Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 299

Undelivered Promises From the HR Profession: A Plea to Return
to a More Defensible Motivation for Embracing Diversity
Ernest. E. Stark
Bellevue University

 Abstract

This paper argues that HR professionals stand to ultimately injure their credibility by making continued unsubstantiated claims regarding return on investment from initiatives to increase workforce diversity.  Building on Williams and O’Reilly’s 1998 original review of over 80 empirical studies concerned with various manifestations of diversity, this paper focuses on studies examining outcomes of gender and racial/ethnic diversity and includes studies omitted by Williams and O’Reilly and studies published since 1997 but before 2002. More significantly, the paper reviews the existing studies reporting the impact of gender and racial/ethnic diversity at the firm and group level of analysis. Finding little evidence of the ability of gender or racial/ethnic diversity to make measurable contributions beyond random chance at the firm or group level, this paper advocates that HR professionals reposition the drive for workplace diversity to proceed from a motivation for social justice and adapt concepts inherent in institutional theory.

Introduction

While the concepts of “diversity” and “multiculturalism” may not be synonymous, the initial concern of corporate human resource professionals for advancement of diversity can be said to have evolved from academia’s endorsement of the philosophical concept of multiculturalism.  Incubated on U.S. campuses, multiculturalism advanced the proposition that while the civil rights movement ended blatant discrimination based on gender, race, ethnicity, religion, and national origin and changed the demographic characteristics of higher education, bringing demographically different groups together in academia without changing the rules was wrong (D’Sousa, 1992).  The existing rules and standards were labeled as manifestly biased by Eurocentrism and white male favoritism (Flowers & Richardson,1996). Across U.S. campuses, an argument was made that new rules needed to be negotiated between demographic groups whereby all were considered equal partners in the debate with equal claims of justice and power (D’Sousa, 1992).

Parallel to the evolution of multiculturalism on U.S. campuses, erosion of the demographic uniformity that had characterized the American workforce for 200 years accelerated both as a result of civil rights legislation and changes in the ethnic and racial composition of the population. Entering the workforce in ever-increasing numbers, women and minorities were reported as having problems adjusting to the workplace culture of their new employers (Hemphill & Haines, 1997). Human resource professionals proposed that the job difficulties reported by the increasing number of women and minorities were not because of their qualifications, but because corporations were patterned on white male heterosexual norms (D’Sousa, 1995).  To address this situation, HR professionals began actively campaigning for businesses to break new ground in accelerating demographic dissimilarity within the workforce (Hemphill & Haines, 1997) and, in doing so, making standards relative throughout the

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 300

workforce so that demographic minority group members not be made to conform to a single organizational standard (Thomas, 1990). Writing for Mosaic, the diversity periodical of the Society for Human Resource Management (SHRM), Gardenswartz and Rowe (1997) forcefully argued that diversity is about culture change.  The fundamental principle behind such change has been reasoned to be that in a multicultural society differences are not a deficit, do not matter (Weaver, 2000), and all cultural values and preferences should be granted equal legitimacy (D’Sousa, 1995).   Simply stated, the original motivation for advancement of diversity throughout U.S. businesses appeared to be the underlying desire for equality and social justice inherent in the concept of multiculturalism.

However, HR professionals became painfully aware that U. S. businesses are noted for their pecuniary motivation and not generally known for allocating scarce resources to moral or social movements.  They were left searching for an argument to convince corporate decision-makers to allocate the resources necessary to advance diversity within the workplace.  In 1991, Cox, Lobel, & McLeod provided HR professionals with a rationale that appeared to establish the focal point from which to leverage diversity in Corporate America.  Their “value-in-diversity” hypothesis argued that diversity would be beneficial for business organizations.  The central assumption underlying the value-in-diversity hypothesis was that a move away from traditional organizational values and norms toward an unqualified valuing of a variety of cultural norms accompanying increased demographic diversity automatically contributes to better productivity and better employee morale (D’Souza, 1995).

HR professionals quickly attached themselves to the value-in-diversity theory and energetically promoted it as an opportunity to harness the potential of a new emerging workforce for increased profitability (D’Sousa, 1995). They became extremely forward in warning of the economic ills about to befall those businesses not actively promoting diversity.  For instance, Thomas (1991) argued that companies had no choice other than to embrace increasing diversity if they wanted to improve employee morale and firm productivity. Fernandez and Bar (1993) issued a strategic imperative for organizational survival by stating that without total acceptance of diversity and a business plan that completely integrated it into corporate strategic plans, a corporation could not succeed in a global market. Lebo (1996) added that in an economic climate characterized by tight budgets and an ever-increasing need to do-more-with-less it is imperative that firms move to unqualified acceptance of diversity. Proceeding from this motivation, HR professionals have been successful in repositioning diversity from a concern for a degree of equity and justice to a hardheaded concern about bottom line issues and profit motives.  However, in the long term, this motivation may not serve either the profession or the concept of diversity well.

Statement of the Problem

The value-in-diversity theory has quickly become a “mantra” among corporate HR professionals.  Weaver (1999), writing for HR professionals, states that investing in a multicultural workforce of professionals will yield long-term dividends for business organizations.  Gray (2002), writing to the same audience, cites the Boston Globe in arguing that corporate diversity is imperative to long-term competitiveness.  Long-Nguyen Robbin (2002, p.6) challenges the same audience of HR professionals by

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 301

asking, “Will your indecisiveness put you behind the competitors who understand and believe the competitive advantage found in diversity?”

The concept of value in diversity has generally been regarded by academics as lacking rigor, theoretical specificity, and a scientific construct primarily because it draws its meaning from the work of organizational practitioners (Nkomo & Cox, 1996).  As Jac Fitz-enz noted, HR professionals are more inclined to report processes than to report outcomes (Bates, 2002). It is the contention of this paper that the central assumption underlying the value-in-diversity hypothesis, that increased diversity within organizations leads to improved productivity and profits, has been widely accepted and publicly promoted by HR professionals without benefit of objective and empirical research. The scant evidence used by HR professionals to advance the value in diversity hypothesis merits closer examination. As Williams and O’Reilly (1998, p.79) clearly noted, “Most of the research supporting the claim that diversity is beneficial has been conducted in laboratory or classroom settings, rather than examining intact working groups within an organizational context.” If, after an extensive review of the existing empirical literature (including both laboratory and field studies) the preponderance of evidence fails to support the value in diversity hypothesis, HR professionals need not abandon their campaign to increase work place diversity but must adopt a more defensible motivation for its advancement.

Method

The purpose of this paper was to more closely examine the central assumption of the value-in-diversity hypothesis.  A meta-analysis was not chosen as the preferred method of research in this study because of often vague and ill-defined variables across the literature. For example, one widely cited study defined the predictor variables as teams having a “majority of males” and teams having “some females” but never clarified how the two groups would differ on the basis of gender.  Such imprecision across existing studies of diversity indicated that a meta-analysis, at this point, would not be desirable but pointed toward the value of a literature review.

The study began by searching for empirical literature reporting the value of diversity at the firm level. The study initially revisited the work of Williams and O’Reilly (1998), who reviewed over 80 studies regarding the effects of various manifestations of diversity as they apply to management and organizations.  This was followed by an extensive literature search conducted by accessing numerous data bases to search for empirical studies (both laboratory and field) on diversity and group performance not included in the Williams and O’Reilly review and works published after 1997 and before 2002. While not necessarily an exhaustive review of the literature, an extensive array of literature was examined in pursuit of the study’s objective. What follows next is a description of how the parameters of the investigation were defined.

Valuing Diversity

Agreement as to an exact definition of what valuing diversity has come to mean within an organizational context proved difficult to achieve.  Where enforcement of the Civil Rights Act of 1964 focused for thirty years almost exclusively on racial and gender issues, valuing diversity has grown to include an even wider array of differences between organizational members such as gender, sexual orientation, ethnicity, tenure, education, and more (Grossman, 2000; Weaver, 2000).  Reflecting the ambiguity of the

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 302

concept, Cross, Katz, Miller, and Seashore (1994) asserted that valuing diversity should expand beyond racism and sexism and focus on heterosexism, classism, ableism, and other forms of discrimination at the individual, identity group, and systems level.   Expansion of the diversity concept to imply that everyone is different and all differences should be valued holds the threat of valuing diversity evolving into a benign and meaningless concept (Bond & Pyle,1998).  Indeed, Grossman (2000) reported an emerging concern that the definition of what it means to value diversity has become so inclusive that it now marginalizes the original issue of gender, race and ethnicity.

Williams and O’Reilly (1998) concluded that most studies of diversity agree that certain demographic characteristics such as race, ethnicity, gender, and age are most visible and likely to be most salient under the majority of circumstances. Thus, salient or visible characteristics such as these are most frequently used for social categorization and most likely to be identified with valuing diversity.  For the purpose of this paper, the definition of diversity was restricted to gender and race/ethnicity, “women and minorities”, and the literature search was conducted accordingly.

Results

Diversity and Organizational Performance

At the heart of the value-in-diversity movement in Corporate America is the proposition that diversity among employees generates human capital that serves as a source of sustained competitive advantage because it creates value that is both difficult to imitate and rare (Richard,2000).  If this proposition is true, evidence should exist demonstrating a positive relationship between gender and racial/ethnic diversity and firm performance.  As argued by Simmons and Nelson (1997), good diversity management practices are undeniably a bottom line issue. Studies examining relationships between gender and race/ethnicity diversity and bottom line indicators of firm performance such as return on equity, increased shareholder value, net operating profits, etc. was sought. However, with four exceptions, little informative literature was identified.

Simmons and Nelson (1997) investigated the relationship between 30 firms recognized for their diversity management practices (their initiatives toward women and minorities) and their return on equity compared to their industry averages.  Their study concluded that the diversity advantage did indeed demonstrate a statistically significance relationship with return on equity, but found little evidence of practical significance in the relationship. That is, there appeared little evidence of a significant effect size. They concluded (p.16), “At this point in time, claims that the issue (increased diversity) is a strategic imperative are not tenable.”

Richard (2000) investigated the relationship between racial diversity and firm performance in the banking industry.  Using a sample of 574 banks, this study found no direct effect of racial diversity on measures of financial performance.  However, Richard did report an interaction effect between racial diversity and business strategy such that firms with racial diversity and a strategy of growth experienced higher return on equity than did firms with the same level of diversity and a no-growth or downsizing strategy.

The most widely cited evidence of a relationship between gender and racial/ethnic diversity and firm performance is Greenberg, Canzoneri, and Smith’s (1998) research report published exclusively for the members of the American Management Association (AMA). In this study, 1000 AMA executives submitted profiles of their senior

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 303

management teams (SMT) and their organization’s recent performance regarding shareholder value, total assets/book value, net operating profits, worker productivity, gross revenues, and market share. The study purported to present evidence that a mixing of gender and racial/ethnic backgrounds in senior management teams consistently correlated with superior corporate performance.  Unfortunately, the results presented in this report were fraught with evidence of poorly constructed research. 

It was never confirmed if the study was submitted for peer review as is common for publication of academic research.  It was evident that the demographic groupings in this study were not exclusive and exhaustive as expected in rigorous empirical research.  Thus, it was difficult to understand how the demographic groupings delineated diverse from non-diverse senior management teams.

While the authors of the study announced that they had found consistent correlations between diversity and firm performance, the results were presented in tabular format devoid of any statement of correlation values.  There was no basis from which to determine how much of what the study claims to have found should be attributed to random chance and how much could truly be linked to diversity in senior management teams.

Most disturbing is that when using the study’s own rubric for determining a relationship between diversity on SMT and firm performance (differences in the percentage of firms reporting increases on various measures of firm performance based on demographic grouping), the results were contradictory and confusing.

An argument might be made that empirical studies of diversity and organization performance have failed to examine the potential of a diverse work force to increase sales to diverse market segments.  Indeed, Cox (1991) advocated that the cultural heterogeneity resulting from increasing diversity leads to more successful marketing and sales to different types of customers.  This seemingly logical assumption does not appear to receive support from a major study of Fortune 500 firms by Gomez-Mejia and Palich (1997). In a ten year study (1985-1994) of the ability of culturally diverse international initiatives to predict firm performance, the authors specifically addressed both the promise and the problems associated with cultural diversity in global undertakings and the development and administration of human resources program.  The study concluded that neither the cultural diversity nor the cultural homogeneity of international ventures made a difference on firm performance.

An argument could be advance that the 2001 settlement of $192.5 million agreed to by Coca-Cola and the 1994 settlement of  $171.6 million agreed to by Texaco for alleged discrimination in employment practices stands a evidence of the value in diversity argument at the firm level (Lawrence, 2001). While neither of these cases went to court, one should conclude that there was sufficient evidence to present a prima facie case of discrimination in employment practices, and neither company desired to assume the burden of responsibility in court. This paper argues that such cases do not represent issues associated with diversity as defined by Weaver (2000), but rather questions regarding affirmative action and EEO compliance. More specifically, this paper argues that these cases represent the financial consequences that must be extracted from firms when they fail to comply with existing laws regarding employment practices. Further, there is no evidence in either of these cases that either gender or racial/ethnic

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 304

diversity made any measurable contribution to corporate performance beyond what should be expected from compliance with existing legal statutes.

An argument might be forwarded that firms have embraced diversity because, whether they can prove it or not, it is economically rational for them to do so. Further, one might assume firms who do not embrace diversity are at a competitive disadvantage, whether they realize it or not. Those who might advance this argument should do so with caution. Walsh (2000) reported that General Electric, whose revenues rank among the top 10 companies in the US, whose stock value nearly tripled in 3 yrs, and whose management practices became the “gold standard” (Walsh’s words) for other companies, had at the time only one African-American among its 31 member senior management team and no females, Hispanics, or a Asians.  It does not appear that the lack of gender or racial/ethnic diversity at the senior management level created a competitive disadvantage in this case.

The few relationships reported between diversity and the bottom line at the firm level are, in the words of Kuczynski (1999), “complex, and not necessarily positive.”  However, the aforementioned complexity suggests the advisability of looking deeper for potential indicators of a relationship between diversity and performance at the firm level.

Diversity and Performance at the Group Level

Finding no conclusive evidence of a relationship between diversity and firm performance, this study proceeded on the assumption that group functioning might serve as a reasonable proxy for organizational performance. That is, it appeared reasonable that a multitude of outcomes at the group level might aggregate at some point to impact a firm’s performance. It was believed wise to reflect on Gladstien’s observation of nearly twenty years ago (1984) that one must consider not only group-produced outputs, but also the consequences of the group for its members and the group’s capacity to perform in the future. With this in mind, studies were sought reporting the relationship between gender and race/ethnicity diversity and both group processes and group outputs.

Gender Diversity and Group Processes

In general, the literature does not provide much encouragement for those seeking evidence of a positive relationship between gender diversity and improved group processes.  Williams and O’Reilly (1998) reviewed a number of studies reporting that mixed-gender workgroups experienced higher levels of conflict, interpersonal tension, and lower levels of friendliness than same gender work groups (e.g. Alagana, Reddy, & Colllins, 1982; Clement & Schiereck, 1973; Holahan, 1979, Sackett, BuBois, & Noe, 1991; Pelled, 1997).  More recent work by Shipley and Aven (1997) reported that the more gender diverse (genders approaching equality in numbers) a work group the more the decision-making style tended to be reflexive, a style in which groups make quick decisions without gathering all information or considering many alternatives. Over an extended period of time, a decision-making style such as this would not appear to have a strong potential to contribute to improved organizational performance. Further, Chatman and Brown (1998) reported finding that variance in the intensity of a friendship between 200 MBA students demonstrated a significant positive relationship with gender similarity and identification with gender.  Admittedly a laboratory experiment, the results of this study suggested that diminished gender identification or diminished similarity

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 305

predicted lower levels of friendship intensity.   Such reduced levels of friendship does not bode well for the generation of social capital that Nahapiet and Ghosal (1998) and Uhl-Bien, Graen, and Scandura (2000) argue facilitates achievement of organizational performance that is otherwise impossible.

However, there are studies indicating that a negative association of gender diversity with group processes might be confounded by other effects.  For instance, Williams and O’Reilly (1998) cited studies (e.g., Abrams, Thomas, & Hogg, 1990; Ely, 1994; Konrad, Winter, and Gutek, 1992) suggesting that the relationship between gender diversity and group process is likely dependent upon the proportion of men and women present in the unit, not simply on group heterogeneity. Work by Allmendinger and Hackman (1995) reported a negative relationship between the percentage of females in the work unit and member perceptions of the quality of their work relationships. More recent work by Knouse and Dansby (1999) has suggested that up to a point where the gender mix of an organization exceeds 30% females, shared perceptions of group effectiveness improves.  Beyond that point, male perceptions of group effectiveness decline while women perceive a higher level of group effectiveness up to 50% gender diversity.  Whether the perceptions of group processes reported in these two studies were substantiated by actual measures was not reported, but reduced perceptions of group effectiveness would appear to have little potential to enhance a group’s ability to contribution to improved firm performance.

With the exception of the Knouse and Dansby (1999) study, research reporting a positive relationship between gender diversity and group processes was not found. However, neither was there extensive replication of the reported negative relationships between gender diversity and group processes.  For instance, neither Pelled (1996) nor O’Reilly, Williams and Barsade (1997) found any relationship between gender diversity and conflict.  In a meta-analysis of case studies, gender was not found to be associated with worker solidarity (Hodson,1995), and Bradley (1997) reported that gender was not a significant predictor of either subordinate or supervisor behavior in performance evaluation interviews, and the study provided no reason to assume any relationship between the level of gender diversity, behaviors involved with performance evaluation, and ineffective group processes.

In summary, the most optimistic conclusion one might reach from the existing literature is that there is no relationship between gender diversity and group processes.  There is certainly evidence that could justify a more negative conclusion.  In either case, there is little empirical evidence to support a position that increasing the gender diversity of a group enhances group processes, especially to a point where it is likely to influence a firm’s bottom line.

Gender Diversity and Group Performance

One of the most cited works in defense of gender diversity is Hoffman and Maier’s (1961) laboratory study of 41 four-person groups that suggested that gender diversity improved the quality of group solutions on five cognitive tasks.  However, Williams and O’Reilly (1998) concluded from their extensive review of laboratory studies that the effects of gender diversity on group performance involving cognitive or creative tasks were, at best, mixed.  Citing work (e.g. Clement and Schiereck ,1973; Kent and McGrath, 1969; Hoffman, Harburg and Maier, 1962) Williams and O’Reilly (1998)

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 306

suggested that homogeneity of gender, and not heterogeneity, may promote creativity.  These studies raise important questions about the potential of increasing levels of gender diversity to enhance group performance in firms requiring high levels of creativity.   In a related vein, a recent laboratory study by Chatman, Berdahl, Boisnier, Spataro, and Anderson (2000) found that majority male groups (with a lone female) out performed gender-balanced groups on math and verbal ability problem sets.  However, on a contradictory note, they also found that the gender-balanced groups outperformed majority-female groups (with a lone male).

In the field, a single study reported evidence of positive group outcomes associated with gender diversity. Wharton and Bird (1996) reported a study of gender in university departments where men in mixed-gender work groups were more satisfied with their jobs than men in all male groups. Reviewing field studies by Pelled (1997), Kizilos, Pelled, and Cumming (1996), and O’Reilly, et al. (1997), Williams and O’Reilly (1998) summarized findings that appear to contradict the work of Wharton and Bird.  They concluded that increasing gender diversity demonstrates not a positive, but a negative, relationship with group performance when the sample is male-dominant and no relationship when the sample is female-dominant.  Their review suggested that gender diversity in male-dominant situations leads to increased levels of emotional conflict (resulting in decreased levels of productivity) and lower levels of prosocial behavior toward the customer (employee behavior beyond the requirements of the job). A more recent study by Harrington (2000) concluded that, as implied by prior research, gender diversity does indeed frequently have a negative direct effect on group performance.  However, Harrington suggested that indirect positive effects of gender diversity, enhancement of task orientation and the expression of diverse opinions might, possibly, counteract such direct negative effect.

There are also studies suggesting that as the gender mix of an organization increases in favor of females, males are more likely to demonstrate reduced commitment and less attachment to an organization (Riordan, 1997; Shipley & Aven, 1997). Tusi, Egan, and O’Reilly (1992) reported that sexual heterogeneity was associated with men’s, but not women’s, intention to leave the organization while Burke and McKeen (1996) reported that the proportion of men in an organization was positively related to the intention of managerial and professional women to quit. Pelled, Ledford, Jr., and Mohrman (1998) reported that increasing gender dissimilarity was negatively associated with male reports of job security while no such relationship was found for females.  Finally, a recent study by Humphrey, Sleeth, Kellet, and Showalter (2000) concluded that the greater the visibility of females within a group, the greater the level of conflict reported by men in the group.

Reduced employee commitment and organizational attachment, increased intention to exit the firm, and increased levels of conflict are not likely to have a positive relationship to a work group’s contribution to firm performance. It is significant to note that a meta-analytic integration of 57 effect sizes from 13 studies failed to find convincing evidence that building gender diverse work teams results in significant gains in team performance (Bowers, Pharmer, & Salas, 2000). In the absence of convincing evidence that gender diversity improves measures of group performance, the literature appears to provide little evidence to support the assumption that gender diversity produces group outcomes

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 307

that, aggregated over time, are likely to make a contribution to a firm’s bottom line beyond random chance.

Racial Diversity and Group Processes

An optimistic view of valuing diversity holds that workplace diversity has positive effects on group processes. For instance, valuing diversity is advanced on the proposition that racial/ethnic diversity will enable differences in perspectives, expertise, and attitudes to be widely communicated across all team members resulting in outcomes that are more acceptable to everyone (Sessa & Jackson, 1995).  Unfortunately, research has not demonstrated the presence of such communication between team members. Rather, communication appears to suffer as a result of racial/ethnic diversity.  More than twenty years ago work by Lincoln and Miller (1979) reported a negative relationship between racial/ethnic diversity and levels of communication. Hoffman (1985) reported that informal meetings among peers and with immediate superiors are fewer in racially diverse groups, thus reducing the diffusion of information across organizational boundaries.  Further, Pelled et al. (1998) reported that race dissimilarity demonstrated a significant negative relationship with access to organizational information. In view of Langan-Fox’s (2001) observation that communication is the major causal variable determining many organizational outcomes, the failure of increased racial/ethnic diversity to improve communication within groups does not bode well for such diversity to make a positive contribution beyond random chance to a firm’s performance.

Williams and O’Reilly (1998) proposed that there are likely barriers in the communication process within racially and ethnically diverse groups that inhibit, but not make impossible, the ability of racially diverse groups to contribute to performance at the firm level.  For instance, their review of the literature suggested that minorities may be more likely to choose same-race friends or be more satisfied when spending time with others of ethnic similarity (e.g. Mehra, Kilduff, & Brass, 1996; Verkuyten, de Jong, & Masson, 1993).  Further support for that position was found in work by Jehn, Northcraft, and Neale (1996) and Chatman and Brown (1998).  Citing other research (e.g. Ibarra ,1995), Williams and O’Reilly (1998) speculated that minorities often have more extensive communication networks designed to tie into other minorities located outside the immediate work unit.

Another optimistic outcome associated with valuing diversity is that increased racial/ethnic diversity will introduce non-Anglo cultures into the American workplace.  Cox et al. (1991) proposed that the introduction of non-Anglo cultures leads to increased cooperation and result in collectivist cultures having the potential to be more productive than the highly individualistic and competitive Eurocentric culture that currently dominates U.S. businesses. In their often cited laboratory study, Cox et al. examined differences between Asian, Hispanic, black Americans, and Anglo-Americans in cooperative behavior using a Prisoner’ Dilemma game.  They reported that minority group members were more cooperative than Anglo-Americans and that diverse groups were significantly more cooperative than the homogeneous groups composed of all Anglos.  Movement toward such collectivism and cooperation by increasing diversity efforts was assumed to have potential to enhance overall firm performance.

However, Williams and O’Reilly (1998) suggested that there are other studies that should temper interpretation of the Cox, et al. (1991) report.  For instance, Espinoza

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 308

and Garza (1985) studied Anglo- and Hispanic-Americans in a laboratory experiment and found that members of each demographic group were equally cooperative when their own group was in the numerical majority.  When in the minority, Hispanics were found to be more competitive, while Anglos were less competitive when they were in the minority. Further, a laboratory study by Garza & Santos (1991) found that in interdependent payoff games Hispanics were inclined to act in ways that benefited their own ethnic group at the expense of Anglos.  Anglos, on the other hand, were driven mainly to earn the most points possible disregarding potential benefits to their own ethnic group. Thus, the results reported in the Cox et al. (1991) study have not been substantiated by additional research.

A pessimistic view of racial/ethnic diversity is found in social identity theory that predicts that increasing racial and ethnic diversity will be associated with conflict and negative influence on group processes (Nkomo and Cox, 1996).  Williams and O’Reilly’s (1998) review of the literature was inconclusive on this matter.  Some studies (i.e. Pelled, 1993); Pelled, Eisenhardt, & Xin, 1997) were cited as providing evidence of a positive relationship between racial diversity and levels of emotional conflict.  However, other studies (i.e. Pelled, 1997; O’Reilly, et al, 1997) were interpreted as finding no strong effect of racial diversity on conflict.  Finally, Watson, Kumar, and Michaelsen (1993) reported a study of newly formed ethnically homogeneous and heterogeneous work groups.  They noted that, initially, the culturally homogeneous work groups reported more process effectiveness than the heterogeneous work groups.  However, reported group process effectiveness improved for both cultural groups over time with essentially no difference (no advantage or disadvantage for either group) at the conclusion of the study.

In summary, the claims of improved group processes associated with racial and ethnic diversity appear to be at best inconclusive, if not unfounded.  However, there may be an underlying reason for this apparent lack of support.  Citing numerous other studies (Riordan & Shore, 1997; Tsui, Egan, & O’Reilly, 1992; Wharton & Baron, 1987), Williams and O’Reilly (1998) suggested that the proportion of racial and ethnic diversity may matter for the positive effects ascribed to racial and ethnic diversity, just as they appear to do for gender effects. That is, increasing racial/ethnic diversity up to a certain ratio may result in improved group processes, but beyond that point important group process are likely to suffer decline. 

Racial Diversity and Group Performance

Many of the optimistic assumptions regarding racial/ethnic diversity and group performance resulted from two laboratory studies.  McLeod and Lobel (1992) reported that while ethnically diverse groups did not necessarily produce more ideas when performing a brainstorming task, the ones they did generate were rated of higher quality than those produced by homogeneous groups.  They interpreted their findings as proof of the positive impact of diversity.  In a second study, Watson et al. (1993) found that at the end of a 17-week term, ethnically diverse groups scored higher on two of four task measures (the range of perspectives considered and alternatives generated) than homogeneous groups when engaged in structured case evaluations. However, it is important to note that in the study’s first three measurement periods, homogeneous groups scored higher than diverse groups on all performance measures. Further, in the final measurement period there was no significant difference in the performance of either the ethnically homogeneous or heterogeneous group.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 309

A somewhat more pessimistic view of the value of racial/ethnic diversity was reported in a laboratory study by Thomas (1999). In this study, 77 undergraduates in New Zealand were placed into groups in which all characteristics except nationality were the same.  Each group was engaged in the evaluation of five case studies and group effectiveness and group performance were measured on each case.  Results of this study found that the culturally homogeneous groups performed better than the heterogeneous groups across all five cases. Thomas interpreted these results as suggesting that cultural heterogeneity resulted in process losses that influenced the performance of the groups and that the culturally heterogeneous groups were unable to overcome these losses to achieve their performance potential.

While a field study conducted by O’Reilly, Williams, and Barsade (1997), reported a positive relationship between racial/ethnic diversity and both creativity and implementation ability of the groups, after controlling for the moderating effects of conflict, Williams and O’Reilly (1998) concluded that most field studies (e.g. Kizilos, Pelled, & Cummings, 1996; Pelled, 1997; Pelled, Eisenhardt, & Xin, 1997) have reported either no effect of racial/ethnic diversity on performance or weak negative effects.  Work by Tusi, Egan, and O’Reilly (1992), Riordan (1997), Shipley and Aven (1997) indicated that racial/ethnic diversity in work groups often demonstrates a negative relationship with organizational commitment, absenteeism, turnover, and job satisfaction, all believed to have a negative relationship with indicators of firm performance.

Williams and O’Reilly’s (1998) review of the literature further suggested that Whites experience more negative effects than non-Whites when faced with racial/ethnic diversity.  The Tsui et al. (1992) study concluded that the more minorities in a work unit, the less commitment Whites were likely to express toward their company, but Whites working in units with no minorities were the most attached to their company.  Interestingly, minority attitudes toward the company were not affected by the number of Whites in the work unit.  In a more recent study, Muller, Finley, Iverson, and Price (1999) reported that White teachers in White teacher and White student dominant schools demonstrated higher job satisfaction than White teachers in schools where both White teachers and White students were in a minority. However, they found no job satisfaction differences among Black teachers across schools with differing racial composition.  Interestingly, the authors reported that Black teachers in non-Black schools were less committed to teaching careers than White teachers in White dominant schools.

In summary, the literature appears to provide little evidence to support the assumption that racial/ethnic diversity produces group outcomes that, aggregated over time, are likely to make a long-term contribution to a firm’s bottom line beyond that of random chance. However, as noted by Williams and O’Reilly’s (1998) and Knouse and Dansby, (1999) the proposed positive effects of racial and ethnic diversity on group performance may rest with the proportional representation of the racial and ethnic subgroups.  As to what the optimal proportion might be, additional study is required.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 310

Why HR Professionals Have Not at Present, and May Never, Demonstrate a Clear Relationship Between Diversity and Bottom Line Issues

There are numerous reasons why continued promotion of valuing diversity under the banner of “economic necessity” should be reconsidered by HR professionals.  First, the profession stands accused of currently lacking the discipline, the tools, and the science to establish such a claim (Bates, 2002).  Second, it is no secret that while businesses measure return-on-investment, diversity has been considered a soft concept whose returns are difficult to gauge at the firm level of analysis (Digh,1999). This paper agrees that a causal effect between diversity and firm performance will be extremely difficult to prove.  Considering all the variables known or suspected of impacting a firm’s bottom line, the potential of diversity to explain a significant portion of variation in a firm’s financial performance would appear very small.  Third, diversity is a rather complex concept, and its application will not likely produce consistent and universal results either at the firm or group level of analysis. Situational moderators may determine when diversity is capable of producing organizational results and when it is unreasonable to expect diversity to do so (Richard, 2000). One should expect some situations to favor homogeneous units and others to favor heterogeneous units. There are even situations when rather than celebrating diversity by emphasizing individualism and distinctiveness among members it is advisable to encourage people to categorize one another as having the organization’s interest in common (Chatman, Polzer, Barsade, & Neal, 1998).

Finally, the assumed contribution of diversity to improved processes and outcomes at the group level may be the result of personality, not gender and racial/ethnic diversity.  Monynihan and Peterson (2001) noted that researchers have traditionally been drawn to the use of demographic variables in organizational research because of the relative ease of data collection and have often assumed that these variables are proxies for the values, perspectives, or cognitive orientation of an individual that affect group processes and performance.  They argue (p. 329) that personality traits are closer antecedents of cognitive and affective orientations than demographic variables and these are likely to be more powerful predictors of group processes and performance than demographic characteristics. They state that empirical studies show that over time, surface level demographic differences in a group will be transcended by deeper-level values associated with personality traits, and as such, personality traits become important predictors of the processes of intact teams.  Thus, if group processes and group performance can aggregate over time to influence performance at the firm level, the relationship may have more to do with the personalities of the employees than their gender or racial/ethnic classification.

A Plea for a Defensible Motivation

While the value-in-diversity argument may have some relationship with bottom line issues, there is relatively little empirical evidence that the reality matches the rhetoric from HR professionals. Unfortunately, HR professionals have demonstrated a tendency to promote the effectiveness of diversity while unable to back up their beliefs with numbers (Thomas, 1998). The profession would be well advised to reflect on an observation made by Amy Hillman, management professor at the University of Western Ontario, as reported by Kucyznski (1999): …there is no clear proof that diversity causes better firm performance in the market place….we think there is something there, but we

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 311

can’t rule out a whole bunch of different factors.  It is the position of this paper that continued promotion of value-in-diversity as a strategic imperative when faced with repeated failure to “show me the money” can only decrease the credibility of the HR profession, a profession that has consistently been given little credibility and respect at the executive level (see Stewart, 1996).

Rather, this paper proposes that the HR profession has at its disposal a much more defensible motivation from which to advance the concept of diversity.  Consider the words of Johnetta Cole (professor of anthropology at Emory University and former board member of Coca-Cola and Home Depot) as reported by Kuczynski (1999): There are likely times when the bottom line is not being aggressively served by diversity initiatives; however, regardless of the bottom line issue, diversity in the workplace may simply be the right thing to do.

Cole’s statement points to a need for a return to a more basic and fundamental motivation, that of a desire for “social justice.” A social justice motivation would acknowledge the immorality of attempting to maintain a homogeneous workforce within a gender and racially/ethnically diverse nation.  Further, this motivation would acknowledge the subtle forms of sexism and racism that permeate our society and argue that businesses have a social responsibility to address and reduce such discrimination by generating new diversity initiatives and supporting existing ones.  A case should be made that in diversity initiatives business organizations contribute to changing core societal values. In doing so, corporate America responds not to a strategic business imperative but to a higher moral imperative. To paraphrase Cole, corporate America moves from “looking good” to “doing good” (Kuczynski,1999).  In some sense, this is a return to the fundamental motivation from which diversity was initially advanced by HR professionals.

HR professionals should not fear that abandoning dubious attempts to make a business case leaves no avenue for advancement of diversity.   A familiarity with institutional theory provides concepts that will likely prove critical in achieving social justice and increasing work place diversity.  First, institutional theory consents to the assumption that organization decision-makers are not bounded by rationality in the decisions they make (Tolbert & Zucker, 1996).  That is, decisions are made from a multitude of motivations, and rationality is open to subjective interpretation.  This allows for the possibility that the decision to embrace diversity does not require a bottom line justification.

Second, institutional theory focuses on the political processes through which actors and their interests and motives are constructed and mobilized in support of certain organizational logics as opposed to others (Reed, 1996).  That is, practices within organizations are seen as structures whereby powerful people are committed to some value or interest. This implies that advancement of diversity would be well served when the motivation for social justice is aligned with the needs and interest of powerful others within the organization.  As Stienchcombe (1968) noted, power has a great deal to do with the historical preservation of patterns and values.  Such a perspective allows the HR professional to avoid losing credibility by promoting continued unsubstantiated economic claims about the value in diversity and focus on the more productive realities of political alignment. In other words, the challenge for HR professionals is to acquire

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 312

the political savvy necessary to secure access to powerful others and the inspiration essential to forge a linkage between the social justice inherent in enhanced workplace diversity and the specific motives and interests of such powerful others.

Third, institutional theory holds that organizational decision-making produces structures and practices that have symbolic as well as action-generating properties (Tolbert & Zucker, 1996). Such structures and practices become invested with socially shared meaning that communicates their legitimacy to both internal and external audiences. That is, as programs, policies, and practices diffuse throughout and across organizational boundaries, they come to be seen as legitimate.  This implies that the most critical role HR professionals can play in the advancement of diversity is to be found in advancing diffusion of the legitimacy of the social justice achieved by embracing diversity.

If there is an economic case to be made for diversity, it will not likely be realized until businesses learn to integrate skilled and talented employees of different gender, race, and ethnicity into a single, cohesive corporate culture that enables organizations to thrive in a global market (Weizmann & Weizmann,2000).   Academics and HR professionals alike should heed the admonition made over a decade ago that until we have greater clarity about the theoretical underpinnings, it will be difficult to design processes to fully capture the value in diversity (Murninghan & Conlon, 1991).

References

Abrams, D., Thomas, J., & Hogg, M. (1990):  Numerical distinctiveness, social identity, and gender salience.  British Journal of Social Psychology, 29, 87-92.

Alagana, S., Reddy, D., & Collins, D. (1982).  Perceptions of functioning in mixed-sex and male medical training groups.  Journal of Medical Education, 57, 801-803.

Allmendinger, J., & Hackman, J. R. (1995).  The more, the better? A four-nation study of the inclusion of women in symphony orchestras. Social Forces, 74, 423-60.

Bates, S. (2002, Aprin). HR: Learn to measure success. HR-News, p. 7.

Bond, M. A., & Pyle, J. L. (1998).  Diversity dilemmas at work.  Journal of Management Inquiry, 7, (3), 252-269.

Bowers, C. A., Pharmer, J. A., & Salas, E. (2000). When member homogeneity is needed in work teams: A meta-analysis. Small group research, 305-327. Thousand Oaks, CA: Sage.

Bradley, L. M. (1997).  Using theory to structure qualitative research: An investigation of gender and identity in performance appraisal.

Burke, R. J., & McKeen, C. A. (1996).  Do women at the top make a difference? Gender proportions and the experiences of managerial and professional women.  Human Relations, 49, (8), 1093-1104.

Burt, R., & Reagans, R. (1997).  Homophily, legitimacy, and competition: Bias in manager peer evaluations.  Working paper, Graduate School of Business, University of Chicago.

Chatman, J. A., & Brown, R. A. (1998).  Demographic similarity, salient social identities, and friendship.  Working paper, Hass School of Business, University of California, Berkeley.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 313

Chatman, J., Berdahl, J., Boisnier, A., Spataro, S., & Anderson, C. (2000). The typical, the rare, and the outnumbered: Distinguishing the effects of historical typicality and the numerical distinctiveness in workgroups.

Chatman, J., Polzer, J. T., Barsade, S. C., & Neale, M. A. (1998). The influence of demographic composition and organizational culture on work processes and outcomes.  Administrative Science Quarterly, 43, 749-780.

Clement, D., & Schiereck, H. (1973).  Sex composition and group performance in a signal detection task.  Memory and Cognition, 1, 251-255.

Cox, T. Jr. (1993). Cultural diversity in organizations. Theory, research, and practice.  San Francisco, CA:  Berrett-Koehler Publishers.

Cox, T. (1991). The multicultural organization. The Academic of Management Executive, 5, 34-47.

Cox, T., Lobel, S., and McLeod, P. (1991).  Effects of ethnic group cultureal differences on cooperative and competitive behavior on a group task.  The Academy of Management Journal, 34, 827-847.

Cross, E. Y., Katz, J. H., Miller, F., & Seashore, E. W. (1994).  The promise of diversity.  Burr Ridge, IL: Irwin.

D,Sousa, D. (1992).  Illiberal education: The politics of race and sex on campus.  New York, NY: Vintage Books.

D,Sousa, D. (1992).  The end of racism.  New York, NY: The Free Press.

Digh, P. (1999). Creating a new balance sheet: The need for better diversity metrics.  Mosaics, SHRM focuses on workplace diversity, 5, 1-5.

Diversity management is a cultural change, not just training (1993, May 26). Chicago: Commerce Clearing House.

Ely, R. (1994).  The effects of organizational demographics and social identity on relationships among professional women.  Administrative Science Quarterly, 39, 203-238.

Espinoza, J., & Garza, R. (1985).  Social group salience and inter-ethnic cooperation.  Journal of Experimental Social Psychology, 21, 380-392.

Ferdman, B. (1992). The dynamics of ethnic diversity in organizations: Toward integrative models.  In K. Kelly (Ed.), Issues, theory, and research in industrial/organizational psychology, (pp. 246-277). Amsterdam: North Holland.

Fernandez, J. P. (1991). Managing a diverse workforce.  Lexington, MA: Lexington Books.

Fernandez, J. P., & Barr, M. (1993).  The diversity advantage: How American business can out-perform Japanese and European companies in a global marketplace. New York: Lexington Books.

Flowers, B. J., & Richardson, F. C. (1996).  Why is multiculturalism good?  The American Psychologist, 51, (6), 609-621.

Gardenswartz, L., & Rowe, A. (1997). Diversity Q & A: Effectively moving beyond affirmative action with a diversity task force.  Mosaics, SHRM focuses on workplace diversity, 3, 5.

Garza, R., & Santos, S. (1991).  Ingroup/outgroup balance and interdependent inter-ethnic behavior.  Journal of Experimental Social Psychology, 27, 124-137.

Gladstien, D. (1984).  A model of task group effectiveness.  Administrative Science Quarterly, 29, 499-517.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 314

Gomez-Mejia, L. R. & Palich, L. E. (1997). Cultural diversity and the performance of multinational firms. Journal of International Business Studies, 28  309-335.

Gray, C. (2002). Employees a-plenty: The emerging workforce of people with disabilities.  Mosaics, SHRM focuses on workplace diversity, 8, 1 (4:7).

Greenberg, E. R., Canzoneri, C., & Smith, C (1998).  Senior management teams, Profiles and performance.  American Management Association, New York, NY.

Greenhaus, J., Parasuraman, S., & Wormley, W. (1990).  Effects of race on organizational experiences, job performance evaluations, and career outcomes.  Academy of Management Journal, 33, 64-86.

Grossman, R. J. (2000). Is diversity working?  HR Magazine, 45, (3), 47-50.

Harrington, B. (2000). Cohesion, conflict, and group demography: A multi-method integration. A paper presented to the annual convention of the Academy of Management. Toronto, Ontario, Canada.

Hemphill, H., & Haines, R. (1997). Discrimination, harassment, and the failure of diversity training: What to do now? Westport, CT: Quorum Books.

Hodson, R. (1995).  Cohesion or conflict? Race, solidarity, and resistance in the work place.  Social Work, 5, 135-59.

Hoffman, E. (1985).  The effect of race-ratio composition on the frequency of organizational communication.  Social Psychology Quarterly, 48, 17-26.

Hoffman, L, Harburg, E., & Maier, N. (1962).  Differences and disagreement as factors in creative group problem solving.  Journal of Abnormal Psychology, 64, 206-214.

Hoffman, L., & Maier, N. (1961).  Quality and acceptance of problem solutions by members of homogeneous and heterogeneous groups.  Journal of Abnormal and Social Psychology, 62, 401-407.

Holahan, C. (1979).  Stress experienced by women doctoral students, need for support, and occupational sex typing: An interactional view.  Sex Roles, 5,  425-436.

Humphrey, R. H., Sleeth, R. G., Kellet, J., & Showalter, E. D. (2000).  How demographic factors and personality traits influence complex task choice and leadership emergence. A paper presented to the annual convention of the Academy of Management. Toronto, Ontario, Canada.

Ibarra, H. (1995). Race, opportunity, and diversity of social circles in managerial networks.  Academy of Management Journal, 38, 673-703.

Jackson, S., Stone, V., & Alvarez, E. (1993).  Socialization amidst diversity: Impact of demographics on work team oldtimers and newcomers.  In L. Cummings and B. Staw (Eds.), Research in organizational behavior, 15. Greenwich, CT: JAI Press.

Jehn, K. A., Northcraft, G. B., & Neal, M. A. (1996).  Opening Pandora’s Box: A field study of diversity, conflict, and performance in workgroups.  A paper presented to the annual convention of the Academy of Management. Cincinnati, OH.

Johnston, W. B., & Packer, A. E. (1987).  Workforce 2000: Work and workers for the 21^st century.  Indianapolis, IN: Hudson Institute.

Kent, R., & McGrath, J. (1969).  Task and group characteristics as factors affecting group performance.  Journal of Experimental Social Psychology, 5, 429-440.

Kizilos, M., Pelled, L., & Cummings, T. (1996).  Organizational demography and prosocial behavior.  Unpublished manuscript.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 315

Knouse, S. B., & Dansby, M. R. (1999).  Percentage of work-group diversity and work-group effectiveness.  The Journal of Psychology, 133, 486-494.

Konrad, A., Winter, S., &  Gutek, B. (1992).  Diversity in work group sex composition. Research in the Sociology of Organizations, 10, 115-140.

Kuczynski, S. (1999). If diversity, then higher profits?  HRM Magazine, 44, 66-74.

Lawrence, L. (2001, January). Coca-Cola agrees to record discrimination settlement. HR News, 1,4.

Lebo, F. (1996).  Mastering the diversity challenge: Easy on the job measures. Del Ray Beach, FL: St. Lucie Press.

Lincoln, J. R., & Miller, J. (1979). Work and friendship ties in organizations: A comparative analysis of relational networks.  Administrative Science Quarterly, 24, 181-199.

McLeod, P., & Lobel, S. (1992).  The effects of ethnic diversity on idea generation in small groups. A paper presented at the Annual Academy of Management meeting, Las Vegas, NV.

Mehra, A., Kilduff, M., & Brass, D. (1996).  Relegation to the margins: Race and gender differences in network positions, friendship strategies, and social identity.  Unpublished working paper.

Moynihan, L. M. & Peterson, R. S. (2001). A contingent Configuration Approach.  In B. M. Staw and Robert I. Sutton (Eds.), Research in Organizational Behavior, 23, (338-356).  New York, NY: JAI Press.

Muller, C. W., Finley, A., Iverson, R. D., & Price, J. L. (1999).  The effects of group racial composition on job satisfaction, organizational commitment, and career commitment: The case of teachers.  Work and Occupations, 26, 187-219.

Murninghan, K., & Conlon, E. (1991).  The dynamics of intense work groups: A study of British string quartets. Administrative Science Quarterly, 36, 165-186.

Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage.  Academy of Management Review, 23, 242-266.

Nkomo, S. M., & Cox, Jr., T (1996). Diverse identities in organizations.  In S. W. Clegg, C. Hardy, and W. R. Nord (Eds.), Handbook of Organizational Studies, (338-356).  Thousand Oaks, CA: Sage Publications.

O’Reilly, C., Williams, K., & Barsade S. (1997). Group demography and innovation: Does diversity help?  In E. Mannix and M. Neal (Eds.), Research in the management of groups and teams, (Vol. 1).  Greenwich, CT: JAI Press.

Pelled, L. (1993).  Team diversity and conflict: A multivariate analysis.  A working paper, School of Business Administration, University of Southern California.

Pelled, L. (1996). Demographic diversity, conflict, and work group outcomes: An intervening process theory.  Organization Science, 7, 615-631.

Pelled, L. (1997).  Relational demography and perceptions of group conflict and performance: A field investigation.   International Journal of Conflict Resolution, 7, 230-46.

Pelled, L. H., Ledford, Jr., G. E., & Mohrman, S. A. (1998).  Individual demographic dissimilarity and organizational inclusion: A field investigation.  Academy of Management Proceedings, OMT: A1-A17.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 316

Pelled, L., Eisenhardt, K., & Xin, K. (1997).  Demographic diversity in work groups: An empirical assessment of linkages to intragroup conflict and performance.  A working Paper, School of Business Administration, University of Southern California.

Reed, M. (1996). Organizing theory: A historically contested terrain. In S.R. Clegg, C. Hardy, and W. R. Nord (Eds.), The handbook of organizational studies,  (32-56). Thousand Oaks, CA: Sage.

Richard, O.C. (2000). Racial diversity, business strategy, and firm performance: A resource-based view.  Academy of Management Journal, 43, 164-177.

Riordan, C. M. (1997). Relational demographic theory: A construct validity study of three measures of demographic similarity. Proceedings of the Academy of Management, 159-163.

Riordan, C., & Shore, L. (1997).  Demographic diversity and employee attitudes: Examination of relational demography with work unit.  Journal of Applied Psychology, 82, 342-358.

Sacket, P., DuBois, C., & Noe, A. (1991).  Tokenism is performance evaluation: The effects of work representation on male-female and black-white differences in performance ratings.  Journal of Applied Psychology, 76, 263-267.

Sessa, V. I., & Jackson, S. E. (1995).  Diversity in decision-making teams: All differences are not created equal.  In  M. Chemers, S. Oskamp, and M. A. Costanzo (Eds.), Diversity in organizations: New perspectives for a changing workplace,  (133-156). Thousand Oaks, CA: Sage.

Shipley, M. F., & Aven, F. F. (1997).  Team member diversity: Contributor to or detractor from member commitment and group decision making?  Proceedings of the Institute of Behavioral and Applied Management, 275-278.

SHRM (1997). The evolution of diversity training.  Mosaics, SHRM focuses on workplace diversity, 3, 1-5.

Simmons, B. L., & Nelson, D. L. (1997).  The diversity advantage?  A paper presented at the annual conference of the Academy of Management. Boston, MA.

Stewart, T. A.  (January 15, 1996). Taking on the last bureaucracy. Fortune, 105-108.

Stienchcombe, A. L. (1968). Constructing social theory. New York: Harcourt and Brace.

Thomas, D. C. (1999).  Cultural diversity and work group effectiveness: An experimental study.  Journal of Cross-Cultural Psychology, 30, 242-263.

Thomas, R. R. Jr. (1990). From affirmative action to affirming diversity.  Harvard Business Review, 68, 107-117.

Thomas, R. W. (1998, August). Diversity programs deemed valuable despite skimpy data.  HR-News, 5.

Tolbert, P. S., & Zucker, L. G. (1996). The institutionalization of institutional theory.  In  S.R. Clegg, C. Hardy, and W. R. Nord (Eds.), The handbook of organizational studies,  (175-187). Thousand Oaks, CA: Sage.

Tusi, A., Egan, T., & O’Reilly III, C. (1992). Being different: Relational demography and organizational attachment.  Administrative Science Quarterly, 37, 549-579.

Ulh-Bien, M., Graen, G. B., & Scandura, T. A. (2000). Implications of leader-member exchange (LMX) for strategic human resource management systems: Relationships as social capital for competitive advantage. In Gerald R. Ferris (Ed.), Research In Personnel and Human Resources Management, 18, 137-185.

Ó the Journal of Behavioral and Applied Management – Winter, 2003 – Vol. 4(3) Page 317

Walsh, M. W. (2000, September 3). Where G.E. falls short: Diversity at the top.  The New York Times, pp. bu 1, 13.

Watson, W. E., Kumar, K., & Michaelsen, L. K. (1993).  Cultural diversity’s impact on interaction process and performance: Comparing homogeneous and diverse task groups.  Academy of Management Journal, 36, 590-602.

Weaver, V. J. (1999). If your organization values professionals, why are they leaving.  Mosaics, SHRM focuses on workplace diversity, 5, 3 (4:5).

Weaver, V. J. (2000). Diversity: One size doesn’t fit all.  Mosaics, SHRM focuses on workplace diversity, 6, 3 (4:6).

Weizmann, H. C., & Weizmann, J. K. (2000). Rewards and business strategy. Chicago, IL: CCH, Inc.

Wharton, A. S., & Bird, S. R. (1966).  Stand by your man: Homosociality, work groups, and men’s perceptions of difference.  In C. Chang (Ed.), Masculinities in Organizations, (97-114). Thousand Oaks, CA: Sage.

Wharton, A., & Baron, J. (1987). So happy together? The impact of gender segregation on men at work.  American Sociological Review, 52, 574-587.

Williams, K. Y., & O’Reilly III (1998).  Demography and diversity in organizations: A review of 40 years of research.  In B. M. Staw and L. L. Cummings (Eds.), Research in organizational behavior, 15, 77-140.