Racial Bias In Supervisor Ratings

The purpose of this research was to determine if there is an existence of racial bias in supervisory ratings. The researchers, studying archival data, dispute the findings of previous research and state they believe there is, in fact, racial bias in supervisory ratings. The researchers contend that previous researchers have incorrectly interpreted the data for three reasons. First, previous researchers have dismissed the final statistics when taking into account the allotted margin of error. Second, others have concluded the results as inconclusive. Finally, some research exclusively challenged that there was no bias in supervisor ratings in regard to race.

The method of this research was solely by archival data using the results of the Sackett and Dubois Study (1991). Sackett and DuBois used evaluations from the United States Army involving 561 black and 1259 white soldiers, also 331 black and 286 white civilian employees. Each employee was rated by black and white supervisors. Sackett and Dubois concluded that all supervisors, black and white, consistently rated white employees higher than black. However, black supervisors rated black employees substantially higher than their white counterparts. The current researchers conclude that this is the ground work for further research involving this topic with different data sets.

A more remarkable part of this research is the recovery of the National Research Council report on test bias in the United States Employment U.S. Employment Service's General Aptitude Test Battery (Hartigan and Wigdor, 1989). The NRC warned that current evidence suggests that supervisors will tend to rate members of their own race higher than others, all the while acknowledging most evaluations are from white supervisors. Sackett and Dubois took this into account and included a black and white supervisor evaluation for each employee.

The largest flaw of the Sackett and Dubois study is the sample of people