In a recent study, a team of scientists investigated how perceived facial dominance, indicated by the facial width-to-height ratio (fWHR), affected student evaluations of teaching. Their study found that higher fWHR enhanced evaluations for male professors but appeared to detrimentally affect female professors’ scores. The findings, published in Economics & Human Biology provide evidence for a gender-based double standard in academia that punishes women more harshly for perceived dominance.
The fWHR is calculated by dividing the width of the face by its height, and studies have suggested that a higher fWHR is often perceived as a marker of dominance. These findings have been observed in different contexts, including leadership effectiveness, financial success among CEOs, and even athletic performance.
The link between fWHR and perceptions of dominance has prompted researchers to explore how these perceptions influence social interactions and evaluations in professional settings. For example, a higher fWHR in male CEOs has been associated with better company performance, possibly due to perceived leadership qualities. Similarly, in the realm of politics, candidates with higher fWHRs are often judged as more competent or strong, potentially influencing electoral success.
The motivation for conducting the new study stemmed from a need to understand how these perceptions translate into the academic sphere, specifically in how students evaluate their professors. This inquiry was particularly pertinent given the ongoing discussions around gender biases in academic evaluations.
“We think it is important to study how perceptions may have real life consequences for individuals,” explained study author Valentina Paredes, an assistant professor of economics at Universidad de Chile. “Nowadays, most universities use student evaluations of teachers to evaluate their professors, and these evaluations are usually taken into account for promotion or hiring decisions. However, our study shows that facial features such as the fWHR, a proxy for perceived dominance, have a differential effect by gender on these evaluations, independent of professors’ quality.”
For their study, Paredes and her colleagues conducted a comprehensive analysis using a large dataset from the School of Business and Economics at a Chilean university. The dataset spanned from 2012 to 2019, including 5,157 courses taught by 688 professors. The main variable of interest, fWHR, was calculated using professors’ official photos taken for the school’s website.
Student evaluations consisted of an electronic survey with 12 questions that assessed various aspects of teaching effectiveness. The survey also included an overall grade that students gave each professor. The researchers constructed three primary measures of teaching effectiveness from the survey data: the principal component of all questions combined, the overall grade, and the average score of three key questions related to teacher quality.
To ensure that perceived dominance measured through fWHR was distinct from attractiveness or teacher quality, the researchers also controlled for beauty scores and a value-added measure of teaching quality. The beauty scores were obtained through both algorithmic and human ratings. The value-added measure was calculated by analyzing students’ individual grades in each class and determining the professor’s contribution to student performance.
For male professors, a higher fWHR consistently led to higher better evaluations across various measures. Specifically, an increase of one standard deviation in the fWHR was associated with an increase in student evaluations of teaching scores by 0.07 to 0.13 standard deviations, depending on the specific evaluation metric used. This suggests that traits associated with dominance are valued positively in male professors by students, aligning with traditional gender norms that favor assertive and authoritative characteristics in men.
In contrast, an increase in fWHR for female professors resulted in significantly worse evaluations. The reductions were substantial, ranging from 0.27 to 0.87 standard deviations across different measures of teaching evaluations. This finding indicates a strong gender bias, where the same characteristics that benefitted male professors were detrimental to female professors.
“There are some characteristics that are not equally valued in women as in men,” Paredes told PsyPost. “In particular, we show that perceived dominance can have a negative effect on the student evaluation of teaching of female teachers, while the opposite is true for male teachers. Also, these results are relevant beyond the educational sector because any context in which individuals such as managers, co-workers, or employees evaluate another individual is subject to this bias.”
“We expected to find a positive effect of the fWHR on student evaluations of teaching scores for male professors and a negative effect for female professors. However, we were surprised by the magnitude of the effect. For example, our model predicts that if perceived dominance had no effect on student evaluations of teaching scores, the likelihood of a female professor receiving the highest evaluation among all teachers in a given term would increase by 13 percentage points.”
The findings from this study provide compelling evidence of gender biases in student evaluations of teaching, mediated by perceived dominance as signaled by facial features. But as with any study, there are limitations to consider. For instance, the study was conducted in a single institution and country, which may limit generalizability to other regions or educational contexts.
“Throughout our study, we use the fWHR as a proxy for perceived dominance, but the fWHR could also correlate with other teacher characteristics,” Paredes noted. “While we acknowledge this limitation in our study, our results suggest that this facial feature is indeed a good measure for perceived dominance. Moreover, even if our results were driven by other teacher characteristics that correlate with fWHR, we can still conclude that these characteristics,unrelated to teacher quality, have an impact on student evaluations of teaching.”
“Our results are not able to tell apart whether fWHR has an effect on student evaluations of teaching because it affects perceptions of dominance or actual dominant behavior. We acknowledge that students might encounter other cues during their interaction with the faculty in the classroom to judge dominance and also to write the student evaluations. More research is needed to further understand the channels and how interactions in the classroom affects the results.”
“Finally, one limitation of our data is the imbalance between the number of male and female professors. We acknowledge that there is limited representation of women in our sample. However, we think it is important to study settings with many courses situated in male-dominated domains, even though these schools have a small percentage of female teachers.”
Future research could explore more direct measures of dominance and their classroom impacts, as well as how other traits, such as voice or posture, influence student evaluations. Furthermore, additional studies could identify interventions to mitigate bias in teaching evaluations, particularly for female professors, to promote fairer evaluation practices.
“We are interested in studying the underrepresentation of women in academia and different tools to increase their representation,” Paredes said. “Course evaluations is one particular aspect where women experience discrimination, but it is not the only one.”
“This is joint work with Francisco Pino, from the Department of Economics at Universidad de Chile, and David Diaz, from the Department of Management at Universidad de Chile,” she added.
The study, “Does facial structure explain differences in student evaluations of teaching? The role of fWHR as a proxy for perceived dominance,” is published in the August 2024 issue of Economics & Human Biology.
