Diversity Trainer Preconceptions: The Effects of Trainer Race and Gender on Perceptions of Diversity Trainer Effectiveness

Abstract

An important issue in designing diversity training programs is deciding who should conduct the training because the demographics of the diversity trainer may affect perceived training effectiveness. A total of 224 subjects participated in 2 experimental studies investigating the effects of diversity trainer race and gender on perceptions of diversity trainer effectiveness. The findings reveal that the race of the trainer, but not gender, influences trainer effectiveness perceptions and that the trainer's personal experience with discrimination and knowledge of institutional discrimination moderate the relationship between trainer race and effectiveness perceptions.

ACKNOWLEDGMENTS

Portions of this research were presented at the 68th annual meeting of the Academy of Management in Anaheim, California, in August 2008 and at the 25th annual meeting of the Society for Industrial/Organizational Psychology in Atlanta, Georgia, in April 2010.

This research was supported in part by the Teachers College, Columbia University President's Grant for Student Research in Diversity awarded to Benjamín Liberman.

We thank Laura Buffardi, Rachel Mendelowitz, Julie High, Stephanie Tornatore, and Jonathan Blodgett for their help with data collection. We also thank Loriann Roberson for her valuable comments on drafts of this article.

Notes

Note. Total sample size ranged from 122 to 124.

*p < .05.

¹Due to the low number of Black and Asian participants, these two groups were combined into one to see whether responses varied by subjects’ minority group membership.

Note. Standard deviations appear in parentheses. All ratings were done on 7-point scales with the exception of likelihood of discrimination, which was on a scale from 0 to 100%. The higher the number, the higher the rating (the more effective, the more credible, etc.). n = 62 in each condition.

^a Means for each dependent measure grouping differ significantly at p < .05, as indicated by univariate analysis of variance tests.

†Means for each dependent measure grouping differ significantly at p < .10, as indicated by univariate analysis of variance tests.

Note. Total sample size ranged from to 95 to 100.

*p < .05.

Note. Standard deviations appear in parentheses. All ratings were done on 7-point scales with the exception of likelihood of discrimination, which was on a scale from 0 to 100%. The higher the number, the higher the rating (the more effective, the more credible, etc.). n = 49 in the White trainer condition and 51 in the Black trainer condition.

^a Means for each dependent measure grouping differ significantly at p ≤ .05, as indicated by univariate analysis of variance tests.

Note. Standard deviations appear in parentheses. All ratings were done on 7-point scales with the exception of likelihood of discrimination, which was on a scale from 0 to 100%. The higher the number, the higher the rating (the more effective, the more credible, etc.). Means for each dependent measure grouping that do not share subscripts differ significantly, as indicated by paired t tests for comparisons between White and Black trainers and by least significant difference tests for comparisons between information conditions. n = 17 in each condition with the exception of the White trainer in the personal experience with discrimination information condition, which has only 15 participants.