Dollars and difference: The 'diversity premium' in organizational demography.

摘要

Is demographic diversity good for organizations? While organizations are becoming increasingly diverse in terms of gender and race, research findings have been unclear or negative about the impact of these changes on organizational performance. This multi-method study attempts to bring more clarity to the debate by building a better model of the demography-performance relationship, specifying the conditions under which demographic diversity produces positive or negative outcomes.;Using both ethnographic and survey techniques, this project examines a large sample of U.S. investment clubs---voluntary associations composed of 10 to 15 people who pool their money to invest in the stock market Clubs composed of men and women together earn significantly higher returns on their investments than clubs composed of men only or women only. This is the "diversity premium" which the study seeks to explain.;A constellation of variables must line up in order for diversity to have positive effects; when those conditions are not in place, diversity's negative effects predominate. First, the task has to benefit from diversity; such tasks are non-routine and occur in a changing environment. Second, the group has to have differences of opinion and a mechanism for discussing those differences. While all investment clubs operate in the volatile environment of the stock market, mixed clubs outperform their same-sex counterparts because they are more willing to express differences of opinion. Same-sex groups are primarily founded by friends, and are less willing to pay the social costs of dissent. Mixed clubs, which are more often based on instrumental networks such as prior work relationships, are much more task-oriented.;This study offers three contributions to the organizational demography literature. First, it brings in prior network ties between group members; these ties have pervasive effects on group composition, process and performance. Second, it separates demographic diversity from diversity of opinion---a distinction which has been obscured in recent research. Finally, while most of the current literature argues that demographically diverse groups perform poorly due to insufficient social cohesion, this study demonstrates that excessive social cohesion---in the form of suppression of dissent---is just as damaging to group performance.