Managing for Diversity Research:  What We Know from Empirical Research about the Consequences of Heterogeneity in the Workplace

Lois Recascino Wise

Professor of Policy and Administration

 School of Public and Environmental Affairs,

Indiana University

Bloomington, Indiana 47405

wisel@indiana.edu

Paper presented at the Metropolis Conference, November 27-30, 2001, Rotterdam

Workshop 9:

Strategies for Diversity Policy and Practice in Multiethnic Working Life

Acknowledgements:  This paper draws on collaborative research with Mary Tschirhart. The Swedish Council for Worklife Research (RALF) and the Futures Studies Institute provided funds for the original data collection.

Bio Sketch

Lois Recascino Wise is professor in the School of Public & Environmental Affairs at Indiana University.  Her research and teaching interests center on the broad area of management policies and practices, with a special focus on the public sector.  She has served as a consultant to public and private sector organizations in the United States and Europe. In the area of managing for diversity, she has contributed to projects for public and private sector organizations independently and through Tessla, AB.  Her work may be found in numerous US and international journals.  She is author of Labor Market Policies and Employment Patterns in the United States (1989) and joint author of Lön för Mödan (Reward for Effort) published in Stockholm (1997). 

Managing for Diversity Research:  What We Know from Empirical Research about the Consequences of Heterogeneity in the Workplace 

Abstract

Many useful lessons for those pursuing managing for diversity practices have been drawn from case study research.  Although empirical hypothesis testing research has often presented researchers with conflicting and ambiguous findings, some insights can be extracted from this body of research that may be equally valuable to those hoping to gain the advantages of a diverse work and reduce the problems related to workplace heterogeneity.  This study draws eight findings for managers from a meta analysis of 106 empirical research studies of the effects of diversity in the workplace.

Managing for Diversity Research:  What We Know from Empirical Research about the Consequences of Heterogeneity in the Workplace

Although we actually knew very little about the effects of various forms of heterogeneity when the 'diversity' movement began in the early 1990s, our knowledge has steadily increased.  Early claims that diversity leads to innovativeness and creativity, which were among the main reasons put forward for organizations to take up the managing for diversity approach, were largely based on much older studies of personality differences.  Although important contributions to our understanding of the management implications of diversity in the work place have been made by those applying case study research  (Dobbs 1996; Sani 2000; Slack 1997; Pomperleau 1994),l reviews of diversity-related empirical research have generally  pointed to the difficulty in drawing lessons  from the existing body of work.  Lack of replication and methodological problems in the relatively small body of research and, more importantly, inconsistency in findings among studies linking different types of diversity to specific work outcomes   made it difficult to draw generalizable conclusions   (Cox, et al., 1991; Tsui, et al., 1995; Jackson, et al., 1991; Wise and Tschirhart, 2000a).    The body of empirical work does, however, offer some insights into how heterogeneity affects work organizations and recent empirical research reinforces those findings.  While we may not know as much as we want to know, the research to date offers unique insights that are useful for managers and policy makers.  These insights can be seen as complementary to those drawn from case study research.

 Since there are many different definitions of the term . diversity,. the paper begins with a brief explanation of how diversity is defined for the purposes of this research.   It then briefly reviews the research methods used to compile empirical hypothesis-testing studies in the meta-analysis.  The discussion then turns to the generalizations and findings about the consequences of diversity that can be drawn from the research.  The conclusion affirms the need for active and strategic approaches to managing for diversity. 

I. THE DIVERSITY CONSTRUCT

 Although practitioners mainly think of human diversity in terms of race, sex, and ethnicity, in the scholarly literature, heterogeneity and diversity embrace a very broad spectrum of individual differences.  Many primary dimensions of human heterogeneity such as age, race, and ethnicity have only recently started to interest researchers.  There is a relatively long history of investigations focusing on the differences between the sexes, and studies examining the effects of differences among personality types date back to the late 1950s and early 1960s (Hoffman, 1959; Hoffman and Maier 1965; Triandis et al. 1965).  Some research examines the consequences of different educational backgrounds, such as degree type, field of study, or alma mater.   A substantial body of organizational research examines differences in organizational tenure, team tenure, and functional background, or other experiential differences.  These studies represent an important share of the work examining the relationship between diversity and work outcomes such as group problem solving, creativity, and cohesion, as well as organizational performance and innovation.  Since many key assumptions about diversity in the workplace are drawn from this research, it seems appropriate to use a broad definition of the construct in a review of the findings.  Still, it should be acknowledge that some writers  contend that use of the term as a . catchphrase. for all perceptions of likeness and otherness diverts attention from disadvantaged groups or glosses over the fact that the groups captured by a broader definition do not face equal prejudice or conditions in the workplace (Caudron and Hayes, 1997).

For socio-political reasons, the term 'diversity' or 'managing diversity' is increasingly applied to studies involving discrimination or bias without examining the diversity mix (Cox, 1995).  Such studies are part of the substantial body of research investigating the extent to which bias and prejudice operate in work and everyday transactions. In practice these personnel policies generally do not involve active management of existing diversity but focus on introducing or protecting equality and opportunities for inclusion.   Efforts to increase workforce inclusiveness generally focus on introducing diverse members into an organization and perhaps making the workplace friendlier to people with different values and needs. They provide an understanding of the patterns of human bias and the role of discrimination in interpersonal relations, but they do not tell us what happens when we mix together people from different backgrounds, with different physical appearances or abilities, or from different educational or cultural groups.  Thus, such studies are not part of this review.

II. THE RESEARCH FRAMEWORK AND METHOD OF THE META ANALYSIS

The meta-analysis of managing for diversity research on which this study is based searched for general patterns in hypotheses, study methods, and results (Wise, et al., 1997).  It identified different streams of research contributing to the managing for diversity literature and generalizable findings that could be drawn from these works. The literature included was limited to empirical studies focusing on predicting work outcomes using human diversity variables.   If a study did not actually measure diversity as a mélange or mix of similar or dissimilar characteristics, it was not included in the final database for the meta-analysis.  The approach was to search on a list of keywords pertaining to diversity dimensions and work-related outcomes.  The keywords for diversity dimensions are found in Table 1 and  the dimensions and outcomes are given in Table 2.

  [Insert Table 1 here]

Meta-analyses involve different techniques that permit researchers to extract the cumulative results from a set of individual studies addressing similar questions or pieces of a theoretical model (Johnson, et al., 1995; Wolf, 1986).  They identify central tendencies, variability, and prediction moderators in a body of research.   The meta-analysis was first conducted in 1997 using seven different electronic databases of articles published in English language journals and a technique called . snowballing. to work backward from reference lists and bibliographies of relevant in-hand studies to find previous research on the same topic 1 (Wise, et al. 1997).  The seven databases are listed in Figure 1. The electronic search was repeated in 1999 to obtain studies in print through December 1998, which produced a total of 106 usable studies (Wise and Tschirhart, 1999).

[Insert Figure 1 here]

The search procedure was designed to locate and retrieve empirical studies addressing the impact of diversity on the workplace, including individual, group, and organizational outcomes.  Case studies and other qualitative research, training literature, and theoretical pieces that did not test the effects of some form of diversity on a work-related outcome were excluded although they represent a disproportionately large share of the research 'hits'.    Other work draws lessons for managers from this body of research.  

Table 2 gives the number of research . findings. by work outcome and diversity dimension from 1961 to 1998.   A single study would typically produce more than one finding about the consequences of diversity on different work outcomes or would examine more than one diversity dimension in the context of a work outcome. A total of 272 different findings were extracted from the studies in the meta analysis.  These findings include statistically significant and insignificant differences.  By area of work-related outcomes, there was some variation across the different units of analysis: for individual outcomes there were 113 findings, for group or team outcomes 105 findings were identified, and for organizational outcomes 54 findings were identified. As the data in Table 2 indicate, the greatest volume of findings pertains to sex or gender diversity (82).  Ethnicity and race diversity represent the second most popular area of study with 58 findings reported.  The dimension with the third highest frequency of findings is age (38).  The frequency of findings for educational diversity and work outcomes was relatively low (22), although there is a substantial body of work on the effects of diversity on educational outcomes  in non-work settings.

[Insert Table 2 here]

III. WHAT DO WE KNOW ABOUT THE CONSEQUENCES OF DIVERSITY IN WORK ORGANIZATIONS?

This section presents and discusses eight generalizations about heterogeneity in work organizations that can be drawn from the empirical scholarly literature.  The effects of diversity on creativity and innovation are not on this list.  Although claims for the positive effects of diversity on the quality of solutions for solving problems, creativity, and receptivity to, or actual implementation of, change are often cited positive as benefits of diversity, the record is still ambiguous.   There was scant evidence of positive effects of heterogeneity on creativity, problem solving, and innovation. An early study by Hoffman and Maier (1961) demonstrated a positive effect of diversity in sex composition on group innovativeness. Rogelberg and Rumery (1996) also found some evidence that slight variations in sex composition made groups more innovative.  Baugh and Graen (1997) report negative effects related to modest levels of diversity in race and gender on perceived team effectiveness.    A study by Cady and Valentine (1999) adds to the confusion about nature of the relationship between  biological diversity and creativity and problem solving.   It reports that diversity in race, age, sex, and function had no impact on the quality of group innovations.  Diversity in sex composition had a negative impact on the quantity of innovative ideas generated, but racial diversity was associated with more innovative ideas

1. The effects of diversity mature over time.

     Empirical studies attempting to assess the consequences of diversity over time indicate that initial observations differ significantly from those taken later on in time (Watson, et al., 1993).  The effects of diversity evolve and mature over time (Amir, 1969).  On the one hand, this may lead to more positive work relationships among diverse group members as the processes of mutual accommodation and integration, which are critical to effective diversity management, take time to develop. As diverse group members increase their contact with each other, gradually stereotypes about people are replaced by a deeper understanding of the individual and his or her actual similarities and dissimilarities to others replace earlier prejudices (Allport, 1954).  On the other hand, as work group histories develop, the costs of diversity may become increasingly apparent to different stakeholders and participants may be less willing to compromise and accommodate differences.    Homogenous groups may seem to work together more efficiently and with less conflict in early observations than mixed groups, but subsequent observations often show that there are no differences between homogeneous and heterogeneous groups at later stages of group development. 

2. The nature of the task affects the potential gain from diversity.   

          The nature of the task to be performed appears to function as an intervening variable in the diversity-work outcome relationship (Watson, et al., 1993).  When the task- related challenge is categorized as creative, problem solving and execution or coordination, it appears that differences in performance are related to group heterogeneity.  For example, Hambrick et al. (1998) report that in groups made up of people from different countries, multinational diversity offers the greatest challenge to groups engaged in tasks that require coordination.  Although such groups may have a richer diversity in task-related knowledge, diversity overall functions as a disadvantage in efforts to develop effective interaction.  At the same time, multinational diversity had no effect on groups performing more bounded computational tasks where the number of correct solutions is limited. Bowers. (2000) recent meta analysis of the effects of diversity in sex, ability and personality on group performance in 13 different studies confirms that the type and difficulty of tasks accounts for whether or not studies find significant effects.

3. Flawed work processes reduced the potential gain from diversity. 

          Ineffective and inefficient work processes reduce the potential contribution of a work group.  Gains that might be expected from a heterogeneous group of workers may be lost in the processes involved in getting the job done.  Communication is a primary example of a work process that undermines the potential gain from diversity (Zenger and Lawrence, 1989).  To the extent that members of certain diverse groups have unique communication patterns, communication functions as an intervening or contextual factor affecting the performance of diverse members.  This effect appears to operate between the sexes (Aries, 1976; Verdi and Whelan, 1992; Webber, 1974; Schmitt and Hill, 1977), and within racially diverse groups (Webber, 1974; Schmitt and Hill, 1977.  Similarly, tasks requiring substantial coordination are affected by diversity in the work group (Hambrick et al. 1998).

4. The diversity that matters is contextual.

          Unique and specific situational factors determine the extent to which individual characteristics are taken into account   in work organizations. Hofstede's pioneering work demonstrated systematic differences in managers. attitudes and behavior were associated with their cultural background and related to their management style.  Similarly, the cultural setting of the work activity influences diversity effects (Hofstede, 1980).   Findings obtained for top managers in an individualistic society like the U.S. were not repeated among executives in collectivist societies (Wiersema and Bird, 1993).  For example, heterogeneity in age, team tenure, and prestige of alma mater was found to have a substantially stronger effect on top management team tenure among Japanese subjects than among similar Americans.  The cultural value of collectivism is one possible explanatory variable.  Members of collective societies may focus less on interpersonal differences within the group and more on member/non member differences than members of individualistic societies, like the United States.  Where the majority falls on the collectivist/individualistic dimension may influence the gain from the introduction of ethnic minorities with collectivist or individualistic orientations.   Similarly, researchers find that even within one-country, variations in the conditions of competition in different organizational contexts affect the extent to which diverse or non-diverse teams produce better quality decisions (Pate, et al., 1998).

5. The diversity that matters is perceptual. 

          Diversity may exist in a group or work organization but have no significant effects on organizational outcomes.  Some studies indicate that members of diverse groups perceive group performance as lower than do members of homogeneous groups (Lucas and Lovaglia, 1998) but many empirical studies find no significant effects related to diversity on organizational outcomes (Bantel and Jackson, 1989).   While sex and race are frequently found to be associated with performance indicators at the individual, group, or organizational level, many studies report that these diversity dimensions had no effect on performance (Pulakos, et al., 1989; Tsui and O. Reilly, 1989; Waldman and Avolio, 1991; Sackett, et al., 1991).  This finding may be due to the fact that people are not sensitive to or aware of the diversity that exists.  For example, members of collective societies may focus less on interpersonal differences within the group and more on member/non member differences than members of individualistic societies, like the United States (Triandis, 1988; Bochner, S., and Hesketh, B., 1994).  If diversity is not physically obvious, group members may be unaware of the diversity unless it is brought forward by some event or disclosure.  This may pertain not only to informational diversity and experiential diversity but also to biological diversity, such as ethnicity (Harrison et al., 1998).  A person. s ethnic or racial origins may not be physically obvious to others. 

         By the same token, even if a particular type of diversity is evident, it may not be salient to group members.  Religious diversity, for example, may be highly important among members of some work groups and irrelevant in other contexts where religious affiliation is not an important personal characteristic.  Despite religious differences members may not perceive others as different from themselves.  Similarly, the perceived significance of ethnic diversity, for example, may vary in different contexts and over time.   Whether actual diversity matters or not, may be a function of the extent to which members are open to different types of diversity (Härtel, et al., 1999).   

6. Findings from one diverse group are not necessarily generlizable to another.

          Not all dimensions of diversity appear to have the same effects.  An outcome can be affected by one type of diversity but not another.   Much of the research regarding diversity pertains to heterogeneity with regard to sex or personality type.  Findings from this body of research are often not replicated when applied to culturally diverse groups, indicating the results are not transferable from one diversity dimension to another (Watson, et al., 1993).   Sackett, DuBois, and Noe (1991) also conclude that findings regarding performance ratings obtained for differences related to groups based on sex are not generalizable to groups based on racial differences.  Similarly, while sex is frequently associated with adverse effects on communication, age and educational diversity do not reveal the same pattern of effect on communication. 

7. Specific types of diversity have different effects on different organizational outcomes. 

          A review of diversity research in the demography tradition led Tsui et al. (1995) to the conclusion that diversity impacts organizational outcomes differently.  For example, diversity in age and organizational tenure generally is found to be positively related to turnover and negatively related to group cohesion and communication.  Similarly, Hambrick et al. (1998) report that  multinational diversity reduces performance in coordinated tasks, groups with the same diversity demonstrate greater creativity by drawing on their wide range of values and cognitions. 

          In reviewing the body of research on diversity and work performance, Wise and Tschirhart (2000b) analyzed the distribution of findings from performance-based diversity studies by diversity dimension and performance outcome. They report differences among specific diversity dimensions and type of performance outcomes.  The diversity dimensions of education, age, and to a lesser extent experience were most often positively associated with performance outcomes.   In comparison, diversity in organizational tenure and team tenure had little or no positive association with performance outcomes.  Moreover, diversity dimensions were more likely to be positively associated with qualitative indicators of performance than with efficiency-based indicators.

8. Diversity increases organizational turnover. 

A large share of the empirical research focuses on the consequences of diversity for organizational or team turnover (Jackson, et al., 1991).   These studies generally find that as diversity increases so does the amount of turnover. For example, increased diversity in tenure in work organizations is associated with higher turnover rates in academic units and among hospital nurses (McCain, et al., 1983).  Compositional differences in age have also been associated with higher organizational turnover (Wagner, et al., 1984).  Individual. s propensity to quit is also related to increased diversity in age, education and race/ethnicity, experience and team tenure (Wagner, et al., 1984; Caldwell and Barnett, 1989; Jackson, et al., 1991).  Increasing diversity on a set of demographic characteristics is associated with decreasing organizational attachment and lower social integration among both majority and minority group members, according to Tsui, Egan, and O. Reilly (1992).

IV. CONCLUSIONS 

The philosophy of managing for diversity is gaining popularity and both public and private sector organizations are investing significant financial resources, time, and energy into diversity missions, programs, and strategies.    Managers who want to integrate a managing diversity approach into their leadership behaviors and into the policies, structure, and culture of their organization look for lessons from the research literature that can be reliably applied to actual work situations.  In considering the conclusions drawn from the body of empirical research, it is important to note that these studies do not necessarily take place in organizations that are actually pursuing a managing for diversity policy.   If research was focused on the effects of heterogeneity in workplaces characterized by these policies, we might see very different effects from diversity.

          The eight points identified here about the consequences of diversity offer some guidelines to managers. For example, that the effects of diversity mature over time, suggests that managers need to treat the diversity that exists differently at different stages of group development. Adler (1991), for example, identifies three stages of group development and argues that existing diversity must be managed actively to limit its dysfunctional effects in stages one and three and to promote its positive effects on creativity and problem solving in stage two.

          One implication of the finding that flawed work process reduce the potential gain from diversity is that it can be assumed that heterogeneous groups will require more frequent communication than homogeneous groups.  Managers need to provide the additional time, opportunity, and appropriate instruments for effective communication in diverse groups.   The number of items on the agenda should be limited, written copies of information discussed should be handed out, and multicultural groups should meet more often to reduce communication  problems.

          If the diversity that matters is contextual, managers need to evaluate the  situation in which work occurs to anticipate and lessen the effects of barriers to effective performance among heterogeneous workers. Other potentially critical contextual factors are organizational size, structure, technology, and organizational communication mechanisms.   Organizational complexity is a factor that may affect the extent to which diversity initiatives can be successfully installed in different organizations.  More complex organizations, for example, show evidence of more frequent interpersonal communication exchanges that would increase the amount of contact between members of diverse groups (Hoffman, 1985).  Similarly, organizational complexity affects the level and types of communication within an organization and in turn the need for cooperation and coordination. Management can anticipate these needs in scheduling work, meetings, and providing necessary equipment and opportunities to improve communication. Organizations can increase effectiveness by promoting an understanding that diverse and non-diverse groups will have different reactions to conditions of competition (Pate, et al., 1998).

          If the diversity that matters is perceptual, the degree of openness or closed-mindedness may be an intervening variable explaining how much room for tolerance of diversity exists within a given workplace (Härtel, et al., 1999).  This would mean that training in valuing diversity would be an important element for establishing a workplace fertile to heterogeneity and preparing the ground for a managing for diversity approach.

If findings from one diverse group are not necessarily generalizable to another, managers must be careful not to treat all diverse groups and minority groups alike.  Invoking the platinum rule, . treat others as they want to be treated,. implies that managers have an obligation to know differences in values and norms among members of different social groups.  This includes understanding and appreciating differences in values and work orientations within broad groups such as Native Americans, Asians, South Americans, etc.  The diversity within groups is often greater than the diversity between groups.

          Some research indicates that culturally diverse groups that improve their group process performance will improve their task performance and exceed the performance of homogeneous groups. A proactive management approach to teams would   ensure that team members get sufficient feedback, set ground rules and focus on behaviors affecting process performance (Watson, et al., 1998).

          If diversity impacts organizational outcomes differently, managers need to have realistic expectations of where they can expect positive benefits from heterogeneity and perhaps more importantly where diversity will reduce effectiveness if not properly managed.  Managers can more carefully attend to those work activities where diversity is likely to produce negative effects so that faulty processes and contextual factors that impede effective work performance can be minimized.  Qualitative gains in greater creativity and innovation may come at a high price if efficiency is a critical variable in organizational performance.  Supervisors need to manage diversity strategically so that it does not substantially reduce efficiency gains.

          Finally, if heterogeneity is associated with higher turnover, it is important to develop activities and incentive systems to reinforce organizational attachment and commitment while at the same time providing information and training that leads to greater understanding and tolerance of human diversity. Reduced effectiveness in diverse teams may be real or merely perceptual, but since for each observer the perception is a reality that affects work satisfaction and organizational attachment, those perceptions are of concern to managers.  

Simply increasing the amount of diversity in an organization is not a good strategy.  Many contextual factors can impede the effectiveness of heterogeneous groups.  Moreover, minority group members will be perceived as outsiders and they and their team members may have lower evaluations of team effectiveness.  The diversity that exists in the work place has to be managed carefully and management should assure that objective indicators of group performance are provided.  

The competitive advantage from a diverse work force comes not from achieving a certain statistical distribution of minorities and women but from the different ways of thinking, life experiences, product and service preferences that these people can bring into an organization.  Organizational structures that do not provide an avenue for these insights and experiences to become part of the problem solving and decision-making framework will see little gain from efforts they make to recruit different groups of people into the work force.  In fact, they may create frustrations that lead to new problems and conflicts in the workplace. 

Organizations need both a vision and a plan for how they will implement diversity.   An organization cannot be neutral on the issue of diversity.  The risk in not fully embracing the philosophy at the top of the organization is that management sends mixed signals to employees and customers that negatively impact success.   If diversity policies are based on the idea that creating a high-performance organization is not possible without leveraging the world. s diverse cultures and rich human diversity, then diversity programs must both involve and empower employees from all levels of an organization.

This conclusion is consistent with management literature indicating that managing diversity means assuring the full range of human and cultural variation within an organization is encouraged to come forward in solving problems and mapping strategies.  It entails a degree of individual empowerment and that is why it is so important for management to make sure that all employees are given the opportunity to voice their ideas and feel their contributions are welcome.  Thus, the main conclusion to draw from this review of the literature is that managers must take responsibility to ensure that diversity is managed actively and strategically. 

Table 1

Keywords for Diversity Dimensions for Electronic Database Search

·           Age, age composition, older worker

·           Cohort, cohort composition

·           Disability, handicap, disabled

·           Education

·           Ethnicity, ethnic composition

·           Race, racial, race composition

·           Sex, gender, gender composition

·           Tenure, tenure composition

·           Team composition

·           Group composition

·           Work background

·           Work experience

Number of findings by work outcome and diversity dimension

Figure 1

ELECTRONIC DATABASES USED

·            Indiana University. s Online Catalog (IUCAT)

·            Expanded Academic Index (EAI)

·            Business Index (BI)

·            Educational Resources Information Center (ERIC)

·            PsychINFO

·            Sociofile

·            Social Planning Policy and Development Abstracts. (SOPODA)

REFERENCES

Allport, G., (1954). The Nature of Prejudice, Cambridge, MA: Addison-Wesly;

Amir, Y. (1969).  Contact hypothesis in ethnic relations, Psychological Bulletin 11: 319-342.

 Berger, J., Rosneholtz, S., and Zelditch. M., (1989).  Status organizing processes, in A. Inkeles, N.J. Smelser, and K. H.  Turner (eds.) Annual Review of Sociology,  Palo Alto, CA:, 6:479-518

Bochner, S., and Hesketh, B., (1994).  Power distance, individualism/collectivism, and job-related attitudes in a culturally diverse work group.  Journal of Cross-Cultural Psychology 25(2), 233-257.

Bowers, C., (2000).  When member homogeneity is needed in work teams: a meta-analysis, Small Group Research, 31 ( 3), 305-27.

Cady, S., and Valentine, J. (1999). Team innovation and perceptions of consideration: what difference does diversity make? Small Group Research , 30, 6: 730-751

Caudron, S., and Hayes, C. (1997).  Are diversity programs benefiting African Americans?  Black Enterprise 27(7), 12-18.

Cox, T., (1995). . A diversity framework,. in M. M. Chemers, S. Oskamp, and M.A. Costanzo (eds.), Diversity in Organizations. Thousand Oaks, CA and London: Sage, p. 246-260.

Cox, T., Lobel S., McLeod P., (1991).  Effects of ethnic and group cultural differences on cooperative and competitive behavior on a group task.  Academy of Management Journal 34(4), 828

Dobbs, M. (1996).  Managing diversity lessons from the private sector, Public Personnel Management, 25 (3), 351-367.

Hambrick, D. C, S. Davison, S. Snell, and C. C. Snow (1998) When groups consist of multinationalities. Organizational Studies 19(2), 181-206

Harrison, D., Price, K., and Bell, M., (1998). Beyond relational demography: Time and effects of surface and deep level diversity on work group cohesion, Academy of Management Journal 41: 96-107

Härtel, C., Douthitt, S., Härtel, G., and Douthitt, S., (1999). Equally qualified but unequally perceived: General cultural openness as a predicator of discrimination in performance ratings, Human Resource Development Quarterly 10:1: 79-89

Härtel, C., and Fukimoto, Y., (1998). Effects of diversity in organizational groups: development of the perceived openness moderator model.

Hoffman, L., (1959), Homogeneity of member personality and its effect on group problem-solving, Journal of Abnormal Social Psychology 58:27-32

Hoffman, L., and Maier, N. (1961).  Quality and acceptance of problem solutions by members of homogeneous and heterogeneous groups, Journal of Abnormal and Social Psychology, 62: 401-07.

Hofstede, G., (1980).  Motivation, leadership, and organization: do american theories apply abroad?  Organizational Dynamics 9(1), 42-63

Jackson, S., (1992). Consequences of group composition for the interpersonal dynamics of strategic issues processing, in Advances in Strategic Management , 8, P. Shrivastava, A. Huff, and J. Dutton (eds.), pp. 345-82, Greenwich, CT:  JAI Press. 

Jackson, S., Brett, J., Sessa, V., Cooper, D., Julin, J., and Peyronnin, K., (1991). Some differences make a difference: individual dissimilarity and group heterogeneity as correlates of recruitment promotions, and turnover, Journal of Applied Psychology, 76: 675-89

Johnson, B., Mullen, B., and Salas, E., (1995), Comparison of three major meta-analytic approaches, Journal of Applied Psychology 80:1: 94-106

Lucas, J., and Lovaglia, M., (1998). Leadership status, gender, group size, and emotion in face to face groups, Sociological Perspectives 41(3), 617-37.

 Pate, S., Watson, W., and Johnson, L., (1998). The effects of competition on the decision quality of diverse and nondiverse groups, Journal of Applied Social Psychology 28(10), 912-23.

Pomperleau, R. (1994).  A desideratum for managing the diverse workplace, Review of Public Personnel Administration, 14(1), 85-100.

Sackett, P., DuBois, C., and Noe, A., (1991).  Tokenism and performance evaluation, Journal of Applied Psychology 76(2), 263-267.

Sani, V. (2000).  A 21^st century reception for diversity in the public sector: A case study, Public Administration Review, 60(5), 395-408.

Slack, J.D. (1997).  From affirmative action to full-spectrum diversity in the American workplace, Review of Public Personnel Administration, 17(4), 75-87.

Thomas, R., (1990), . From affirmative action to affirming diversity,. Harvard Business Review 68(2), 107-117

Triandis, H., (1988).  Collectivism vs. individualism.  In G.K. Vernma and C. Bagley and H. Triandis, eds., Personality, Attitudes and Values.  London: Macmillan, pp. 60-95.

Tsui, A., Egan, T., and Tsin, K., (1995).  Diversity in research:  lessons from demography research. In Martin Chemers, S. Oskamp, and M. A. Costanza (eds), Diversity in Organizations.  Thousand Oaks, CA: Sage Publications, pp. 191-219

Wanous, J., and Youtz, M., (1986), Solution diversity and the quality of group decisions, Academy of Management Review, 29(1), 149-159

Watson, W., Johnson, L.,  and Merritt, D. (1998). Team orientation, self orientation, and diversity in task group, Group and Organization Management 23(2), 161-88.

Watson, W., Kumar, K.  and Michaelsen, L. (1993).  Cultural diversity. s impact on interaction process and performance: comparing homogeneous and diverse task groups.  Academy of Management Journal 36(3), 590-602.

Wise, L., and Tschirhart, M. (1999).  Can managing for diversity in the public sector be informed by the empirical literature on diversity?  An assessment and agenda.  Paper presented at the 5^th National Public Management Research Conference, College Station, TX: Texas A& M University, Dec. 5.

Wise, L., and Tschirhart, M., (2000a). Examining empirical evidence on diversity effects: how useful is diversity research for public sector mangers? Public Administration Review, 60(5), 386-394.

Wise, L. and Tschirhart, M. (2000b).  What diversity research tells us about heterogeneity and work performance. Paper presented at Workplace Diversity:  A Research Perspective on Policy and Practice, Brussels, 13-15 June.

Wise, L., Tschirhart, M., Hosinski, D., and Bandy, L., (1997).  Diversity Research: A Meta-Analysis.  The Institute for Futures Studies, Stockholm, Sweden

Wolf, H. (1986). Meta-Analysis: Quantitative Methods for Research Synthesis, Beverly Hills, CA: Sage

Zenger, T., and Lawrence, B., (1989). Organizational demography, Academy of Management Review, 32 (2), 353-76.