Are College Students “Naturally Biased” Against Female Instructors?

A new study claiming to demonstrate bias in how college students evaluate female instructors has been making a lot of waves in the media recently. The study, published in the journal Innovative Higher Education, found that students in an online course gave lower ratings to instructors who were presented as female compared to those who were presented as male.  In response, Slate ran an article entitled “Best Way For Professors To Get Good Student Evaluations? Be Male,” in which they called the results of this study “astonishing” and concluded that “men still get bonus points for showing up male.” Likewise, Jezebel ran an article entitled “Students Give Male Instructors Better Evaluations, Says Science,” in which they claimed that this study demonstrates that “college students are naturally biased against female instructors.” But are college students in general really so hostile to the idea of being taught by women? Looking across all of the science out there on this topic, you’ll find that the story is much more complicated than these media reports let on.

Let me first clarify that the goal of this post is not to deny that sexism exists in academia (it certainly does) or to suggest that the playing field is perfectly equal for male and female professors (it isn’t). Rather, what I want you to see is that the picture is more complex than the media is reporting and that the study they’re using to support their claims has some major limitations that prevent it from telling us anything about how college students in general feel about female professors.

Let's start with a closer look at the study. The first thing you should know is that it involved just 43 students who evaluated assistant instructors from one online anthropology course administered over the summer at one university in the southern United States. So, we aren't actually talking about how students rated their professors, but rather how they evaluated their teaching assistants (i.e., the people who facilitated group discussion and graded papers--not the people who gave lectures and directed the course). This is an important distinction because we shouldn't just assume that students evaluate professors and TAs according to the same set of standards. However, we must also keep in mind that we’re only talking about data from 43 students from one academic discipline who attend college in a specific part of the country where all of the –isms (racism, sexism, heterosexism, etc.) tend to be more common to begin with. In other words, this is hardly a generalizable sample (and, for the statistically inclined, the power is obviously very low).

There are some important issues with the study design that we should consider as well. In this study, students were assigned to one of 4 discussion sections. Each section was led by a teaching assistant whom students were led to believe was either male or female based upon their name. Sometimes the assistant’s actual gender matched the gender perceived by the students, while other times it didn’t. When students perceived the assistant as female, the TA was actually male 60% of the time. When students’ perceived the assistant as male, the TA was actually female 52% of the time. So, what we're mostly looking at in this study is how students evaluated male TAs who were masquerading as female and female TAs who were masquerading as male.

The problem here is that, in the "perceived female" condition, the researchers simply combined students' evaluation data for actual women with that of men who were pretending to be women. Likewise, in the "perceived male" condition, they combined students' evaluation data for actual men with that of women pretending to be men. In my view, these data should be analyzed separately because, when combined, it makes the findings very difficult to interpret. The thing that is unclear from them is this: in the perceived female condition, were the actual women rated just as low as the men who were pretending to be female? Or was it the case that students liked the actual female TAs just fine, but disliked the men who pretended to be women? If it's the latter case, the conclusions from this study would be quite different (i.e., they certainly wouldn't suggest that students are "naturally biased" against female teachers in general). To be clear, I'm not saying that the findings wouldn't be cause for concern in this case--rather, I simply think it's important to point out that people are drawing sweeping inferences from a study in which the findings are not all that clear.

There are a number of other issues with this study that should also arouse caution (e.g., there is a selection bias because not all students completed the end-of-semester evaluations, the authors never reported how many male and female students were in each condition, the cutoff point for statistical significance was bumped up to .10 instead of the standard .05 for many tests, etc.). In short, I think it should be clear how problematic it is for anyone to say that this particular sample or set of findings is generalizable and for the media to be making such bold claims based upon this study. On that note, I would encourage you to be wary about where you get your science news in the future because far too many outlets tend toward sensationalism, and Slate and Jezebel are repeat offenders in this area.

So what has other research on this topic found? Is there evidence of a pervasive bias against female instructors at the college level? If you look at early research conducted in the 1970s and 80s, there are more than a few studies that would lend support to that idea. For instance, a 1987 study published in the Journal of Educational Psychology found that both male and female students evaluated female instructors less favorably than male instructors.

However, the world has changed a lot since then, and so have the research findings. Overall, however, the results are kind of a mixed bag. For instance, a 1995 study found that male and female students provided similar ratings for male instructors, but that female instructors received higher evaluations from female students and lower evaluations from male students. In contrast, a 1999 study reported that male students did not differ in how they evaluated male and female instructors, whereas female students provided higher ratings to female instructors and lower ratings to male instructors.

Yet other studies have found no difference in the way that male and female students perceive instructors of different genders. For instance, in a study I co-authored earlier this year in the journal PLoS ONE, we examined how a sample of over 350 college students perceived male and female instructors who teach psychology courses focused on women’s and gender issues. In our work, we found no difference in how male and female students evaluated instructors of either gender.

One other thing that’s worth pointing out is that even in recent studies that have found differences in the way that instructors are evaluated based upon their gender, those differences have been relatively small.

In sum, the available data do not appear to support the claim that college students are “naturally biased” against female instructors. Again, this is not to say that no students hold such bias. Many do, but this is a case where it is essential to consider the context. Not all academic disciplines, courses, class sizes, universities, and student bodies are equal. For instance, some research suggests that sexism of this nature may be more likely to emerge in large lecture classes in the South relative to other contexts. This is the kind of thing we should be paying attention to and reporting on. Identifying where the problems are and what’s causing them is the only way to solve them—making gross over-generalizations and sensationalized claims serves only to distract us from the real issues and it hurts our ability to address them.

Want to learn more about Sex and Psychology ? Click here for previous articles or follow the blog on Facebook (, Twitter (@JustinLehmiller), or Reddit ( to receive updates.

Image Credit:

You Might Also Like: