My teaching evaluations arrived and female students statistically gave me worse scores than male students. Is there any possible explanation?

Question

I'm a new assistant professor and I finished my first semester of teaching. It's my first class, but I think it went good: the students seemed happy and interested, I had many of them (both male and female) doing more than required, and some of them later approached me to tell me they learned a lot.

The scores arrived: numerically they ended up being lower than the average of my school, which was disappointing, especially because in the descriptive part about the strengths there were many positive comments about me and the class, and in the negative part, they were mostly neutral or just slightly negative (and I agree with these, they are about my inexperience and baby steps). There was nothing nasty, ad hominem, or abusive. So based on the descriptive part and the overall experience, I expected the numerical score to be much higher, but it's fine and that's not the main topic of this question. I read a lot about student evaluations here on SE Academia and how pointless they can be, so while I care about feedback as a new lecturer, I had low expectations about them and I was going to take them with a grain of salt.

But the scores are also reported by gender. I am a male lecturer, and female students rated me much lower than male students (who rated me highly). The sample size and the difference are so high that I cannot attribute it to a statistical error and dismiss it. This left me astonished as it's the last thing I'd expect, and I'm trying to understand why and where is this coming from. I'm very well aware of the gender equality issues in academia, so I am quite conscious about treating everyone equally well, I'm very careful about avoiding everything that might even remotely imply sexism (for example I don't make jokes, and I never even mentioned anything gender related), and I'm careful about never making any physical intrusion (not even a handshake or sitting too close to a student during discussions). I just cannot think of any reason why students of one gender would give a much different score than the other, and what made female students less happy with the course and me. I know that everyone is biased when talking about themselves, but recently I even had a former female colleague telling me out of the blue I'm the one of the most respectful and pleasant people she worked with.

While you didn't take my class and don't have insights, but is there something in general I might be unaware of, and can improve? Is it possible that there is a teaching style to which different genders are receptive in a different way? Is it just possible that there is simply a discrepancy between how different genders give different scores? It may sound silly, but is it me being too respectful backfiring and seen as condescending by female students? This will obviously be difficult to answer without an insight, so I am at least trying to understand general reasons.

In the comments part of the teaching evaluations there is nothing negative about me or the course that I can interpret being gender related (they are mostly about the organization of the teaching or the course being demanding, and not much about me). The positive comments that are about me say that I am very nice and the course is well organized. This leaves me confused.

I searched online about gender bias in student evaluations, but almost all articles are about the lecturer's gender, with female lecturers on average receiving lower scores than males. I cannot find research on the gender bias of students. I also cannot find anything similar here on SE Academia. I read other question as well (this and this), but didn't get the answer.

In one of the articles I read that some students are so subjective that they evaluate their (especially female) lecturers based on looks. I have experienced this first hand since one of the students wrote that I am the best looking teacher in the school. So even considering that I am always groomed and nicely dressed I am not sure that a group of students would rate me badly based on my appearance.

Some additional notes:

• the gender ratio in the class is balanced, about half of the students are female
• there was not a single conflict with any student during the semester, which would be amplified among a group of students and reflect badly on the evaluations

Answer 1

I am a female student from Japan and will humbly attempt to answer your question, although I have never attended your lectures if you never taught in my country.

Subjectively, I find that females are on average better at explaining things than males are, but I strongly suspect that males subjectively find that the opposite is true. This has nothing to do with any kind of prejudice: I really better and easier understand explanations by females than by males, on average. Given a choice between a male and female teacher and no further information, I would be strongly tempted by my whole life experience to choose the female.

I guess the root cause of this phenomenon is that the male brain and the female brain are, as research shows, wired differently, and I guess this gives rise to somewhat different thinking patterns or styles of reasoning and learning. So it appears to me that male teachers are better for male students and that female teachers are better for female students, because people of the same sex tend to think in more similar ways.

Such things are difficult to describe, so it is no wonder that you got no hints in the comments of the teaching evaluations, but let me try. I will now try to summarize my own observations about typical shortcomings of male teachers as viewed from my perspective. I emphasize that what I will write is my own observations and impressions based on my own personal experience.

First of all, male teachers tend, in my experience, to explain things by using abstract general ideas and concepts, whilst my thinking processes are mainly driven by analogies to particular instances. Male teachers tend to make very few analogies in their explanations and to operate mainly in terms of purely abstract deductive logic.

Second, the male approach to explain things seems to be turned upside down with respect to what I consider natural. To understand something, I need to be shown a set of simple specific examples that I could use as keys to inferring some general ideas. Contrary to my needs, male teachers start with abstract general principles and methods, build a multi-layered system of definitions and principles, and only then show how that system works in particular instances. In other words, I prefer to go from particular to general, from simple to complex, step by step, and male teachers prefer the other way around.

Third, males tend to have a kind of negative attitude in thinking. Facing a problem to solve, they seem to tend to ask themselves, "What do I have to do to solve the problem? What general principle applies to situations like this one?" I am a much more positive person and tend to ask myself something different, "What steps can I do now to get closer to getting the problem solved?" Males seem to tend to think in terms of necessities, whilst I tend to think in terms of opportunities and making choices. The male approach simply does not resonate with how I am used to think.

Fourth, male teachers tend to be succinct in their explanations, as if they were afraid to overburden students with information. In contrast, I always want more clarifications, more analogies, more examples, more insights from different angles of view, etc. This helps me understand and memorize things.

Fifth, male teachers tend to rarely repeat things or provide alternative explanations, and the result is often disastrous: a failure to understand an idea renders the remainder of the lecture superfluous if the teacher builds on that idea. Apparently, male teachers presume that students will instantly and fully understand everything, every sentence and idea told to them. What makes it worse is that male teachers tend to build their lectures in a highly organized, systematic manner, so it is really critical to understand everything they say. This, in turn, requires utmost concentration and quickly makes me tired. I feel more relaxed when taught by a female.

Sixth, male teachers tend to do their job rather formally, basically repeating what is stated in books, so I often see no point in attending lectures. After all, I can read books myself. What I want from teachers is bright examples, interesting analogies, etc. - everything that can help me interpret and memorize theories written in books.

Seventh, male teachers tend to care little as to why a student is confused and asking a question. Instead of starting a dialogue with a confused student, talking in the student's terms, and building on what the student knows, a typical male teacher blindly imposes his own way of reasoning, essentially repeating what already has been said during the lesson.

Finally, male teachers tend to be simply more boring to me. They talk slower and less expressively, show less emotions, are less empathic, less interact with students, rarely appeal to emotions, and are less involved in general.

The above shortcomings are probably not shortcomings at all from the perspective of male students. As I said, males and females seem to tend to have different thinking patterns or styles of reasoning and learning. So please kindly do not take my post as sexist. I am not saying that any sex is superior. We are just different, as shown by the research mentioned above.

Also, the points made above are tendencies or generalizations rather than strict laws of nature. There are exceptions, and I had a few male teachers whose ways of teaching I really liked. Those teachers were largely devoid of the shortcomings outlined above.

I would like to conclude my post with an illustration of my points. How would a typical male school teacher explain what is the force of elasticity in physics? He would say something like, "The force of elasticity is the force that arises in a body in response to its deformation and acts to return the body to its original shape. An example is the action of a spring." A typical female teacher would rather come up with an explanation of the following kind: "Imagine Tokyo Tower. Have you ever visited it? Great, now suppose King Kong approaches Tokyo Tower, puts his hairy hand on the tip of the tower and pulls it towards himself, thereby bending the tower. To do so, King Kong acts with a force on the tower, and the tower acts with an equal force on King Kong's hand, according to Newton's third law. And that force - exerted by the tower on King Kong - is the force of elasticity. The same name applies to the force exerted by a stretched spring. Or a rubber band. Or just anything that is being deformed in any way and is resisting because of its internal structure. That's what we call the force of elasticity. It's what King Kong has to overcome to bend Tokyo Tower." Learn to explain things in this way, and female students will probably like you more :)

I humbly hope that my post was helpful in shedding some light from the other side of the fence, so to speak.

Answer 2

I would hesitate to draw any conclusions whatsoever from one semester. File this away, and check whether it becomes a pattern, and if it does then think about it then when you have more information.

Although there's not a single incident that you're aware of, I wouldn't be at all surprised if something being "amplified among a group of students" is exactly what happened. It's easy to have a miscommunication with a student that leaves them with a negative impression of you, and for that to spread among their friends. Unless there's a trend in other classes I wouldn't jump to the conclusion that this has anything to do with gender specifically, it may well just be that one group of friends which is mostly or entirely female didn't like you.

Also, I commend you for thinking about this and for seriously considering whether you need to change something in your teaching. I just think you might be overthinking it at this point.

Answer 3

Here are a few possible lines of thought. Of course, we would need much more data and context to know if any were actually the case.

---

Don't forget the null hypothesis. If you carried out this survey for other lecturers (of each gender) for the same course, other courses, other departments, other universities, would we indeed find that your result is anomalous, and by how much? We may have no reason to think it isn't, but unless we know, we don't really know what phenomenon is under discussion.

How do you know your survey is valid? Which factors in your teaching did you measure? Are they measurable numerically?

---

As you suggested and Noah responded to, "amplification" is a possibility. Classroom culture is a factor at every level of teaching. I found this especially during practice teaching in a middle school, where the same students and teacher are together all day and it's hard to have multiple atmospheres or moods coexist in the classroom. The verbal and tacit feedback they get from each other promotes uniformity. One class decides they like a teacher, and it's easy to teach the group even if a few students privately feel differently. One class decides they don't, and it becomes very hard to win them back as a group.

Even when the whole class doesn't respond alike, subgroups often respond alike. In the high school classes I'm teaching now, there are friend groups that break along some classic lines, and each group rather than each student tends to have a certain attitude towards the course and the teacher. I could imagine that gender is a possible subgrouping. Such a group was created just recently in my English course on the basis of content when we watched and discussed a movie on how women are depicted in advertising. The overall class cohesion meant it was a temporary divide, but a subtler cause of division could be harder to heal.

---

Being overly respectful is not likely to be seen as condescending. But standoffishness (even unintentional) is not the same thing as respect. I have an idea of the careful line you're trying to hold by not sitting too close and not shaking hands, and it's a good thought. But consider that a conscious attempt not to overengage can come off as stiff or cold. Some people avoid others because they dislike them or feel disdain for them, which is not the impression you want your students to have.

If you feel more confident that you can comfortably interact with the young men without crossing boundaries than you do with the young women, there might well be a difference in how warm each group perceives you to be. You might even be unconsciously reducing the number of interactions with the girls. Students are highly sensitive to cues about whether their teachers like them.

---

The same neutral approach could also be an issue in terms of content. Qui tacet consentire videtur. You don't always know what you're saying by not saying anything. I was teaching English in a high school and we were reading essays on gun control. Suddenly, the Stoneman Douglas shooting happened, and the government responded about arming teachers and whatnot. To be silent about it would be to invite the students to construe — on their own, without evidence — what I thought of the issues. So the discussion was opened.

Note that in such a discussion, you don't have to even reveal your own opinions, but give the floor respectfully and affirm what's said and the right to speak of various people, especially those who tend to be intimidated or talked over by the class as a whole. But you can also make direct contributions if you want. It's a great idea to prepare students on these issues that they'll grapple with as adults, bringing insight from whichever field you're working in. Anthropology, biology, history, literature, statistics, business — many fields have inroads if you want to show your hand, rather than hide it, for the positive effect it might have on skeptics.

---

I think any more requires knowing about your results, your students, and your class.

Here's a suggestion if you want richer, non-useless student feedback from these students, since the people who were there have more insight than you can convey to us here. You could invite a few students who you think were highly engaged and who had good connections with their peers to sit and share more detailed feedback on the class, if they're interested. "After reading through the anonymous surveys, I'm hoping to explore some points further. I think you might have a valuable perspective to offer, based on your participation in class."

You could then ask some open, non-leading questions, not about your results themselves ("Why was there a gender divide?") but about possible factors ("How did you perceive the teacher-student interaction? Did you sense any favouritism?").

You can also ask colleagues, who will sometimes hear from students or even inquire about how the new hire is doing. They may have some things to share through the grapevine. Another answerer's suggestion of inviting a colleague to sit in on a class and take notes on your teaching is also great. I think here too I'd avoid giving them too strong a lead to look for, beyond the premise that you want feedback; whatever they notice will be more significant if it comes up without prompting.

Do keep in mind what you've said — that you don't know what the significance of your results is. (Not that it stops you from making positive moves.)

Answer 4

It could be almost any of the suggestions already suggested, but my hunch from the information provided (in particular, that the complaints "are mostly about the organization of the teaching or the course being demanding"), is that it may tend to be one of two things:

• Communication style. You may organize and explain ideas in a style that more typical men in your culture use more and follow more easily than typical women in your culture do.

• The minor/major point mentioned in the comment by @darijgrinberg . For example, if your male students are mostly majoring in related subjects, but your female students are mainly taking your course to meet some requirement but they're not particularly interested in the subject, or if they have less experience with courses in your field, your course may seem less intelligible to them because you may not be explaining things for people less familiar with the assumptions and language of your subject, making the teaching seem more disorganized and harder from their perspective.

Answer 5

I have posted this as a comment, which has been referenced twice in the discussion; since it now has been moved to chat, I'm reposting it here for better visibility:

Can you check the major/minor composition of your course by gender? If it is significantly skewed, you may have your answer there. I'm pretty sure students will evaluate a (well-taught) class much lower if they view it as a hurdle on their way to something interesting, and that's much more likely if the class is not in their major.

Here is a concrete quasi-example. There are well-documented huge differences in gender composition between majors. If you are teaching (say) a theory-of-probability class that is required for mathematicians (43% female), computer scientists (18% female) and medicine students (84% female), then a woman sitting in your class will be much more likely than a man to view it as a painful hurdle on their way than as an exciting intellectual journey. There is only so much you can help this with good teaching. Students don't like the non-major classes they have to take, and this pulls their ratings of the teacher down. (I don't have any good data for this, but the mechanism is obvious to me and my own experiences confirm this.)

Note that you can probably check whether this applies to you; most universities let you easily see what programs your students are in, what other classes they have taken, etc.

---

Another possible explanation was mentioned in a deleted answer; I'm going to repeat it here to put it in context. There is a rather significant sex skew in visuo-spatial intelligence with women scoring (on average) the lower (roughly speaking) the more 3-dimensional it gets. This may and may not be contributing to your female students' problems (as well as those of others who have low visuo-spatial abilities). My personal impression is that the significance of visuo-spatial intelligence in learning mathematics is rather minor: With my own lack of it (despite being male), I found a few "topology in pictures" books incomprehensible and am probably getting less mileage out of most graphics than other people; but there is no field of mathematics I found myself blocked from entering. You can easily learn linear algebra without ever seeing a picture. However, you may be significantly exposing yourself to the sex difference if

1. you are teaching material is inherently 3-dimensional (say, some parts of physics, engineering, etc.), or

2. you are relying heavily on spatial visualization.

In mathematics, at least, these cases are easily avoided (just give references to more formalized versions of your arguments). I don't know how it is with more applied disciplines.

Answer 6

One possible way to investigate this further is to ask a trusted colleague to observe one of your classes and give feedback on that. This can be done privately, outside of any officially documented peer-review requirement (which may relax you both and give more honest feedback.) Some people also video record a lecture and watch it with a colleague for possible feedback -- this also leverages the fact that sometimes we're our own harshest critic.

Even if the one observation doesn't bring the gender-equity issue into focus, there might be one or more other things that pop up for improvement, so it could be a good use of time either way.

Answer 7

I'm going to do a bit of a frame challenge on this one: before going down this path, you should actually make sure that there's actually meaningfulness to the disparity.

Let me give you an example that happened to me personally. I just started a new job, and within a few months, we had an All-IT meeting. The company had conducted a survey of everyone that worked at the company, and had found that IT's 'Rate-Your-Happiness' responses were lower than the non-IT and were looking at ways of improving IT happiness. But there was a problem. IT and Non-IT people are different. Ask both to give a "Rate Something X-Out-Of-5", and you're going to have different averages between those two groups. And sure enough, it turned out that it wasn't that IT was unhappier than Non-IT - it's just that they handed out numbers differently. If you looked at the non-numerical Yes/No questions ("Would you be likely to recommend working here to a friend or family member?" for instance.), the two groups had nearly identical responses.

I'm not saying that's necessarily what's happening here. But right now, for your value difference to be meaningful, you have to first assert:

• There are no fundamental differences between men and women that would result in how they rate things in general. I have no clue on this one. And googling didn't help out, either. I wouldn't be surprised if there are differences in the average "Rate something X-out-of-10" between men and women in general. And I wouldn't be surprised if there were no differences at all.
• There are no fundamental differences between men and women that would result in changing how they rate teacher performance between genders. This is debatable. Googling was tough (the results were almost uniform in 'Male versus Female Prof' instead of 'Male versus Female Student'), but I found Bachen, McLoughlin, and Garcia (1999) as well as Basow (2000) that indicated female students rated male profs lower than female profs. While it's not definitive, it's not a stretch to believe that it would indicate that a male prof's reviews would be lower from the female portion of students versus the male portion.
• There are no fundamental differences in campus/college experience between men and women at your university and your degree program. Just for an extreme, we'll take where I graduated - which was a college that had twice as many male students as female, and where the female-to-male ratio in the particular STEM field I was in was a crazy 1-to-50! The women in that field had a wildly different college experience from the men, to the point where I'd hesitate to even guess the scope of it. Expecting how they evaluate professors in general to be identical to the male students is a bit unfair - simply due to the drastically different environment they were in compared to the male students.

So what should you do? Honestly, I wouldn't take the numbers into account at all. Instead, look for anything like "What would you improve about this class?" or "What things did you not like about how the class was conducted" - items with specific points to address.

Because at the end of the day, even if that numerical difference is meaningful, it doesn't actually help you improve what you're doing. To improve, you have to find out why you're given a low score. (Put another way: if you try to solve the problem without actually knowing what's causing the problem, you're just blindly guessing.)

Answer 8

I might as well weigh in on this, with all of the caveats mentioned by everyone else.

IMO, the most telling sentence in your question is this:

It may sound silly, but is it me being too respectful backfiring and seen as condescending by female students?

Let me reframe this: it's possible that you're trying too hard to be respectful to your women students and they're picking up on your discomfort with them and it's making them a little uncomfortable too. (Ask a black friend about white people trying too hard to not seem racist.)

Remember that undergrad evals are largely beauty pageants, so you might be coming off as vaguely unlikable to the women and it's skewing your numbers. A colleague probably wouldn't notice this subtle dynamic.

PS I'm guessing you're in a heavily male-dominated field. Your question beautifully illustrates why gender imbalance is a problem.

Answer 9

Can you get the overall statistics? Specifically:

• Is there any bias observed assuming all answers asked for all courses (at your deaprtment / faculty / university / field?
• Is there any bias observed assuming all answers of your students?

How are you measured? 1-5 stars, 1-10 points? There are zounds of way how to interpret such scales. Can you get histograms how each student evaluated?

I dare to say there are people who never give 10/10 (because there's always something to do better, right). I also dare to say some use only 3-4 values from the 1-10 scale: 10-awesome, 9-acceptable, 8-bad, 7-rubbish; others may use same sorting but with 3 point lower readings...

When I was in highschool the marks were:

1. Excelent (Výborně)
2. Remarkable (Chvalitebně)
3. Good (Dobře)
4. Acceptable (Dostatečně)
5. Unacceptable (Nedostatečně)

You can see that mark 4 is acceptable to pass the exam. Still they who had average around 3 were considered underperformers, those with averages around 2 were okayish and those scoring 1.6 and better were considered excelent students...

See? There are way too many factors in the game, the biased performance readings may be caused by the biased audience, by different scale interpretations...

Also remember: Coincidence does not mean Causality.

Were there any strong complaints? Were your lectures hard to follow? Did they consider you xenophobic? If the answers are No, don't bother yourself and focus on improving your feeling of the courses.

Answer 10

Ask others in your department who have taught the same course whether their scores had a gender difference. If there is an administrator (Vice Chair, say) who keeps records of student ratings, discuss with them. It may turn out that this is a known phenomenon; if so it is nothing to do with you.

It still may be something of concern, but something for the whole department to do something about, not just you.

Answer 11

Unless you are an experienced statistician with access to a good amount of studies and/or a good knowledge of the results of meta-analysis in respect to performance evaluations, please refrain from reading anything into this.

By education and cultural background, female students may have a different scale of what the evaluation scale means (when women rate themselves, and men rate themselves the difference is significant, and I assume that some of this may also be the case here). So it's very likely that by your approach and data basis you are even able to validly conclude that you are rated especially badly by females.

Answer 12

The sample is not independent so I think that you cannot say much about statistical significance. One group of friends might be of size 20 and one person complaining about you might change the whole group's perception.

There is the bias of thinking that what is easy or intuitive for you is the same for others. The genders have their differences in learning. Men are generally more visual. Your teaching style may suit men better. Mitsuko gives in her answer examples that I see stemming from the gender differences in learning.

There are also cultural differences. Existing expectations of what a representative of gender should be and how they may behave are still true. You try not to be a sexist might distort how you perceive society. Society feeds some behaviour to the genders. Women are accepted to blame external factors for their failure whereas from men that would be seen as weakness, so men tend to blame internal factors. It is easier to see this from my perspective as the difference is probably more radical. In my nation, there is mandatory military service for men. This changes dramatically what men expect and accept from the professor as they are a year older and have better discipline. More men go through the military probably also in your country. Men are taught to cope and overcome the pressure. If men do not manage they take themselves the responsibility from not being successful. They would not blame you for requiring them to do more work than they expected or are credited for. This would fit your case as you say that you had the students doing more work than required.

Answer 13

A few points. First, ask your department if they can give you the historical average scores for the course (before you taught it) broken down by gender. This will give you a sense as to whether this effect is typical (related to the subject, university, etc) or you personally.

Secondly, as @Elizabeth_Henning mentions, this quote has a few clues

It may sound silly, but is it me being too respectful backfiring and seen as condescending by female students?

Elizabeth points out that you might be trying too hard and it's making the female students uncomfortable. This is certainly possible. But here is an additional point on this. Make sure you are trying equally hard in the company of the male and female students. I had a professor that was so scared of committing sexual harassment that he always acted super formal around female students. While there is nothing wrong with acting formal around female students, the problem was, he was quite chummy with the male students. He would tell jokes with them, maybe even give them a pat on the back. While probably nothing this extreme is happening here (you say you are very careful about this sort of thing), be sure to treat the genders similarly. If you are formal with the female students and more informal with the male students this will look like you are even favoring the male students. Probably not happening here, but its worth pointing out for others.

Answer 14

Possible paths for investigation that don't seem to have been fully fleshed out:

(1) Correlation is not causation: are there any statistically significant differences between the male and female students taking your classes, such as what prerequisites they have taken, what year and major they are in, what related classes they have taken? For example, it may be that your teaching is more geared to students in a male-dominated major than a female-dominated major, who are both required to take your class, but with different educational backgrounds.

(2) Are you treating male and female students the same way? Are you sure? It might be worth having a colleague sit in on a few of your classes and count how many times you call on male and female students, for example, and record (on a simple scale) the tone you take with each of them.

(3) Are your teaching materials gender-neutral? Are you sure? For example, a book I read not all that long ago about a woman's experience at Harvard Business School included an anecdote about lesson module with a business plan that talked about the average single woman's hopes and dreams for marrying a successful man. You don't sound like someone who would be that unaware, but there may be more subtle issues with your teaching materials.