Evidence for gender bias in interpreting online professor ratings Deanna Robertson and Sara Sohr-Preston Department of Psychology, Southeastern Louisiana University INTRODUCTION METHOD RESULTS Table 1 Summary of F values from ANCOVA College students use tools such as the website Rate My Professors (RMP) to evaluate their professors and seek information about potential professors for future classes. Ratings on these websites indicate whether students are satisfied with the education they receive from their professors. Relatively little is known, however, about potential biases present in such professor ratings. One area of concern is gender bias in professor ratings. Emerging research has suggested that college students tend to rate male professors more favorably than female professors (e.g., Abel & Meltzer, 2007). Given that these publically accessible ratings may influence students’ decisions about enrolling in classes and even administrator decisions regarding hiring and tenure/promotion, potential biases should be examined. Participants included 300 students (62 males, 238 females; 73.20% Caucasian; mean age = 19.62, SD = 1.72; mean GPA = 3.15, SD = .62) who completed a survey on demographic information, ratings of a hypothetical professor, and ratings of their current psychology professor (not included in this investigation). The professor ratings were created for the current study and were meant to encompass topics mentioned on sites like RMP. Furthermore, the hypothetical professors were listed with average scores of the type presented on RMP. Both male and female professors were listed with the exact same scores: 3.2 out of 5 for Helpfulness 4.2 out of 5 for Clarity 4.0 out of 5 for Easiness 3.8 out of 5 for Overall Quality For the current study, it was hypothesized that there would be differences between the ratings for male and female hypothetical professors. Of the participants, 296 provided complete data for items included in the analyses summarized. Comparisons were conducted using analysis of covariance (ANCOVA), with professor sex entered as the independent variable, professor rating as the dependent variable, and participant sex as a covariate. Results of analyses of covariance (ANCOVA), which controlled for the participant’s own sex, revealed statistically significant (p < .05) mean differences for speaking loudly enough to be heard, knowing students by name or face showing videos, and giving extra credit. In all cases, females were rated higher. See Table 1 for summary of ANCOVA results. Explains topics clearly .12 Speaks loudly enough to be heard 4.19 Dresses professionally .14 Arrives on time .72 Has clear grading policies 3.30 Test cover lecture material .97 Tests cover readings 1.12 Is helpful 1.10 Is polite to students .01 Smiles 1.93 Knows students by name or face 5.55 Is consistent .81 Exhibits respect toward students 2.49 Is physically attractive 2.32 Is physically fit .57 Uses current examples .44 Shows videos 4.74 Performs in-class demonstrations .76 Gives extra credit 5.21 Has attractive hairstyle 2.76 Tells funny jokes .24 Shares relevant examples .09 Wears flattering clothing 1.80 Appears to be in good mood .90 Gives handouts 3.28 Breaks down complicated topics Displays confidence .21 Seems to enjoy teaching Hands back quickly 1.66 Is available outside of class .74 Discussion of relevant topics 2.08 CONCLUSIONS The findings of this investigation support that there are some potential differences in ratings when comparing male and female professors on the same scales. Since the ratings were of professors not yet met, and descriptions were similar except for sex, they can be interpreted as reflecting expectations of potential professors. It may be beneficial to examine how students’ ratings affect class registration and how well the student does in the class. The current study provides basis that there may be a gender bias in how students evaluate and anticipate characteristics of female professors compared to male professors. Given the preliminary nature of these findings, further replication and expansion of this study are encouraged. OBJECTIVES The current investigation compared students’ ratings of either a male and female hypothetical professor addressing a variety of specific areas of evaluation (summarized in Table 1), while controlling for participant sex. Note. Bolded F values are statistically significant (p < .05) CONTACT Deanna Robertson – deanna.robertson@selu.edu Dr. Sara Sohr-Preston – sara.sohr-preston@selu.edu