The Effect of Diversity on Turnover: A Large Case Study

Abstract

Using longitudinal data collected in 1996–98 from over 800 similar workplaces owned and operated by one corporation, the authors examine how workplace diversity and employee isolation along the dimensions of gender, race, and age affected employee turnover. This design controls for much of the variation in job characteristics and labor markets that have confounded other studies of diversity. The authors use the non-linearity of diversity to distinguish its effect from the main effects of demographic groups and from isolation (being in a numerical minority). The study examines how diversity and isolation by race (white, black, Hispanic, Asian), sex, and age affected different groups. The authors find no consistent evidence that diversity itself increased turnover. In contrast, isolation from coworkers and from customers was often associated with higher turnover.