An Examination of Potential Racial and Gender Bias in the Principal Version of the Interactive Computer Interview System

Abstract

Abstract The primary object of this study was to determine whether racial and/or gender bias were evidenced in the use of the ICIS-Principal. Specifically, will the use of the ICIS-Principal result in biased scores at a statistically significant level when rating current practicing administrators of varying gender and race. The study involved simulated interviews of fifty-two participants who currently are practicing principals in a large, urban school district located in the southeastern region of the United States. Participants in the study were evenly split according to race, with twenty men and thirty-two women participating. Independent t-tests were conducted to investigate the differences between race and gender and ANOVA analysis was conducted comparing results according to organizational level (elementary, middle, and high school). Regression analysis was also conducted on the subscales of vision, instruction, management, collaboration, and integrity, examining the influence of race and gender on the subscale score. The study revealed evidence of the possibility of some racial bias in the instrument. The research indicated that there were significant differences according to race in the content area of ensuring effective management of the organization. In this case a significance of .02 was found. Research indicated that there were slight differences found in terms of gender and organizational level. Regression analyses indicated that the variables of race and gender overall had a slight influence over the content areas measured by the ICIS-Principal. Race did, however, appear to be more important in the results than gender.