Racial attitudes and the selection interview : a factorial experiment

This study investigated the impact of the racial attitudes of interviewers on the ratings given black and white job applicants. The study was particularly concerned with the situation in which high-prejudiced interviewers operate under organizational policies prohibiting racial discrimination. The theoretical framework used to specify the research problem was organizational role-taking. A review of the empirical literature indicated the likelihood of an interaction between applicant race and applicant quality. In order to control for applicant quality and applicant race simultaneously, videotapes of simulated job interviews were produced. Applicant quality was controlled by producing a high quality applicant script and a low quality applicant script. A black male and a white male, role-playing applicants, produced two interviews each. Each applicant role-played the high quality applicant and the low quality applicant. This procedure standardized applicant performance across races, producing black and white applicants with equal qualifications and equal interview performance. The impact of four independent variables on the ratings of applicants was investigated using a posttest-only control group design. The four independent variables were applicant quality, applicant race, interviewers' racial attitudes, and organizational policies nondiscrimination. Each of these independent variables was fully crossed with the other three independent variables in a 2 x 2 x 2 x 2 ANOVA design. Subjects were 176 white business administration students at the University of North Carolina--Greensboro. Six hypotheses were formulated to predict the impact of the four independent variables on the evaluative ratings given applicants. Main effects were predicted for applicant quality, applicant race, and interviewers' racial attitudes. Two-way interactions were predicted for applicant race and applicant quality as well as applicant race and interviewers 1 racial attitudes. A three-way interaction was predicted for applicant race, interviewers' racial attitudes, and organizational policies concerning nondiscrimination. A seventh hypothesis predicted that interviewers would have more confidence in ratings given low quality applicants than high quality applicants. The main effect for quality was significant, accounting for 50 percent of the variance in ratings given applicants on the Evaluative factor. The main effect for race was significant but not in the predicted direction. Black applicants were rated significantly higher than white applicants. The interaction of race and interviewers' racial attitudes was significant but not in the predicted direction. High-prejudice interviewers rated black applicants higher than white applicants. The interaction of race and quality of the applicant was partially supported. Low quality black applicants were rated higher than low quality white applicants, while there was not a significant difference in the ratings of high quality black and white applicants. The hypotheses predicting a main effect for racial attitudes and a three-way interaction for applicant race, racial attitudes, and organizational policies concerning nondiscrimination were not supported. The hypothesis predicting that subjects would have more confidence in their ratings of low quality applicants was not supported. Four conclusions can be drawn from this study: 1. Applicant quality is the most important factor contributing to applicant ratings. Applicant quality accounted for approximately 50 percent of the variance in applicant ratings. 2. Some reverse discrimination seems to be present in the ratings. However, reverse discrimination only accounted for approximately one percent of the variance in ratings. 3. Reverse discrimination seems to take the form of leniency in the ratings of low quality black applicants. 4. High-prejudice subjects were more lenient in rating black applicants than were low-prejudiced subjects.