to hearingdifferentperspectives on the usefulness of the
event. As soon as the event was successfully recast as out-
reach, however, a program-relatedactivity,they were able to
see the relevance of race and the importanceof hearinga
specificallynonwhite perspective.
Ourdirect queries about the qualityof race relationsin the
Sales Divisionof the financialservices firm, in which employ-
ees held an access-and-legitimacyperspective on their diver-
sity, revealed few problems and a general sense that black
and white employees experienced littletension in their cross-
race interactions.As one white participantsaid, "It'snot to
249/ASQ, June 2001
say there's never any discomfort, but I've been very sur-
prised-I've never runacross an uncomfortablesituation
here." Similarly,a blackemployee described interactions
"between everyone" as "reallygood" and a general sense
that people ask questions about those from other cultures in
a way that does not offend. "People are different,"another
explained, "butwhen the need arises they can work togeth-
er." The dynamics withinthe Sales Divisionbetween Retail
Operationsand ExternalDeposits, however, revealed a more
complicatedstory.
The racialdifferentiationbetween these two departments,
both in their staffing and in their clientele, resulted quite
clearlyin a two-tiered system in which the white department
received better treatment and higherstatus relativeto its
blackcounterpart.Participantshad much to say about this,
and what they said did not reflect the sanguine sentiments
we heardwhen we asked about race relationsmore directly.
Yet there were unequivocalracialovertones, as well as
explicitreferences to race, in their discussions of the relation-
ship between these two departments.And despite the sym-
metry between blacks and whites in positions of authority,
the relationshipbetween these two departments seemed to
reproducethe asymmetric divisionof power and status with-
in the Sales Divisionthat characterizessocietal race relations
more generally.
Most people agreed that there were very few differences
between the kinds of tasks the two departments performed.
Nevertheless, more than one participantreferredto the fact
that there were "two banks"withinthe firm:RetailOpera-
tions and ExternalDeposits. One participantfrom External
Deposits explainedthat, in her view, this had come about
because the previous manager,who had an ambitiousagen-
da and insisted on providingthe highest qualityservices,
duplicatedfunctions that alreadyexisted in Retailwhenever
she encountered a level of qualitythat she judged as too low:
And so you had this sort of cracker-jackgroupof people who

worked for her ... that were in the absolute perfect job for the sort
of white, smart, dedicated, loyalworkaholic.And not the perfect job
for the sort of black,hard-working,needs a salary,will do a good
job, but not that kindof worker... and there was absolutelyno
time for people who wanted a 9-to-5 job.
This status differentialbetween the two departments and the
resentments it fostered were palpable.There was a percep-
tion among those in RetailOperationsthat management
looked more favorablyon ExternalDeposits, that External
Deposits got "special privileges"and was "more presti-
gious," and that people there were paid more "because
they're white, even though the work is the same." By con-
trast, participantsin both departments referredto Retail
Operationsas "the other side" of the firm, "the darkside."
One blackparticipant,now an officer in RetailOperations,
described an experience he had when he was the lone black
member of ExternalDeposits several years earlier.This expe-
rience illustrateshow racialstereotypes shaped interactions
between blacks and whites in a mannerthat may have rein-
250/ASQ, June 2001
Cultural Diversity
forced, at least for some, the appropriatenessof the racial

divisionof laborbetween the two departments:
We were at a staff meeting talkingabout the problemswe were
havingas a departmenttryingto be all things to all people. And I
rememberedthis thing my boss had said about a year earlierthat
we have to select the battles that we want to fight, and I took that
to mean that we have to decide strategicallywhat we will pursue
and what we won't pursue. And I just happenedto thinkabout that
quote, and so I said, "Ithinkthat we ought to be realcarefulnot to
bite off more than we can chew." . . . I got a response where the
person said, "Well,what do you propose? We do nothing?"So I
saw rightthen and there that I was misunderstood.I said, "No, of
course not. I'msaying that we need to select the battles we want
to fight and fight those." . . . And being pretty new to the organiza-
tion then, I felt that it wasn't the righttime for me to be forthright
about what I meant.... [Wihena white man disagrees, he's being
strong. He's being taken with respect. When a blackman disagrees,
he's being negative and whiny, militantand kindof like MalcolmX.
So you have to be reallycarefulabout how you walk that line so
that you don't get labeled and you don't sabotage your career.
Inthis story, the white employee interpretedher blackcol-
league's comments as consistent with the view that blacks
were not a good culturalfit with the aggressive, workaholic
norms of this department.Concernedthat his objections to
her interpretationmight reinforceadditionalnegative racial
stereotypes about him, the blackcolleague remainedsilent.
Thus, race-basedstereotypes importedfrom the largercul-
ture shaped these employees' interpersonalinteractionsin a
way that reinforceda view of this departmentas appropriate-
ly culturallywhite and elite.
This particularmanifestationof the access-and-legitimacyper-
spective, in which two raciallysegregated, parallelentities
were formed to service differentracialand economic seg-
ments of the market,fostered a good deal of resentment and
competitiveness between the two departments,which was
often expressed explicitlyin racialterms. One participant
described the senior officer in charge of RetailOperationsas
"a little bit resentful when his territoryis encroached on by
white people [i.e., ExternalDeposits]." Anotherdescribed the
"culturalbarriers"to integratingthe two departments, or
even to fostering a more cooperative spirit,which might
replace the "distrust"that seemed to characterizetheir rela-
tionships. Stillanotherattributed"the tensions between the
two sides" to "the logistics, the race, the professionalmix,
and just the natureof how the departmentsare compiled."
Hence, althoughthese participantsoften spoke positivelyof
race relationsin the firm,the racialsegregation inside the
Sales Divisionmirroredhierarchicalrace relationsand racial
tensions in the wider culture.
Participants'descriptionsof race relationsin the consulting
firm, in which all work groups held a discrimination-and-fair-
ness perspective on diversity,were nearlyunanimouslynega-
tive. People of all races described relationshipsbetween
white and AfricanAmericanemployees, who made up the
majorityof the nonwhite staff, as "tense," "cynical,""hos-
tile," and "distrustful,"and described their own feelings as
"disappointed," "hopeless, " "helpless," and "powerless."
251/ASQ, June 2001
Differences in people's characterizationsof the problemtend-
ed to fall along a combinationof racialand hierarchicallines.
Blackexecutives and whites across the hierarchytended to
agree that employees of color were too quickto bring
charges of racismagainst white people. One AfricanAmeri-
can executive was frustratedby her observationthat any
time management met to discuss a problemconcerningan
employee of color "people [of color]are up in arms and say-
ing it's racism."A white managervoiced the same senti-
ment: "Whenevera person of color loses his job, there is an
immediate perceptionthat the decision to terminatethe
employee was a racistone."
At the same time, there was a widely shared fear among
whites that any form of conflict or confrontation,especially if
perceived as instigated by a white person in relationto a per-
son of color,would automaticallyimplicatethe white person
as racist. One white managerexplained, "Iwould find it diffi-
cult to challenge a person of color because I like to thinkof
myself as not being prejudicedand would hate to be said to
be prejudiced."Anotherdescribed the mountingpressure he
felt, as a white male, "to show the correct attitudes towards
race relations,"which he believed meant he was expected to
agree with everythingpeople of color said: "Thereis a level
of psychologicalintimidation;you don't question decisions or
performance."As a result, white managers felt it had
become "increasinglydifficultfor supervisorsto providefirm,
fair,constructivesupervisionto people of color,who are
prone to charge racism if they are criticized."Where he
"once felt that the firm'scommitments to fight racismwere
honorable,"one white manager now felt they were "getting
to the point where we're not just fightingracism;we're set-
ting up other standardsfor letting people get away with
whatever bullshitthey want to get away with."
On the flip side, middle-and lower-levelstaff of color resent-
ed their white colleagues' conflict-avoidantstance and fears
of confrontation,as the cynicaltone of the following com-
ment illustrates:"Thereis a real sense on the partof some
white people that whatever they're going to do they're going
to get in trouble.They'regoing to get accused of being a
racistwhich is almost the worst possible thing that could
happen to a white person here, short of dismemberment."
Many people of color argued that by keeping them from
receiving honest feedback and getting the kindof supervision
they deserved, this stance was itself racist.They felt that, as
a result, they never knew when the "hammermay fall,"
when "the trap door will drop." In a recent incident,a black
woman, who had been an employee at the firmfor ten years,
was summarilyfired for poor work performanceand required
to vacate the premises that afternoon.Though many conced-
ed that her performancewas problematic,people of color
nevertheless organizeda formalprotest of management's
failureto "confronther [earlyon] with her poor performance
and treat her as if she were a normal,equal person." In
another incident,many employees of color signed a petition
to protest the disciplinaryaction taken against a black
employee who was held responsible for money stolen from
his department,arguingthat the theft had occurredonly
252/ASQ, June 2001
Cultural Diversity
because inadequatesupervisionhad prevented him from tak-

ing the necessary precautions.As one blackexecutive
explained,these kindsof events "confirmedpeople's worst
fears about the insensitivityof management to the well-being
of employees of color."Both the white staff and the black
executives in the firmemphasized privatelythe complicityof
people of color in these incidents. One AfricanAmerican
executive lamented that people of color,once fired or disci-
plined, become "purerthan snow" and often fail to recognize
that their own behavior"is not always so desirable."Another
was more cynical,arguingthat "blacksknow they can milk
these [white] people because they [white people] are so
afraidof confrontation."
When the discipliningsupervisorin such incidents was a per-
son of color,other people of color often interpretedhis or her
actions as the result of manipulationand corruptionby white
management. Several blackparticipantsdescribed times
when they believed whites had purposelyused blackman-
agers to handle problems with blackstaff to avoid having
their own confrontations.Two invokeda plantationmetaphor
to capturethis dynamic,in which the "owners" (executives)
used the "house niggers" (blackmanagers)to look after the
"fieldniggers" (blacksupportand technical staff). Interesting-
ly, the ultimateoppressors in this metaphor-the "own-
ers"-were blackas well as white in this firm.This is consis-
tent both with the similarityin views we found between
blackexecutives, on the one hand, and whites, on the other,
and with the perceptionmany blackemployees shared that
blackexecutives "must have sold out in some way" and did
not identifywith the blacks they supervised.
Finally,we were struck by the fact that most of the public
debates about "racialincidents"at this firm centered on the
treatment of people of color ratherthan on the work-related
problemsthat instigatedthat treatment. Forexample, many
people, both white and black,believed that the woman who
was fired in the incidentabove had routinelyand inappropri-
ately biased affirmativeaction searches in favor of candidates
of color in her role as an administratorin the Affirmative
Action department.And the man who was disciplinedfor the
theft rana functionwithin a departmentthat had long been
losing money for the firmthroughinefficiencies and poor
management. Neitherthe qualityof her performance,nor the
efficiency of his department,however, was centralto the
publicdebates that ensued, leaving importantquestions
about these aspects of their work unanswered.
Thatthe tensions in race relationsin this firmwould be
played out aroundcharges and counterchargesof racismand
intimidationseemed ironicin lightof work groups' diversity
perspective in this firm,which emphasized fairtreatment as
its primarygoal. Yet because it providedonly a fairness-
unfairnesslens for viewing differences in point of view that
fell, for whatever reasons, along race lines, this perspective
seemed to foster the very kindsof tensions it sought to
quell. Differences in work-relatedpoints of view were seen
as a problemof primarilymoraland ethical dimensions. This
in turn limitedthe kindof discourse in which people could
engage, especially across races. Finally,the perceptionthat
2531ASQ,June 2001
upper-level blacks identified more with whites in the firm
than with blacks fueled tensions between upper- and lower-
level people of color, mitigating against constructive intra-
group relations as well.
Feeling Valued and Respected
Employees in work groups that held an integration-and-learn-
ing perspective on diversity reported feeling valued and
respected by their colleagues. This was the case to a person
for both current and past program staff in the law firm,
where there was a sense that the firm "placed a value on
the whole person." As one attorney of color put it, "The
assumption about you is that you are competent." Other pro-
gram staff of color corroborated this view. One said, "There
is a lot of support for me to achieve. They really support and
respect their staff of color in a way that I have not seen at
other women's public interest law firms." To the extent that
white people reported feeling marginalized at times from the
central work of the organization, they also reported that "it
isn't so bad." As one woman explained, "it doesn't consume
me in the way that I think it would if I felt out of place here
and questioned whether the organization really wanted me. I
don't feel like that. I feel like there's enough support, and I
have enough self-confidence about my role here that it
doesn't consume me."
There was also a general feeling of well-being and a sense of
having the respect of one's colleagues among employees of
both Retail Operations and External Deposits in the Sales
Division of the financial services firm, where an access-and-
legitimacy perspective prevailed. "I get appreciation here,"
explained one black participant. "People always check in, and
it makes me feel warm inside. It's nice to know someone is
recognizing what you do; and what you do, no matter how
small, makes a difference." Another black participant said, "I
talk to these individuals as people, regular people, and they
talk to me as a regular person, not like I belong to a particular
racial group." In a similar vein, other black participants felt
that "most dismissals have been legitimate" and that "if you
do your job well, you'll be recognized and promoted for it."
As with race relations, however, these accounts of how peo-
ple felt and were treated as individuals in their interpersonal
interactions with others did not square with many of the
things they said about how they felt and were treated as
members of their respective departments. Whites in External
Deposits had a clear sense of their privilege and the value
they brought to the firm. Blacks in Retail Operations, howev-
er, were less sure about where they stood. As one black offi-
cer in Retail Operations said, "the jury is still out." He
explained,
One of the things that I take a measure of pridein is the fact that
we can all live and work together. And that's OK.But I thinkwhere
sometimes the problemcomes in is in the divisionof the duties.
Youknow, how do you perceive me? Do you perceive me as some-
one who bringssomething to the table, who is a decision maker?
Someone who understandsour customer base and whose thoughts
should be taken seriously?Or do you see me as someone who is
good at operationallymakingthings work and makingsure that the
paperworkis together and makingsure that the files are in order
254/ASQ, June 2001
CulturalDiversity
and making sure that the report is complete and typed and photo-
copied and all that stuff?
Althoughmany described opportunitiesfor promotionregard-
less of race, the divisionof laborin the Sales Division,which
followed directlyfrom its access-and-legitimacyperspective
on diversity,again made it clear to members that there were
two tracks-one for whites and one for blacks. Infact, when
one senior blackofficer on the retailside of the firm realized
that he had no blackmale officers, he "pulled[the lone black
member of ExternalDeposits] out of there and made him an
officer over on the retailside," with a sense that his career
would otherwise have stagnated. When asked about the
challenges and opportunitiesaffordedby a diverse workforce,
this new officer in RetailOperationsdescribed the difficulties
he had faced in ExternalDeposits in getting recognizedfor
his contributionto doublingthe department'sportfolioin two
years: only the two whites headingthe departmentwere pro-
moted. When he finallyreceived his own promotiontwo
years later,it was on the retailside, where his supervisor
more easily recognizedand more readilyrewardedhis talents
and skills. He accepted it with gratitudeand excitement at
the opportunitiesthat lay ahead for him but nevertheless
voiced his concerns about the lower status his new depart-
mental affiliationnow conferred.Thus, the message about
the degree to which people felt valued and respected in
these two departmentswas a complicatedone. Although
uniformlypositive for the whites in ExternalDeposits, the
experience was mixed among blacks.
Inwork groups holdinga discrimination-and-fairness perspec-
tive on their diversity,people of color reportedmore directly
negative experiences in this regard.In the consultingfirm,
every one of the programand support staff members of color
we interviewedreportedfeeling undermined,devalued, or
disrespected in one way or another.The sense of having
been denied honest, trustworthyfeedback, for example,
which led to a perceptionof standardsas ambiguous and
management as capricious,was the source of these feelings
for many. One blacksupport staff member felt that incidents
such as the abruptfiringof her blackcolleague sent a clear
message: "We are not going to make an attempt to orient
ourselves to you or deal with you likeyou are a woman or
intelligentbeing, but when we get tired of you we are going
to get ridof you however we decide."
It was the belief that their competence was underestimated
or overlooked, however, that producedby far the greatest
sense of injuryfor most of the people of color we inter-
viewed. They described being passed over for jobs they felt
more qualifiedto do than the white candidates who were
ultimatelyhired,ignoredwhen they felt they had knowledge
or skills to offer, and presumed automaticallyto lackthe skills
requiredto do theirjobs competently. One blacksupportstaff
member observed, "There'sjust no way that you can be
blackand just know what you're talkingabout or be able to
learnsomething well enough for them to say, 'go ahead, try
it, and we'll see how it works."' Anotherexplained, "There's
a tendency to put more credence in what is said by white
people, not to act on something, till it's confirmedby a white
2551ASQ,June 2001
voice." A Latina who worked on the program staff described
her experience with lack of respect: "I find to this day that
I'm treated with condescension on issues that I may know
more about than they do.... Until [white people] discover [an
idea], until they express it with their own words and their
own style, it's as if it doesn't exist."
Many shared the sense of having either to be white or to act
white to be taken seriously. For example, several attributed
what they perceived to be the unfair discipline of the black
man held responsible for the stolen money to the fact that he
"is black-his attire, his mannerisms-he has a street style. I
don't think they can really see past that." As one Latina
explained, "A lot of the tensions have to do with a difficulty
in recognizing that the habits, the ways of doing things have
been set by white people. And there hasn't been enough
recognition that just to include people of color isn't really
enough." Because of their color-blind ideology, however,
racial differences were taboo subjects for discussion, and it
was therefore illegitimate to recognize, solicit, or offer work-
related perspectives that were informed by differences in
people's cultural backgrounds. A number of the participants
of color also described feeling "depressed" and "dispirited"
at what they felt was the "paternalistic" or "patronizing"
attitude toward people of color generally and themselves in
particular.About the white program staff members who do
economic development in Africa, for example, one black
manager said, "They treat black people like they're little
pygmy children."
The paternalism that staff of color perceived in their white
colleagues' attitudes toward them appeared to stem at least
in part from whites' belief that the firm should uphold its
moral commitment to affirmative action, even if it meant low-
ering standards for employees of color. One white manager
explained that he was "leaning over backward to be gener-
ous and fair and understanding." In doing so, he felt it was
incumbent upon him to excuse staff of color for problems
like tardiness, recognizing "that it may be far easier for me
given my particularcircumstances, living in the suburbs, to
be able to maintain a schedule than it is for one with multiple
pressures of being black and inner city." Contrary to this
man's intentions, it was precisely this kind of charitable view
that many blacks in the firm resented. It is consistent with a
discrimination-and-fairness perspective on diversity in which
whites interpret and respond to their perceptions of cultural
differences within a moral frame: blacks were to be forgiven
for their deviations from (white cultural) norms of acceptable
behavior, as these deviations were merely understandable
reactions to the unjust circumstances of their lives.
We heard comparatively little from black executives or from
whites in any position about the ways in which they might
have felt devalued in the organization. Black executives tend-
ed to comment on how blacks lower down felt devalued but
said little about their own feelings in this regard. This is con-
sistent with the fact that they were generally aligned with
their white counterparts in their perceptions of the firm and
its problems. And although one white male described feeling
"denigrated" for being perceived as "not living up to the
256/ASQ, June 2001
Cultural Diversity
affirmativeaction goals of the firm,"whites did not register

complaintsabout the level of respect accorded to them.
Significance of CulturalIdentity
Consistent with the integration-and-learning perspective's
emphasis on culturalidentityas a potentialsource of insight
and skill,both currentand past programstaff of color in the
law firmdescribed their racialgroup membershipas a signifi-
cant factor in shaping how they approachedand carriedout
their work. One Asian Americanattorneyexplained, "I have a
differentperspective on the work because I'ma woman of
color,and I am interested in cases that, for example, would
open doors to women of color that have traditionallybeen
open only to white women. A white woman is naturallyless
likelyto consider those cases." Programstaff of color also
routinelyrelated stories about how their culturalknowledge
and skills enhanced their abilityto do their work by, for exam-
ple, helpingthem to establish rapportwith clients. One Latina
described how she had convinced a reluctantMexican
woman, who was a key witness in a case, to testify:
Itwas partlythat I spoke the language, but I don't thinkit could
have happenedwith an Anglowho spoke Spanish, because it had
so much to do with understandingwhat was going on in this
woman's mind.And being able to anticipateand just plug into what
was happeningwith her.... It was a tense situation,but I was not
afraidof heranger.
White programstaff also described their racialidentityas
havinga significantimpacton them at work, but in different
ways from their colleagues of color.Whites did not see their
race as a source of skillor insight into their work; neverthe-
less, they were both aware of and articulateabout how being
white influencedthem. "Ithinkthat all of us who are white
here do thinkabout being white," one attorneyexplained.
Some spoke of the opportunitiesbeing white affordedthem
at work. Because of "people's racism,"one white woman
explained, "it'sprobablyeasier being white in settings that
are often predominantlywhite." She had observed, for exam-
ple, that in meetings outside the firm, lawyers would immedi-
ately assume that she was the lawyerand that her Asian
Americancolleague was not, when the reverse was true; she
attributedthis to the greater authorityand status they auto-
maticallyattributedto her as a white person. A numberof
whites also commented on how diversityin the program
staff, in particular,moving the women-of-colorprojectfrom
the peripheryto the center of their work, had affected their
own sense of what it means to be white. One white attorney
felt that it had changed the way she thought about herself as
a white middle-classwoman and forced her to examine her
own racism and stereotypes. Anothercommented on how
diversifyingthe staff as they had had made her "less defen-
sive" about being white because race issues were open to
discussion. She explained, "Ithinkbefore the change [in
racialcomposition]if you'd asked me these [interview]ques-
tions I .. . [think] I would have felt more defensive. Like 'Oh
God, she's tryingto find out if I reallyam a racistor some-
thing likethat..."
257/ASQ, June 2001
More generally, employees of all races reported feeling that
they could show more of who they were at work than they
had been able to do in other work settings. A Latina member
of the program staff told us, "It's my first work experience
where the different perspectives I bring are not the only ones
in the office, and they are appreciated and accepted. Talking
about my life or bringing those perspectives is not something
that I have to worry about."
Racial identity among people of color in work groups with an
access-and-legitimacy perspective on diversity, in contrast,
was full of contradiction and ambivalence. In her advice to
other firms wishing to become more racially diverse, one
young black financial services employee in Retail Operations
summarized the quandary of being black in this setting: "Try
not to let the race thing be an issue," she urged. "I know
that's just like asking an elephant not to be gray. . . . I really
don't know how that could work, but it just needs to happen,
is all I can say." At the same time, she advised blacks in par-
ticular to "just remember who you are, and believe in your-
self and where you stand." Her advice was thus paradoxical:
erase the reality of race yet hold onto your black identity. We
suspect that in this kind of setting, in which racial diversity
assumes a highly circumscribed role-it has positive value
only insofar as it provides access to and legitimacy with a
diverse clientele-there is a mixed message about what it
means to be black. On the one hand, it bestows value on
blacks; on the other hand, it upholds an essentially assimila-
tionist vision in which white culture remains the dominant
culture. This mixed message raised concerns about losing
one's identity as a black person despite its avowed value in
the group.
In light of the mixed message the access-and-legitimacy per-
spective sends about the value and significance of being
black, it is not surprising that the meanings that black
employees in Retail attributed to their racial group member-
ship were often contradictory. When we asked black
employees about the salience or significance to them of their
identity group memberships at work, they typically respond-
ed by saying that "race is not a problem." The notion that
their racial group membership might have had a positive
impact on their work or their experiences at work, as it did
for program staff of color in the law firm, was conspicuously
absent in their responses, although they clearly understood
the importance of having black employees in Retail to provide
credibility with the firm's black clientele. For example, when
asked about the impact of her own racial identity at work,
one black employee was adamant that race was irrelevant.
She also remarked later in her interview, however, that "if
they put all of [External Deposits] down here [in Retail] for a
week ... they would be really whipped and surprised, and
they would probably run back to their department and never
look back . . . because that's an all white department." Her
reaction to an incident in a staff meeting we had witnessed,
in which a white male manager expressed strong disagree-
ment with a position that senior management endorsed, also
belied her declarations of racial equity: "Ithink that there are a
lot of people who wish they could have been that outspoken,"
258/ASQ, June 2001
Cultural Diversity
she said, "andthe discussion [amongblackmanagers]was

that had that been a blackperson he probablywould not be
here today."Herstatements taken together thus were contra-
dictory:race is irrelevant,but blacksare better suited to the
work in Retail,and whites enjoy greaterfreedom of expres-
sion. These kindsof contradictionssuggested that racialidenti-
ty may well have been a source of ambivalencefor blacks.
Unlikewhite programstaff in the law firm,white employees
in ExternalDeposits had littleconsciousness of their racial
identityat work. With the exception of the white manager
who attributedher "fit"with the cultureof the firmto her
race, no whites in ExternalDeposits reportedtheir racial
group membershipas salient in shaping their experiences or
how they expressed themselves at work. One white employ-
ee who now worked for ExternalDeposits, but who had for
many years been either the only white or one of a few in
Retail,said that she was "never conscious that no one was
white on the first floor [where Retailis located]. [Untila black
colleague suggested it,] it never occurredto me that I might
have been transferredto [ExternalDeposits] because I'm
white." That racialidentityfigured prominentlyin blackSales
Divisionemployees' reportsof their experience and seeming-
ly little in white employees' reports is predictablegiven the
precepts of the access-and-legitimacyperspective, which
minimizepeople's experience of diversitywhile seeking to
gain its most immediateand instrumentalbenefits.
Consistent with other outcomes that were associated with
the discrimination-and-fairness perspective, people of color
across work groups in the consultingfirmtypicallycharacter-
ized membership in their racialgroup as a source of power-
lessness and disenfranchisement.One blackmanager
explained, "It'slike a struggle between good white people
and bad white people, and basicallywe're observers, and we
just are rootingfor good white people to win." Consistent
with this observation,several described feelings of self-doubt
they often experienced as people of color and even ques-
tioned whether their apparentfailingsmight be due to their
own shortcomings as members of their racialgroups. As one
Latinaexplained, "So many of us find that it's a sink-or-swim
situation.... And I thinkthat those of us who are partof the
minorityhere feel that because of our temperament we're
not strong enough, so that in the sink-or-swim,we sink."
Similarly,anotherfelt that her boss ignoredher completely,
and she questioned whether "that'sa reflectionon me as a
Puerto Rican,or something I myself have made easy, you
know, sort of like my personalitygives room for him to feel
comfortabledoing that."
Althoughmany employees of color, particularly members of
the supportstaff, wished that they were not seen as "black,
Hispanic,or whatever," but were instead seen simply "for
who they are," there were a few members of the program
staff who saw their racialgroup membershipas a source of
culturalvalues for which they wished to be recognized. Nev-
ertheless, because they felt that whites were "afraidto rec-
ognize that there are differences in culture"and would find
such expressions "veryproblematic,"these employees, who
were, in any case, rarein this firm,typicallydid not express
259/ASQ, June 2001
their culturaldifferences. some employees of color tended to
resent this, accusing their seniors of "justbecoming carbon
copies of [whites] and . . . not really giving [whites] . . . a true
sense of the feelings of a person of color."Despite these
criticismsand the similaritiesin points of view we found
between white executives and executives of color, it was not
clear from our interviews whether there were any people of
color in the organizationwho, in fact, felt assimilated.
Most white employees, to the extent that they discussed the
significanceof their racialgroup membershipat all, discussed
it only as a basis for feeling intimidated,apprehensive,or
reluctantto speak out about race-relatedissues. They tended
to describe themselves as "oblivious"to what people of
color were experiencing, "perplexed"by their complaints.
Others were somewhat more reflective.The white executive
director,for example, recognizedthat in race relations,
"althoughthere is a wish to say that everybodystarts out in
the same place, and you should just deal as one infinitely
valuable human being to another . . . all kinds of power stuff
gets in there." Yet she had littleto say about how she, as a
white person in charge, might interveneto make race rela-
tions in the firm better. "Thereare only so many things
somebody who's white and in a leadershipposition can do
directlyon that subject," she said. "[Youjust have to] be the
best person you can be in terms of tryingto make the pro-
gram go the best way you can make it go." Althoughshe
recognizedthat this was "not sufficient," it was "aboutall I
know to do." Consistent with the discrimination-and-fairness
perspective, she was, as a white person, limitedto the moral
realmas a way of understandingthe role her racialidentity
might play in her abilityto address racialissues.
WORKGROUPFUNCTIONING
We found that the perspective on diversitya groupof people
held influencedhow they expressed and managed tensions
relatedto diversity,whether those traditionallyunderrepre-
sented in the organizationfelt respected and valued by their
colleagues, and how people valued and expressed them-
selves as members of their culturalidentitygroups;these, in
turn, influencedpeople's sense of self-efficacy and work
groupfunctioning.Allthree types of work group diversityper-
spectives were successful in motivatingmanagers to diversi-
fy their staffs, but only the integration-and-learning
perspec-
tive providedthe kindof rationaleand guidance people
needed to achieve sustained benefits from diversity.3Table3
summarizes the intermediategroupoutcomes of the three
diversityperspectives and their effects on group-functioning,
as detailed below.
Workgroups with an integration-and-learning perspective
were high functioning.At the law firm,all of the staff we
3 intervieweddescribed the firm'sprogramas successful, and
This is not to say that concerns about dis- virtuallyall attributedat least partof its success to program
crimination are unimportant nor that using staff's abilityand willingness to bringthe interests and per-
cultural diversity to gain access to and
legitimacy with different market seg- spectives of people of color "intothe centerpiece of the
ments is illegitimate; rather, our research organization."As one woman explained, "[Diversityin the
suggests that these alone as the primary
basis for a group's diversity strategy will programstaff] has affected the work in terms of expanding
likely undercut the group's effectiveness. notions of what are women's issues and takingon issues and
260/ASQ, June 2001
Cultural Diversity
Table 3
Intermediate Outcomes Mediating Effects of Diversity Perspectives on Group Functioning
Discrimination-
Mediators Integration-and-learning Access-and-legitimacy and-fairness
Quality of intergroup Conflict resulting from Conflict resulting from dif- Intractable race-related
relations cultural differences in ferential power and status conflict stemming from
point of view; different accorded different entrenched, undiscussible
groups accorded equal races/functions; little status and power imbal-
power and status; open open discussion of con- ances; no open discus-
discussion of differences flict sion of conflict or
and conflict differences
Feeling valued and All employees feel fully Employees of color ques- Employees of color feel
respected respected and valued for tion whether they are val- disrespected and deval-
their competence and ued and respected equal- ued as members of
contributions to the ly; perceive devaluation of minority racial/ethnic
organization functions staffed predom- groups
inantly by people of color
Significance of own racial Source of value for peo- Source of ambivalence for Source of powerlessness
identity at work ple of color, a resource for employees of color; for people of color;
learning and teaching; a whites not conscious source of apprehension
source of privilege for for whites
whites to acknowledge
Group functioning
Enhanced by cross-cultur- Enhanced by increased Inhibited by low morale of

al exposure and learning access and legitimacy; employees, lack of cross-
and by work processes inhibited by lack of learn- cultural learning, and the
designed to facilitate con- ing and exchange inability of employees of
structive intergroup con- between racially segre- color to bring all relevant
flict and exploration of gated functions skills and insights to bear
diverse views on work
framingthem as women's issues in creative ways that would

have never been done [with an all-whitestaff] and doesn't
get done by other women's organizations.It's reallychanged
the substance and in that sense enhanced the qualityof our
work."This result clearlyhinged on the open and direct way
in which the staff managed racialdifferences and conflicts,
the fact that they respected people and sought their contribu-
tions as members of their respective racialgroups, and the
fact that both white employees and employees of color were
able to consider and share with their colleagues how their
experiences as members of those groups influencedthem at
work. This approachto diversityencouraged and enabled pro-
gram staff of color to bringskills and capacities to the firm
that gave them access to importantinformationin their own
communities and helped them buildrapportwith clients,
thereby helpingto expand the firm'sclient base. Equally
important,however, was the emphasis on cross-cultural
exposure and education so that staff members were continu-
ally expandingtheir own capacities. The integration-and-learn-
ing perspective made identitya source of insightthat was
transferableto a broadrange of employees, not just to those
who were members of "diverse"groups. Diversity,thus,
was a resource on which all programstaff could draw.
Inaddition,by incorporatingdiversityinto the core work of
the organization,this perspective affordedall employees
some measure of access to and legitimacywith their clients,
regardless of employees' respective culturalidentities. One
261/ASQ, June 2001
white member of the programstaff, for example, explained
that the firm'sreputationas a raciallyintegratedfirm had
increased her credibilityin minoritycommunities and her abil-
ity to work in them. Similarly,a former member of the pro-
gram staff who is AfricanAmericanfelt that she personally
gained credibilitywith the firm'sAsian clients when a Japan-
ese Americanattorneyjoined the staff. This credibility
allowed staff members to network much more widely across
communities, which providedthem with a much richer,
broaderbase of information;this, in turn,gave them a better
perspective on the problems they were addressing, enhanc-
ing the qualityof their analyses. Finally,this perspective cre-
ated a model of workingin coalitionwith a numberof public
interest, civil rights,and other "people-of-color"groups,
which helped to facilitatea series of mutuallybeneficial,
cross-organizationalcollaborations.
Ourdata suggested that while the access-and-legitimacyper-
spective enhanced a work group'sabilityto reach more
diverse market,it was limitingin other ways. The financial
service firm'sgoals were to make a profitfor the company
and to develop and revitalizethe economy of the local com-
munitywithinwhich it was situated. The Sales Division's
access-and-legitimacyperspective on diversityhad indeed
advanced these goals by giving members some measure of
access to and legitimacywith both the local communityto
whom they appealed for personal investments and commer-
cial ventures, as well as the nationalcommunityto whom
they appealed for socially responsible investments and the
purchase of other kinds of competitive money-management
products.And most informantsagreed that ExternalDeposits
had grown the firm'sassets well beyond expectations. Nev-
ertheless, many were concerned that RetailOperationshad
thus far been unableto reach its growth potentialin the local
communityand that ExternalDeposits' capacityto sustain its
growth would be severely limitedby increasingcompetition
in its nationalmarkets. Ourdata suggested that, despite the
benefits the access-and-legitimacyperspective had garnered
for the Sales Division,this perspective also contributedto the
problems Divisionemployees faced in at least three ways, all
of which were relatedto the racialdivisionof laborit seemed
inevitablyto create.
First,some participantsreportedthat RetailOperations'
lower status in the organizationcompromised the qualityof
service Retailclients received. One woman who had worked
in both departments speculated that the reason for making
the two departments separate in the first place was to draw
"a very distinct line" between their respective customers.
Whether the result of fewer resources in Retail,such as
time, or Retailemployees' diminishedsense of entitlement
for their clients, most people acknowledged, often with clear
racialovertones, that Retailclients received a lower qualityof
service than clients in ExternalDeposits: "Customersin
Retaildon't get that special touch that ExternalDeposits' rich
white clients get," one customer service agent in Retail
lamented. Reiteratingthese concerns, the managerof Retail
Operationsprovidedanecdotal evidence to suggest that her
customers were "overshadowedby the hoity-toitytreat-
262/ASQ, June 2001
Cultural Diversity
ment" others got and were takingtheir business elsewhere

as a result.
Second, referringto the duplicationof efforts in the two
departments-a direct result of how the access-and-legitima-
cy perspective was manifest in this division-the managerof
ExternalDeposits explained, "It'sreallyinefficientto have
what are essentially two banks here." It could take one of
her employees "seven hours to do something himself that
he could have taken to Retailand gotten done much more
quickly,"she explained, but for "the competitiveness and ani-
mosity between the two departments." Moreover,she felt
that this competitiveness threatened to compromise the
qualityof service some customers received. Referringto the
recent additionof a corporatebankingfunction in Retail
designed to service corporateaccounts citywide, together
with her own department'srecent efforts to develop socially
responsible investments within the city, the managerof
ExternalDeposits was concerned that the line between their
client bases would become increasinglyblurred:"Historically,
the Retailside has been defined as [the neighborhoods].Any-
thing else in the city by rights should be mine if we use that
definition.Right?So what happens if I get a law firmdown-
town that needs corporatebankingservices, and I bringthem
in? Whose account is that? I reallycan't service it, but Retail
that's their stock and trade." It was her feeling that with
better relationshipsand less disparitybetween the two
departments,these kinds of conflicts could be avoided and
customers would receive the qualityof services that was
their due, ratherthan being caught up in a battle over whose
account was rightfullywhose.
Finally,there were inefficiencies in the perfectionist "white"
culturethat had come to characterizeExternalDeposits
because they were unableto learnfrom RetailOperations.
Criticalof the cultureher predecessor had builtin her efforts
to service the needs of her more affluent, more demanding
clientele, the currentmanagerof ExternalDeposits explained,
It'svery hardto make money with all that perfectionism.A letter
would be edited four times before it went out the door.... In my
opinion,that just isn't necessary.... [Tiheaverage bankcustomer, I
think,wants somebody who's steady, loyal,knows their business
inside and out and works hard.I don't necessarilywant someone
who, every time a customer calls they'll design a new productfor
them.... And we did an awful lot of that.
This managerfelt stronglythat in this respect, among others,
there might be something to learnfrom the way RetailOper-
ations functioned, but the "culturalbarriers,"created by their
longstandingdifferences made it difficultfor them to collabo-
rate. "They'revery guarded,"she explained. "Theydon't
believe that I reallywant to know what they're saying."
The discrimination-and-fairness perspective appearedto have
a negative effect on work groupfunctioningin a varietyof
ways. Inthe consultingfirm,althoughdifferentgroups laid
blame in differentplaces for the fact that whites were reluc-
tant to disagree with people of color,challenge them, or pro-
vide feedback to them, most agreed that it compromised
both their own and their department'sabilityto reach their
263/ASQ, June 2001
potential. "Because a lot of the problems here have not been
dealt with openly,"a white managerexplained, "they have
been allowed to fester, and people who are incompetent
remainincompetent." In a similarvein, a blacksupport staff
member lamented her inabilityto get "correctivecriticism"
from her white supervisor,"whichwould only furthersupport
my desire, not only to do my job well, but also to gain as
much knowledge about my job and any other technical skills
as might be necessary to enhance my work."
The numerous "racial"incidents and subsequent organizing,
memo-writing,and meeting cost the organizationnot only
the time and energy of the people of color who engaged in
these activities on company time but the moraleof everyone
who suffered from the tense work environmentas a result.
As one white managerexplained, "the tension [overthe fir-
ing of the blackemployee] was palpablein the organization,
which made it harderto come in to work boundingwith
enthusiasm. These incidents affect everyone's morale;you
bounce back, but only untilthe next one erupts." In response
to a differentincident,another described the whole organiza-
tion as "grindingto a halt because of the morale problem."
People of color also found it drainingand time-consuming
always to have to wonder whether their treatment was race-
relatedor not. As one woman explained, "Itreallyhampered
me in the beginning,and I started to question myself all the
time." Others described how management's apparentlackof
interest in their ideas not only made them feel devalued but
was potentiallycostly to their departmentsas well. One mid-
level managersaid he "hada vision" for the function he
supervised but found it difficultto get the ear of "the people
who can make a difference." He said that althoughhe tried
to look past "the possibilitythat this was because of race," it
was difficult.He found management's inattentionboth per-
plexingand depressing and, as a result, had decided no
longer to offer his point of view.
To the extent that whites associated diversitywith positive
outcomes, it tended to be because they felt they had
"learnedan immense amount about race" or that the pres-
ence of people of color had helped them attaintheir "ideals
of equalityand justice." There were a few white program
staff, however, who also felt that diversityhad had a positive
impacton their programsbecause members of other cultural
groups were able to assist them in their programwork with
culturallysimilarclient groups. One person, for example, saw
the value of involvingLatinostaff in the firm'sCentralAmeri-
can work because they had useful insights into race relations
there. Those programstaff of color who also saw the possi-
bilityof such connections, however, typicallydescribed their
colleagues' resistance to their using insights derivedfrom
their particularculturalperspectives. Moreover,when they
did try to make such connections, they, like theirwhite coun-
terparts,would adopt the discrimination-and-fairness moral
framingof differences in the ensuing debate, which was ulti-
mately unproductive.An AfricanAmericanprogrammanager
who headed economic development activities in Eastern
Europetried to get his colleagues to consider reorganizing
the firm'sdevelopment work accordingto similaritiesin coun-
264/ASQ, June 2001
Cultural Diversity
tries' development experiences ratherthan geographical

area. Poland,he argued, had more in common with certain
Africanand LatinAmericancountries than with other Euro-
pean countries and therefore could benefit more from exper-
tise developed in Africaand LatinAmericathan in Europe.As
an AfricanAmerican,he felt he was perhaps less committed
to the firm's "Eurocentric"orientation,which he believed led
his colleagues to assume erroneously-and to the firm's
detriment-that white countries have more in common with
each other than with nonwhite countries. He never succeed-
ed in generatinga constructivediscussion of this idea, how-
ever, because the exchange quicklydegenerated into a
debate about which view-the firm'sor his-was more racial-
ly motivatedand therefore racist.This framing,in which he
participated,foreclosed opportunitiesfor learningabout how
his departmentmight do its work more effectively.
DISCUSSION
Our research showed how three diversityperspectives differ-
entiallyaffected the functioningof culturallydiverse work
groups. The crucialdimension along which the three diversity
perspectives variedwas whether and how culturaldiversity
was linkedto the group'swork and work processes. Inthe
integration-and-learning perspective, culturaldiversityis a
potentiallyvaluableresource that the organizationcan use,
not only at its margins,to gain entree into previouslyinacces-
sible niche markets, but at its core, to rethinkand reconfigure
its primarytasks as well. It is based on the assumption that
culturaldifferences give rise to differentlife experiences,
knowledge, and insights, which can informalternativeviews
about work and how best to accomplish it. Inthe work
groups we studied that embraced this perspective, this view
of the role of racialdiversityencouraged group members to
discuss openly their differentpoints of view because differ-
ences-including those explicitlylinkedto culturalexperi-
ence-were valued as opportunitiesfor learning.This
process communicatedto all employees that they were val-
ued and respected and encouraged them to value and
express themselves as members of their racialidentity
groups. These aspects of the way they functionedafforded
opportunitiesfor cross-culturallearning,which enhanced the
group'swork.
Inthe access-and-legitimacyperspective, culturaldiversityis
a potentiallyvaluableresource, but only at the organization's
marginsand only to gain access to and legitimacywith a
diverse market. Inthe work groups we studied that
embraced this perspective, this view of the role of racial
diversityled to race-based staffing patternsthat matched the
racialmake-upof the marketsthey served. This fostered per-
ceptions of white-staffed functions as higherstatus than
functions staffed by people of color;raciallysegregated
career tracks and opportunities,which fostered concerns
among staff of color about the degree to which they were
valued and respected; and ambivalenceon the partof people
of color about the meaning and significanceof their racial
ten
identityat work. The resultinginterracial/interfunctional
sions appearedto inhibitlearningand people's abilityto be
maximallyeffective in their work.
265/ASQ, June 2001
Finally,in the discrimination-and-fairness perspective, cultural
diversityis a mechanism for ensuringequal opportunity,fair
treatment, and an end to discrimination;it articulatesno link
at all between culturaldiversityand the group'swork and, in
fact, espouses a color-blindstrategy for managingemployees
and employee relations.Inthe work groups that embraced
this perspective, this view of the role of racialdiversity
restrictedthe discourse about race to one in which employ-
ees negotiated the meaning of all race-relateddifferences on
moralgrounds. Questions and concerns about fairness led
inevitablyto strainedrace relationscharacterizedby compet-
ing claims of innocence, with each group assuming a defen-
sive posture in relationto the other (Steele, 1990). Racial
identitythus became a source of apprehensionfor white peo-
ple and feelings of powerlessness for many people of color.
This made it difficultfor people to bringall relevantskills and
insights to bear on their work, thus compromisingtheir ability
to learnfrom one another and to be maximallyeffective.
Implications
Ourresearch makes three theoreticalcontributions.First,we
providea social theory of how work groups make sense of
their culturaldiversityand how this shapes members' identi-
ty, intergrouprelations,and the conduct of work. A central
construct of the theory is a group'sdiversityperspective. A
diversityperspective providesthe cognitive frames within
which group members interpretand act upon their experi-
ence of culturalidentitydifferences in the group. Using these
frames, members of culturallydiverse work groups collective-
ly construct and participatein interculturalidentitygroup rela-
tions withinthe group, which influences members' sense of
how much others in the groupvalue and respect them, as
well as their sense of what their own culturalidentitymeans
at work. These group processes and member experiences in
turn have implicationsfor the group'scapacityfor learning
and adaptivechange in its work and thus for members'
sense of self- and group efficacy. Althoughthe research liter-
ature contains speculationabout the motivationsan organiza-
tion may have for diversifyingits workforce,some of which
resonate with the differentperspectives on diversitywe iden-
tified, it neitherdevelops them nor recognizes them as
among the "untested subjective concepts" that may inter-
vene between the demographiccomposition of groups, on
the one hand, and their effectiveness, on the other
(Lawrence,1997: 20). When a work groupviews culturaldif-
ferences among its members as an importantresource for
learninghow best to accomplish its core work, group mem-
bers can negotiate expectations, norms, and assumptions
about work in service of their goals, and conflicts that arise
are settled by a process of joint inquiry(Argyrisand Schon,
1978). Inwork groups in which it is legitimatefor group
members to bringall of their relevantknowledge and experi-
ence to bear on the core work of the group-including knowl-
edge and experience that is linkedto their culturalidentity-
members are more likelyto feel valued and respected in the
groupand to receive more validationfor their culturalself-
identities (Tylerand Lind,1992). This heightens group mem-
bers' feelings of effectiveness (Lindand Tyler,1988) and
266/ASQ, June 2001
Cultural Diversity
motivationto achieve (Hackman,1992). Moreover,because

these groups are inquiry-oriented, ratherthan competitive,
and characterizedby a high degree of trust, risktaking,and
psychologicalsafety, there are greater opportunitiesfor com-
petency-enhancingcross-culturallearning(Argyrisand Schon,
1978; Edmondson,1999). By contrast, when a work group
views culturaldifferences as havingthe potentialto make
only a marginalor negative contributionto work, the domi-
nant culturalgroup likelydefines the prevailingexpectations,
norms, and assumptions about work, and conflicts, if not
suppressed, are settled by power. Groupsthat do not autho-
rize members to use their culturalexperiences as a resource
for learningconvey mixed messages at best about the
degree to which all members and culturalidentitiesare val-
ued, creatingtension, competitiveness, and distrust in the
group.This impedes learningand limits members' sense of
self- and group efficacy. These observationsare consistent
with research that suggests that a work group'ssuccess
often hinges on members' abilityto engage differences in
knowledge bases and perspectives (Bailyn,1993; Jehn,
Northcraft,and Neale, 1999) and to embrace, experience,
and manage, ratherthan avoid, disagreements that arise
(Gruenfeldet al., 1996; Jehn, 1997). Previoustheorizing(e.g.,
Cox, 1993) notwithstanding,our findings suggest that cultural
diversityin the senior ranksof an organization,which existed
in all three of the firms in our sample, is not sufficient to pro-
duce the kindof shift in power relationsthat enables these
constructivegroup processes to occur.
Second, our research shows how organizationsmediate the
impactof largersocial processes on organizationalfunction-
ing (Zucker,1987). In contrastto the distributionof power
between racialgroups in society, all three of the organiza-
tions in our study had significantnumbers of people of color
in positions of power, yet their differentwork group perspec-
tives on diversitysuggested differentstrategies for managing
this situation,which in turn had differentconsequences both
for the balance of power between racialgroups inside the
organizationand for the work group'sfunctioning.The assimi-
lationiststrategies adopted by work groups that embraced
either the discrimination-and-fairness or the access-and-legiti-
macy perspective seemed simplyto replicateasymmetric
power relationsbetween racialgroups in the largersociety,
inhibitingeffective functioning.By contrast,the integrationist
strategies adopted by work groups that embraced the inte-
gration-and-learning perspective seemed to foster more sym-
metric relationsof power as well as more effective function-
ing. In makingthese connections, we extend the growing
literatureon organizationaldemography,which has begun to
recognize and highlightthe distributionof power within orga-
nizationsas an importantdemographicvariablemoderating
the impactof societal conditionson organizationalbehavior
(Ely,1994, 1995; Ragins, 1997; Lauand Murnighan,1998;
Thomas, 1999) to includework groups' perspective on their
demographicmake-upas well.
Third,our research suggests that just as some organizations
attemptingto diversifyhave done so from a discrimination-
and-fairnessperspective on diversity,so, too, has much of
267/ASQ, June 2001
the organizationalliteratureassumed this perspective in its
approachto understandingdiversity.Both in organizations
and in organizationalresearch, this perspective has been lim-
iting. Forexample, scholars implicitlytake a discrimination-
and-fairnessperspective on diversitywhen they characterize
culturalaspects of identity,such as race and gender, as high
on the dimension of visibilityand low on the dimension of
job-relatednessin explainingthe negative effects of diversity
on groupfunctioning(e.g., Pelled, 1996; Jehn, Chadwick,and
Thatcher,1997). These scholars typicallyposit that because
these characteristicsare easily observable, they are more
accessible as a basis for categorizationand hence are more
likelythan less visible differences to motivate intergroupbias
and feelings of hostility,anxiety,and frustration(Tsui,Egan,
and O'Reilly,1992; Strangoret al., 1992; Pelled, 1996). Fur-
thermore,they posit that because these characteristicsare
not job-related-"they do not reflect task perspectives and
technical skills" (Pelled, 1996: 619)-they do not spark "dis-
agreements about task issues includingthe natureand
importanceof task goals and key decision areas, procedures
for task accomplishment,and the appropriatechoice for
action" (Pelled, 1996: 620). Consistent with the discrimina-
tion-and-fairnessperspective, this approachassumes that
aspects of identitysuch as race and gender are relevantonly
insofaras they triggerothers' negative reactions;they are
therefore a potentialsource of negative intergroupconflictto
be avoided in service of the task. By contrast, the socially
constructed view of culturalidentitywe take in this research
recognizes the role social context plays in shaping what is
both visible and job-relatedand gives at least as much weight
to the meaning people attributeto their own demographic
characteristicsas to the meaning they attributeto others'.
This approachenabled us to identifyconstructivepossibilities
for the role of culturalidentityprecludedby approaches with
a more static conception of identity.
These contributionsaside, our research raises a numberof
questions concerningissues researchers should consider and
the methodologies they use to investigate diversitydynamics
and groupfunctioning.First,althoughwe would hope our
results might generalizeto organizationsinterested in benefit-
ing from a culturallydiverse workforce,our sample is not rep-
resentative of all such organizationson a numberof potential-
ly importantdimensions. Perhapsthe most relevant
dimension that differentiatesthe organizationsin our sample
from many others is that they are all drivenby social and eco-
nomic goals related in one way or anotherto communities of
color,which likelyexplains their interest in culturaldiversityin
the first place. We have no data from this study to assess
directlywhether or how firms whose mission is not so readi-
ly linkedto diversitywould reap the benefits we found to be
associated with the integration-and-learning perspective. We
suspect, however, that even in firms in which the work con-
tent is less obviously relatedto the culturalcompetencies
afforded by a culturallydiverse work group, the insights and
perspectives of such a group can nevertheless informits
work processes, as they did in the work groups we observed
that adopted an integration-and-learning perspective. Another
factor that differentiatesthe organizationsin our sample from
268/ASQ, June 2001
Cultural Diversity
others to which we would liketo generalize our results is

that they were all relativelysuccessful in their affirmative
action attempts; all had achieved significantlevels of diversity
in hierarchicaland functionalpositions traditionallyoccupied
by white men. Thus, it remains unclearwhether or how
diversityperspectives influencefirms that have yet to
achieve these levels or in which educationaland occupational
status distinctionsfall along culturalidentitylines, as they
currentlydo in most organizations.Furtherresearch should
explore whether and how the diversityperspectives we iden-
tified-and/or others-have helped other organizationsto
recruitand retainhigh levels of workforcediversityand with
what consequences. These efforts should include research in
organizationsthat, unlikethose in our sample, are more pure-
ly profit-drivenor, at least, less drivenby social and economic
goals explicitlyrelatedto communities of color and in organi-
zations that have achieved varyingdegrees of success in
their efforts to diversify.
Second, our data collection design allowed us to generate
ratherthan test theory.The connections we propose here
among the constructs we identifiedare, therefore, necessari-
ly speculative. We are unableto determine what role, if any,
contextualfactors that happened to covarywith diversityper-
spectives may have played in producingeither the group
processes and individualexperiences we observed or the dif-
ferent levels of functioningwe associated with them. Two
such factors, which may be confounded with diversityper-
spectives, are the size and status composition of the work
groups. The groups in which we observed the integration-
and-learningperspective, in both the law and financialser-
vices firms, were small-four and seven people, respective-
ly-and relativelyhomogeneous with respect to members'
professionalstatus. It may be that in small work groups,
especially those in which members are of similarstatus, the
problems caused by diversityare more easily overcome (Lau
and Murnighan,1998). Clearly,the hypothesis our research
points to-that a work group's perspective on the role of cul-
turaldiversitymediates the impactof that diversityon its
functioning-remains to be tested and refinedwith other
samples of organizations.The group processes and individual
experiences we propose here as the mediatingfactors that
linkthe group'sdiversityperspective to its functioningalso
requirefurtherempiricalinvestigation,and researchers need
to learnmore about how those factors work in differentorga-
nizationalsettings.
Third,among the groups we observed, we found three per-
spectives. After initiallydefiningthe diversityperspective
construct, we were open to findingadditionalperspectives
when we returnedto the data to conduct a more thorough
content analysis, but we did not find any. Nevertheless, there
may well be additionalperspectives or groups in which no
single perspective prevailsbut where, instead, there are
hybridor competing perspectives. At this point, we are
unableto speculate furtherabout these possibilitiesbut rec-
ommend being open to them in future research. To assess a
group's diversityperspective, it is necessary to collect data
from at least a representativecross-section of the group's
269/ASQ, June 2001
members.4 Researchers should aim to assess not only the
group'sexternallyespoused values and beliefs but those that
are internallyenacted as well-its basic assumptions, which
often remainconcealed or unconscious (Schein, 1984; Barley,
1991; Martin,1992). We recommend our method of observ-
ing behavioralinteractionsamong group members from
which one can infernormativebeliefs and content analyzing
responses to open-ended interviewquestions.5
Finally,we need to learnmore about how and underwhat
conditionswork groups develop and change their perspec-
tives on diversityand, in particular,how they change to the
more promisingone of integrationand learning.Ourcasual
and systematic observations of many organizationssuggest
that both the discrimination-and-fairness and the access-and-
legitimacyperspectives are more common than the integra-
tion-and-learning perspective. More theoreticaland empirical
development is needed to understandfullythe integration-
and-learningperspective's potentialfor connecting organiza-
tions' culturaldiversityto their core work and work process-
es. With such theory, organizationswill be better positioned
4 to gain the promised benefits of culturaldiversity.
It appears from our data that in order for
a diversity perspective to produce the REFERENCES
results we have observed, a single diver-
sity perspective must prevail in a work
Alderfer, C. R Argyris, C., and D. A. Shon
group, with no systematic differences
along either hierarchical or racial lines. 1980 "The methodology of organi- 1978 Organizational Learning: A
Where we observed differences in per- zation diagnosis." Profession- Theory of Action Perspective.
spective within a group, it tended to be al Psychology, 11: 459-468. Reading, MA: Addison-
those in the lower echelons of the organi- 1987 "An intergroup perspective Wesley.
zation's formal hierarchy who deviated on group dynamics." In J.
from the majority point of view. To the Lorsch (ed.), Handbook of Bailyn, L.
extent that there is active resistance from 1993 Breaking the Mold: Women,
Organizational Behavior:
below to using diversity in service of the
190-219. Englewood Cliffs, Men, and Time in the New
work, whether from the access-and-legiti- Corporate World. New York:
macy or integration-and-learning perspec- NJ: Prentice-Hall.
1992 "Changing race relations Free Press.
tive, these perspectives might be difficult
to implement. Our observations tentative- embedded in organizations: Bantel, K. A., and S. E. Jackson
ly suggest, therefore, that for a single Report on a long-term project 1989 "Top management and inno-
diversity perspective to prevail in any with the XYZCorporation." In vations in banking: Does the
given work group (1) a majority of mem- S. E. Jackson and Associates composition of the top team
bers, including but not limited to those in (eds.), Diversity in the Work-
formal positions of authority and power,
make a difference?" Strategic
place: Human Resources Ini- Management Journal, 10:
share and be able to articulate the per-
spective, (2) no systematic differences in tiatives: 138-166. New York: 107-1 24.
perspective exist as a function of mem- Guilford Press.
bers' cultural identities, and (3) to the
Barley, S. R.
extent that there are differences, the rela-
Alderfer, C. R, and K. K. Smith 1991 "Semiotics and the study of
tively few dissenting views be held by 1982 "Studying intergroup relations occupational and organization-
those with relatively little formal authority embedded in organizations." al cultures." In P. Frost, L.
or power in the group. Administrative Science Quar- Moore, C. Lundberg, M.
terly, 27: 35-65. Louis, and J. Martin (eds.),
5
Reframing Organizational Cul-
Assessment might also involve, for exam- Alderfer, C. P., R. Tucker, C. Alder-
ple, presenting vignettes for group mem- ture: 39-54. Newbury Park,
fer, and L. Tucker
bers to interpret in ways that reveal the CA: Sage.
1980 "Diagnosing race relations in
assumptions and beliefs underlying their management." Journal of
group's behavior. These approaches are
Belenky, M. F., B. M. Clinchy, N. R.
less subject to rationalizationand self-
Applied Behavioral Science, Goldberger, and J. M. Tarule
conscious manipulation and are therefore 16: 135-166. 1986 Women's Ways of Knowing:
less likely to be influenced by self-presen- Ancona, D. Gladstein, and D. F. The Development of Self,
tation and social desirability concerns
Caldwell Voice, and Mind. New York:
than some other, more direct methods Basic Books.
(Martin, 1992). Because each of the per- 1992 "Demography and design:
spectives we identified, stated on its own Predictors of new product Blalock, H. M., Jr.
terms, appeals to laudable goals and team performance." Organi- 1957 "Percent non-white and dis-
makes a reasonable argument for diversi- zation Science, 3: 321-341. crimination in the South."
ty, surveys with Likert-type scales on American Sociological
which people indicate their level of agree- Review, 22: 677-682.
ment with the different rationales and
normative beliefs associated with each
perspective would be inappropriate.
Cultural Diversity
Blau, P M. Dworkin, A., J. Chafetz, and R. Foeman, A. K., and G. Pressley

1977 Inequality and Heterogeneity. Dworkin 1987 "Ethnic culture and corporate
New York:Free Press. 1983 "The effects of tokenism on culture: Using black styles in
job commitment among organizations." Communica-
Bochner, S., and B. Hesketh
urban public school teachers: tion Quarterly, 35: 293-307.
1994 "Power distance, individual- A test of Kanter's approach."
ism/collectivism, and job-relat- Paper presented at the Amer- Gilligan, C.
ed attitudes in a culturally ican Sociological Association 1982 In a Different Voice. Cam-
diverse work group." Journal meetings, August 1983. bridge, MA: HarvardUniversi-
of Cross-Cultural Psychology, ty Press.
25: 233-257. Eagly, A. H., and B. T. Johnson
1990 "Gender and leadership style: Gruenfeld, D. H., E. A. Mannix,
Calas, M., and L. Smircich A meta-analysis." Psychologi-
K. Y. Williams, and M. A. Neale
1993 "Dangerous liaisons: The 1996 "Group composition and deci-
cal Bulletin, 108: 233-256.
'feminine-in-management' sion making: How member
meets globalization." Busi- Edmondson, A. familiarity and information dis-
ness Horizons, 36 (2): 71-81. 1999 "Psychological safety and tribution affect process and
learning behavior in work performance." Organizational
Chatman, J. A., J. T. Polzer, S. G. teams." Administrative Sci- Behavior and Human Decision
Barsade, and M. A. Neale
ence Quarterly, 44: 350-383. Processes, 67: 1-15.
1998 "Being different yet feeling
similar: The influence of Ely, R. J. Gutek, B. A.
demographic composition and 1994 "The effects of organizational 1985 Sex and the Workplace. San
organizational culture on work demographics and social iden- Francisco: Jossey-Bass.
processes and outcomes." tity on relationships among
Administrative Science Quar- professional women." Admin- Hackman, J. R.
istrative Science Quarterly, 1992 "Group influences on individ-
terly, 43: 749-780.
39: 203-238. uals in organizations." In
Clement, D. E., and J. J. Schiereck 1995 "The power in demography: M. D. Dunnette and L. M.
1973 "Sex composition and group Women's social constructions Hough (eds.), Handbook of
performance in a visual signal of gender identity at work." Industrialand Organizational
detection task." Memory and Psychology: 199-267. Palo
Academy of Management
Cognition, 1: 251-255. Journal, 38: 589-634.
Alto, CA: Consulting Psychol-
ogists Press.
Cox, T. H., Jr. Ely, R. J., and D. E. Meyerson
1993 Cultural Diversity in Organiza- Harding, S.
2000 "Theories of gender in organi-
tions: Theory, Research, and 1986 The Science Question in
zations: A new approach to
Practice. San Francisco: Feminism. Ithaca, NY:Cornell
organizational analysis and
Berrett-Koehler. University Press.
change." In B. M. Staw and
Cox, T. H., Jr., and S. Blake R. I. Sutton (eds.), Research Harlan, A., and C. Weiss
1991 "Managing cultural diversity: in Organizational Behavior, 22: 1981 Moving Up: Women in Man-
Implications for organizational 105-153. New York:Elsevier agerial Careers. Wellesley,
competitiveness." Academy Science/JAI. MA: Wellesley College: Cen-
of Management Executive, 5 ter for Research on Women.
Epstein, C. F.
(3): 45-56. 1988 Deceptive Distinctions. New Helgesen, S.
Cox, T. H., Jr., S. A. Lobel, and Haven, CT:Yale University 1990 The Female Advantage:
P L. McLeod Press. Women's Ways of Leader-
1991 "Effects of ethnic group cul-
Fiedler, F. E. ship. New York:Doubleday.
tural differences on coopera- 1966 "The effect of leadership and Hoffman, L. R.
tive and competitive behavior cultural heterogeneity on 1959 "Homogeneity of member
on a group task." Academy of
group performance: A test of personality and its effect on
Management Journal, 34: the contingency model." group problem-solving." Jour-
827-847. Journal of Experimental Social nal of Abnormal and Social
Deaux, K., and J. Ullman Psychology, 2: 237-264. Psychology, 58: 27-32.
1983 Women of Steel. New York:
Fiedler, F. E., W. A. T. Meuwese, Hoffman, L. R., and N. R. F. Maier
Praeger. and S. Oonk 1961 "Quality and acceptance of
Di Stefano, C. 1961 "Performance on laboratory problem solutions by mem-
1990 "Dilemmas of difference: tasks requiring group creativi- bers of homogeneous and
Feminism, modernity, and ty." Acta Psychologica, 18: heterogeneous groups." Jour-
postmodernism." In L. J. 100-119. nal of Abnormal and Social
Nicholson (ed.), Psychology, 62: 401-407.
Flax, J.
Feminism/Postmodernism: 1990 "Postmodernism and gender
63-82. New York:Routledge. lzraeli, D. N.
relations in feminist theory." 1983 "Sex effects or structural
In L. J. Nicholson (ed.), Femi- effects? An empirical test of
nism/Postmodernism: 63-82. Kanter's theory of propor-
New York:Routledge. tions." Social Forces, 62:
153-1 65.
Fletcher, J.
1999 Disappearing Acts: Gender, Jackson, S. E., and Associates
Power, and Relational Prac- 1992 Diversity in the Workplace:
tice at Work. Cambridge, MA: Human Resources Initiatives.
MIT Press. New York:Guilford Press.
271/ASQ, June 2001

Jackson, S. E., and M. N. Ruder- Lind, E. A., and T. R. Tyler Nkomo, S. M.
man (eds.) 1988 The Social Psychology of Pro- 1992 "The emperor has no clothes:
1995 Diversity in Work Teams. cedural Justice. New York: Rewriting 'race' in organiza-
Washington, DC: American Plenum. tions." Academy of Manage-
Psychological Association. ment Review, 17: 487-513.
Martin, J.
Jehn, K. A. 1992 Culture in Organizations. New Pelled, L. H.
1997 "A qualitative analysis of con- York:Oxford University Press. 1996 "Demographic diversity, con-
flict types and dimensions in flict and work group out-
Martin, R Y.
organizational groups." comes: An intervening
Administrative Science Quar- 1985 "Group sex composition in process theory." Organization
work organizations: A struc-
terly,42: 530-557. Science, 7: 615-631.
tural-normative model." In
Jehn, K. A., C. Chadwick, and S. B. Bacharach and S. M. Ragins, B. R.
S. M. B. Thatcher Mitchell (eds.), Research in 1997 "Diversified mentoring rela-
1997 "Toagree or not to agree: the Sociology of Organiza- tionships in organizations: A
The effects of value congru- tions, 4: 311-349. Greenwich, power perspective." Acade-
ence, individual demographic CT:JAI Press. my of Management Review,
dissimilarity, and conflict on 22: 482-521.
workgroup outcomes." Inter- Maznevski, M. L.
1994 "Understanding our differ- Ridgeway, C. L.
national Journal of Conflict
ences: Performance in deci- 1988 "Gender differences in task
Management, 8: 287-305.
sion-making groups with groups: A status and legitima-
Jehn, K. A., G. B. Northcraft, and diverse members." Human cy account." In M. Webster,
M. A. Neale Relations, 47: 531-552. Jr. and M. Foschi (eds.), Sta-
1999 "Why differences make a dif- tus Generalization: New Theo-
ference: A field study of McGrath, J. E., J. L. Berdahl, and ry and Research: 188-206.
H. Arrow
diversity, conflict, and perfor- Stanford, CA: Stanford Uni-
mance in workgroups." 1995 "Traits,expectations, culture,
versity Press.
and clout: The dynamics of 1991 "The social construction of
Administrative Science Quar-
diversity in work groups." In
terly,44: 741-763. status value: Gender and
S. E. Jackson and M. N. Rud- other nominal characteris-
Kanter, R. M. erman (eds.), Diversity in tics." Social Forces, 70:
1977 Men and Women of the Cor- Work Teams: 17-45. Wash- 367-386.
poration. New York:Basic ington, DC: American Psycho-
Books. logical Association. Ridgeway, C. L., and J. Berger
1986 "Expectations, legitimation,
Kent, R. N., and J. E. McGrath Mednick, M. T. and dominance behavior in
1969 "Task and group characteris- 1989 "On the politics of psycholog- task groups." American Soci-
tics as factors influencing ical constructs: Stop the
ological Review, 51: 603-617.
group performance." Journal bandwagon, I want to get
of Experimental Social Psy- off." American Psychologist, Rosener, J.
chology, 5: 429-440. 44:1118-1123. 1990 "Ways women lead." Harvard
Business Review, Nov-Dec:
Konrad, A. M., S. Winter, and Meglino, B. M., E. C. Ravlin, and 119-125.
B. A. Gutek C. L. Adkins
1992 "Diversity in work group sex 1989 "A work values approach to Ruhe, J. A.
composition: Implications for corporate culture: A field test 1978 "Effect of leader sex and
majority and minority mem- of the value congruence leader behavior on group
bers." In P. S. Tolbert and process and its relationship to problem-solving." Proceed-
S. B. Bacharach (eds.), individual outcomes." Journal ings of the American Institute
Research in the Sociology of of Applied Psychology, 74: for Decision Sciences, North-
Organizations, 10: 115-140. 424-432. east Division, May: 123-127.
Greenwich, CT:JAI Press.
Milliken, F. J., and L. L. Martins Ruhe, J. A., and J. Eatman
Larkey, L. K. 1996 "Searching for common 1977 "Effects of racial composition
1996 "Toward a theory of commu- threads: Understanding the on small work groups." Small
nicative interactions in cultur- multiple effects of diversity in Group Behavior, 8: 479-486.
ally diverse workgroups." organizational groups." Acad-
Schein, E.
Academy of Management emy of Management Review,
1984 "Coming to see a new aware-
Review,21: 463-491. 21: 402-433.
ness of organizational cul-
Lau, D. C., and J. K. Murnighan Morrison, A. M. ture." Sloan Management
1998 "Demographic diversity and 1992 The New Leaders: Guidelines Review, 25 (2): 3-16.
faultlines: The compositional on Leadership Diversity in
Shaw, M. E.
dynamics of organizational America. San Francisco:
1983 "Group composition." In H.
groups." Academy of Man- Jossey-Bass.
H. Blumberg, A. P. Hare, V.
agement Review, 23:
325-340. Murnighan, J. K., and D. E. Kent, and M. Davies (eds.),
Conlon Small Groups and Social Inter-
Lawrence, B. S. 1991 "The dynamics of intense action, 1: 89-96. Chichester,
1997 "The black box of organiza- work groups: A study of UK:Wiley.
tional demography." Organiza- British string quartets."
tion Science, 8: 1-22. Administrative Science Quar-
terly,36: 165-186.
272/ASQ, June 2001
CulturalDiversity
South, S. J., C. M. Bonjean, W. T. Thomas, D. A., and R. D. Ely Williams, K. Y., and C. A. O'Reilly
Markham, and J. Corder 1996 "Making differences matter: III
1982 "Social structure and inter- A new paradigm for manag- 1998 "Demography and diversity in
group interaction: Men and ing diversity." HarvardBusi- organizations." In B. M. Staw
women of the federal bureau- ness Review, Sept.-Oct.: and R. I. Sutton (eds.),
cracy." American Sociological 79-90. Research in Organizational
Review, 47: 587-599. Behavior, 20: 77-140. Stam-
Thomas, R. R., Jr.
ford, CT:JAI Press.
Spangler, E., M. A. Gordon, and 1991 Beyond Race and Gender.
R. M. Pipkin New York:American Manage- Wood, W.
1978 "Token women: An empirical ment Association. 1987 "Meta-analytic review of sex
test of Kanter's hypothesis." differences in group perfor-
Toren, N., and V. Kraus mance." Psychological Bul-
American Journal of Sociolo-
1987 "The effects of minority size letin, 102: 53-71.
gy, 84: 160-170.
on women's position in acad-
Steele, S. emia." Social Forces, 65: Yoder, J. D.
1990 The Content of Our Character. 1090-1100. 1991 "Rethinking tokenism: Look-
New York:St. Martin's. ing beyond numbers." Gen-
Tsui, A. S., T. D. Egan, and C. A.
der and Society, 5: 178-192.
Strangor, C., L. Lynch, C. Duan, O'Reilly IlIl
and B. Glass 1992 "Being different: Relational Ziller, R. C., and R. V. Exline
1992 "Categorization of individuals demography and organization- 1958 "Some consequences of age
on the basis of multiple social al attachment." Administra- heterogeneity in decision-
features." Journal of Person- tive Science Quarterly, 37: making groups." Sociometry,
ality and Social Psychology, 549-579. 21: 198-211.
62: 207-218.
Tyler, T. R., and E. A. Lind Zimmer, L.
Taylor, S. E., and S. T. Fiske 1992 "A relational model of authori- 1988 "Tokenism and women in the
1976 "The token in the small ty in groups." Advances in workplace: The limits of gen-
group: Research findings and Experimental Social Psycholo- der-neutral theory." Social
theoretical implications." In J. gy, 25: 115-191. Problems, 35: 64-77.
Sweeney (ed.), Psychology
Valian, V. Zucker, L. G.
and Politics: Collected Papers:
1 10-1 17. New Haven, CT: 1998 Why So Slow? The Advance- 1987 "Institutional theories of orga-
Yale University Press. ment of Women. Cambridge, nization." Annual Review of
MA: MIT Press. Sociology, 13: 443-464. Palo
Thomas, D. A. Alto, CA: Annual Reviews.
1993 "Racial dynamics of cross- Wanous, J. P., and M. A. Youtz
race developmental relation- 1986 "Solution diversity and the
ships." Administrative Sci- quality of group decisions."
ence Quarterly, 38: 169-194. Academy of Management
1999 "Beyond the simple demogra- Journal, 29: 149-159.
phy-power hypothesis: How Watson, W. E., K. Kumar, and
blacks in power influence L. K. Michaelsen
white-mentor-black-prot6g6 1993 "Culturaldiversity's impact on
developmental relationships." interaction process and per-
In A. J. Murrell, F. Crosby, and formance: Comparing homo-
R. J. Ely (eds.), Mentoring geneous and diverse task
Dilemmas: Developmental groups." Academy of Man-
Relationships within Multicul- agement Journal, 36:
tural Organizations: 157-170. 590-602.
Ogden, UT:Erlbaum.
273/ASQ, June 2001


Ely y Thomas - 2001 - Cultural Diversity at Work The Effects of Diversi

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Ely y Thomas - 2001 - Cultural Diversity at Work The Effects of Diversi

Caricato da

Copyright:

Formati disponibili

Administrative Science

Cultural Diversity at Work: The Effects of Diversity Perspectives on Work Group

The online version of this article can be found at:

Johnson at Cornell University

Email Alerts: http://asq.sagepub.com/cgi/alerts

>> Version of Record - Jun 1, 2001

he or she identifieswith, values, or expresses a particular

diversityincreases the availablepool of resources-net-

group's policies and practices.We also show how these per-

and nationality.We focused on race because it was the

especially attentive to how context might shape people's

Interviewscentered on fourtypes of questions. First,we

Support Staff Middle Managers/Professionals Senior Managers

Percent Number Percent Number Percent Number

239/ASQ, June 2001

The law firm had developed a successful practicein its first

At first, we were like, "[industry name] workers? That's men and

[Therehas to be] a kindof group process of makingsure that

As a result, white members of the programstaff had to learn

If [the firm]were all white, our relationshipswith the community

Formanagementto come into a blackneighborhoodand undertake

I thinkif we were all black,we'd have a lot of obstacles. We

Despite this apparentsymmetry, however, the access-and-

. . . if you perform and exceed expectations, regardless of color, you

I think[the firm],from my vantage point, has made tremendous

Many members of the organization,both white and of color,

Valueof culturalidentity High;a resourcefor Moderate;a resourceonly Low;it is a basis for

cussed. It's not hidden undera rock."One former attorneyof

white attorneys, who sometimes felt less centralto the

Likesometimes people are puttingtogether panels and for good

And so you had this sort of cracker-jackgroupof people who

forced, at least for some, the appropriatenessof the racial

because inadequatesupervisionhad prevented him from tak-

affirmativeaction goals of the firm,"whites did not register

she said, "andthe discussion [amongblackmanagers]was

Enhanced by cross-cultur- Enhanced by increased Inhibited by low morale of

framingthem as women's issues in creative ways that would

ment" others got and were takingtheir business elsewhere

tries' development experiences ratherthan geographical

motivationto achieve (Hackman,1992). Moreover,because

others to which we would liketo generalize our results is

Blau, P M. Dworkin, A., J. Chafetz, and R. Foeman, A. K., and G. Pressley

271/ASQ, June 2001

272/ASQ, June 2001

273/ASQ, June 2001

Potrebbero piacerti anche