CAMBRIDGE, Mass., Sept. 7, 2006 -- A new study shows that diversity training programs have roundly failed to eliminate bias and increase the number of minorities in management, despite the fact that many corporations have spent increasing amounts of money on them since the 1990s.
In a paper to be published in the American Sociological Review, Frank Dobbin, professor of sociology in Harvard University's Faculty of Arts and Sciences, Alexandra Kalev of the University of California, Berkeley, and Erin Kelly of the University of Minnesota concluded that such efforts to mitigate managerial bias ultimately fail in their aims. In contrast, programs that establish responsibility for diversity, such as equal opportunity staff positions or diversity task forces, have proven most effective.
"For the past 40 years companies have tried to increase diversity, spending millions of dollars a year on any number of programs without actually stopping to determine whether or not their efforts have been worth it," Dobbin says. "Certainly in the case of diversity training, the answer is no. The only truly effective way to increase the presence of minorities and women in managerial positions is through programs that create organizational responsibility. If no one is specifically charged with the task of increasing diversity, then the buck inevitably gets passed ad infinitum. To increase diversity, executives must treat it like any other business goal."
Although the subject has received much theoretical attention, this study is the first to examine the efficacy of diversity programs based on the actual change in minority representation in management positions. In order to study the efficacy of these programs, Dobbin and his colleagues examined reports submitted to the Equal Employment Opportunity Commission by private sector establishments and surveyed a sample of these establishments on the history of diversity programs within the company.
The researchers categorized these programs into three groups: organizational responsibility programs such as task forces or staff positions; managerial bias programs such as diversity training; and programs that create networking or mentoring opportunities for women and minorities. The researchers then evaluated these programs based on the change in proportional representation of black and white women and men in managerial positions.
The data showed that these programs operate with different degrees of efficacy based on the demographic groups, but organizational responsibility programs proved the most effective. Diversity task forces yielded the greatest results, increasing the proportion of white women in management positions by 14 percent, black women by 30 percent, and black men by 10 percent.
Diversity training aimed at reducing managerial bias may actually increase it: Such programs were followed by a 6 percent decline in the proportion of black women in management. White women benefited modestly with a 6 percent increase.
Programs aimed at reducing social isolation showed limited success. Social networking improved representation of white women, but lowered that of black men. Mentoring programs showed a strong positive effect for black women.
Across the board, diversity programs benefited white women the most, followed by black women, with black men benefiting the least.
"Although the likelihood of minorities holding management positions has increased, the raw percentages of minorities in management remain quite low," Dobbin says.
Dobbin, Kalev, and Kelly's work was funded by the National Science Foundation, the Russell Sage Foundation, and the Robert Wood Johnson Scholars in Health Policy Research Program.
Journal
American Sociological Review