122. Differential Grading Policies

Students generally receive lower grades in STEM classes than they receive in other disciplines. In this episode, Dr. Peter Arcidiacono joins us to discuss how these differences in grading policies across departments can help to explain the relatively low proportion of female students majoring in many STEM disciplines. Peter is a Professor of Economics at Duke University.

Show Notes

Transcript

John: Students generally receive lower grades in STEM classes than they receive in other disciplines. In this episode, we investigate how these differences in grading policies across departments can help to explain the relatively low proportion of female students majoring in many STEM disciplines.

[MUSIC]

John: Thanks for joining us for Tea for Teaching, an informal discussion of innovative and effective practices in teaching and learning.

Rebecca: This podcast series is hosted by John Kane, an economist…

John: …and Rebecca Mushtare, a graphic designer.

Rebecca: Together, we run the Center for Excellence in Learning and Teaching at the State University of New York at Oswego.

[MUSIC]

Rebecca: Our guest today is Dr. Peter Arcidiacono. Peter is a Professor of Economics at Duke University. Welcome, Peter.

Peter: Thank you.

John: Welcome.

Peter: Thanks for having me.

John: Our teas today are…

Peter: Green tea. I got it specifically for this conversation.

Rebecca: Excellent. I have English afternoon.

John: I’m drinking Spring Cherry Green tea.

Rebecca: Oh, that’s a little switch for you.

John: I’m looking forward to the spring now that it’s 10 degrees outside here in Oswego in January when we’re recording this.

We’ve invited you here to talk about your December 2019 working paper co-authored with Thomas Ahn, Amy Hopson, and James Thomas, on Equilibrium Grade Inflation with Implications to Female Interest in STEM Majors. In this paper, you know that disciplines with harsher grading policies result in higher earnings than the earnings received by those majoring in disciplines with easier grading policies. Can you first describe what you mean by harsher grading policies?

Peter: Well, harsher grading policies… you look at average grades in STEM classes and economics, they tend to be a lot lower than in the other social sciences and in the humanities. In the school we look at, which is University of Kentucky, it’s a third of a grade lower. And that holds true across many other colleges I’ve looked at, be it the Duke University or Berea College, it just seems to be a general pattern that the STEM classes and economics tend to have lower grades. And in fact, there was a very nice book written about this with Duke data called Grade Inflation: a Crisis in Higher Education or something like that written by Valen Johnson a few years back, and it was making that exact point about the big grading differences across disciplines.

John: How does your model explain the grading differential across departments?

Peter: Yes, so that to me is where the big contribution of the paper lies, though. I don’t think our model’s particularly satisfactory in this regard. I think lots of people have noticed in the past these differences in grading standards. Where I think our contribution is, is thinking about how those grading standards impact different groups of individuals, but also in what you just asked about: What are the sources of this? And I think a portion of it is certainly demand for courses. So, at the school we look at, the STEM classes are twice as big as the non-STEM classes. An analogy from my own experience… when I first got to Duke, no one signed up for my grad labor class. And that was a bit of a disaster, because that would mean I’d have to do a new prep for some other course. And what I did is I bribed two students to take the class. And the way I did that was I said, “The class is going to be: we’re going to write a paper together.” And it was a great experience for them. They did get a paper, they had a revise and resubmit when they went out in the job market. And what do you know, they both got A’s in the class, but now I’m a little more popular. Not everyone gets A’s and I’m not co-authoring papers with everyone in the class. So I think part of it is these differences in demand for the classes affect things. You hear about weed-out classes, they’re weed-out classes because there’s plenty of students who are interested in taking that class, so they can afford to do that. If you’re in a discipline that has low demand, you’re not going to have a weed-out class because you’re not gonna have any students. So, I think that’s a portion of it. A portion of it can also be things like differences in preferences, maybe it is the case that some disciplines are just meaner than other disciplines. And a third component may be that it’s just easier to grade those classes, it might be easier to grade a math problem than an essay. I’m a little skeptical of the last one, in part because when I talked to high school students, it doesn’t appear to be the case that the math classes are that much easier than the humanities classes, for example, or grade that much harder than. And I think part of that is in high school, you have to take those humanities classes and when you get to college, you have a lot more choice and that influences things.

John: This also seems to relate to the academic labor markets. Because there’s so much demand for classes, there’s often more demand for faculty. So, faculty are not under as much risk of losing their job if they don’t fill their classes. While in markets in those disciplines where grades tend to be lower, labor markets tend to have a little more slack in them. And faculty perhaps may, using the analogy you used before, have to be a little more generous to make sure their classes fill so they can keep their jobs, particularly when there’s a large component of adjunct faculty and contingent faculty.

Peter: I think that’s exactly right. I’ve heard some economics professors brag about their low teaching evaluations. I just thought they can get away with that in other disciplines.

John: And we see that in many STEM fields as well, at least anecdotally… certainly not on our campus…[LAUGHTER] But certainly many STEM disciplines do have that reputation of bragging about keeping their grades low and having high standards and such things.

Peter: That’s right. You know, in my view, it’s when you change what you grade around, that doesn’t necessarily mean that the standards… how much work is required in the class may stay the same. And that’s where, I think what’s interesting about men and women, is that women actually study much more than men. So, if you change the workloads, that’s more of a deterrent for men than it is for women. So, you can think about different mixes of either giving high grades but with high workloads, or lower grades, with lower workloads and student life, on average being different between those things, but they won’t be indifferent between men and women. At least on average. Obviously, this is just an on average statement. There are some women who are not interested in studying hard, and some men who are. On average, that appears to be the case.

John: And you do know, though, that the STEM disciplines not only assigned lower grades, but they also require larger workloads based on the survey data you were looking at, I believe.

Peter: That’s right, substantially higher workloads. And that was my own experience… that where I went to college, I guess economics hadn’t quite crossed over to STEM at the time, but I started as a chemistry major, and there the formulas are a lot more complicated and you had to memorize them. In economics, I didn’t have to memorize the formulas and they were much simpler than those in the chemistry major and I had to work a lot harder in those chemistry classes.

Rebecca: Can you talk a little bit about how the difference in grading policies impacts underrepresented groups in these different disciplines? So in STEM, for example, where you’re saying there’s these stricter grading policies, there’s less women, there’s less underrepresented groups in general.

Peter: That’s right.

Rebecca: How do those things connect? Can you talk us through that a little bit?

Peter: Yeah, actually I had caused quite a stir here at Duke back in 2011, talking about underrepresented minorities, because you do see, for example, African Americans at Duke… They came in just as interested in the STEM classes in economics as white students, but they left at a much higher rate. And so the negative headline was something like, you know, “Potentially racist study says Black students are taking the easy way out,” and that was not the case. It was that they’re coming in with different levels of preparation, due to pre-college inequalities. We have serious differences in pre-college education. And that’s resulted in this coming in and then finding that it’s very hard to succeed in the STEM disciplines. For women, the story is different. And there, because women come in just as prepared as men and are interested in studying a lot more. For them, it’s not the workloads that are affecting this, but more the lower grades themselves, given that they’re working so hard, and then to see these low grades seems to have a bigger effect on women than men. I think one of my very interesting result from some of the work that Stinebrickner and Stinebrickner have done is actually asked men and women, “What grades do you expect you’re going to get?” And then they’ve showed them their grades later and said, “Now, that didn’t work out quite as you expected.” They got lower grades. Why do you think that was, and men were much more likely to say it was bad luck. And you could see that they actually believed it was bad luck, because they continue to take the same classes where they were getting low grades, and they didn’t change their study habits. Whereas women were much more likely to say that it was something about them, either that they needed to study more, or they weren’t as prepared as they had hoped to be for the class. And then you can see that they ended up being more likely to switch to different types of classes and to change their study habits.

John: I think you also found that women tended to not just equally prepare, but I think women came in with somewhat higher GPAs, if I remember correctly.

Peter: They do come in with higher high school GPAs, for sure, and they get better grades in college than the men, both in STEM classes and in non-STEM classes.

John: So the difference seems to be that women are more sensitive to the lower grades or they tend to blame themselves for the lower grades, while the men tend to be less cognizant of the fact that they may be the cause of the lower grades.

Rebecca: I feel like I’ve been socialized to blame myself for not being adequate.

Peter: Yeah, it’s been remarkable, you know I have five boys. And they’re not socialized that way, despite my best efforts to get them to really care about their grades and to think that studying even matters. Here I am, a researcher who knows that studying matters, and yet some of my boys are convinced that if they do poorly, it’s bad luck. It has nothing to do with their studying.

John: One of the things I assigned in my labor class is a reading from, I think it’s a sociologist, who noted that women seem to be more concerned with what she termed “local status,” while men tend to be more concerned with “global status.” Basically, the argument is that men tend to be more concerned with their income expectations and the impact that the courses in the majors have on their grades, while women tend to be much more concerned with the relative grades and seeing grades as a perception of their worth or their value in some sense, which makes them much more sensitive to that grading differential between majors. While men tend not to be as concerned about their local standing, their relative status, while they tend to be more concerned with how the degree will impact their future earnings.

Peter: And I think it’s compounded by the fact that I don’t think people know when they go to college how different the grades are. I think that they figure it out rather quickly, but at first they just don’t know that they’re graded so differently. I mean by the time they figure it out, they’ve already made the switches, which is too bad.

Rebecca: So what do we do so that there’s more representation in these disciplines?

Peter: Well I think at a minimum, that universities should at least have an obligation to tell people, “Here’s what the average grades are in these different classes,” so that students can make informed choices. I personally am in favor of doing something where, like they did at Wellesley, where they put in something that either capped the average GPA in the class, or having all the classes that are above a certain size give the same median grade so that people are not having their decisions distorted by the fact that one department is giving lower grades than another department. It’s not as though they’re learning less in these STEM classes, just because they’re getting these lower grades. They’re actually working very hard in these classes, so, it would be nice if that was somewhat remedied.

Rebecca: I wonder if even knowing the information across institutions and knowing that people get lower grades in these fields, no matter what school they’re at, would actually be useful to those groups.

Peter: Oh, I think it’d be very helpful. Yeah.

Rebecca: Not just within your own institution.

Peter: That’s right. Yeah, I think that’d be great. Because then you know what you’re getting into. And maybe when you get that B in the economics class, you’re not feeling like you failed, that that is a much more typical grade than what you expected.

John: I think a lot of people might object to the notion that there should be the same GPA on the grounds that you’re forcing students into this curve where if you happen to have a class that does really well you might be lowering their grades, which doesn’t always happen in the STEM field, but I think there might be some resistance to that. But the notion of transparency and making people more aware of that early on might help reduce that differential effect by gender.

Peter: That’s right. And I think also making departments aware… the implications that the level of their grading has on things like the gender composition, should matter. I don’t want the way I grade my class to be a deterrent for women taking economics classes. I think that that’s bad. And I’ve changed the way I grade a bit in response to this. If other disciplines are going to give much higher grades, and I want to have a good gender representation in my class, I’m still assigning the same amount of workload. So I’m not making the classes easier, but I am changing the level that I’m curving around, and I typically teach a class that has over 100 students in it.

John: And this is especially important because a large share of the gender differential in earnings is tied to major choice, because the fields that women are more likely to enter tend to offer much lower wages on average, while the STEM fields tend to offer much higher wages. So the differential impact on major choice has quite a bit of an effect on the gender wage gap.

Peter: That’s right. And this is potentially a very low- cost way, at least from a pure financial standpoint, to changing that. It’s not going to solve all of the gender issues in economics or in STEM fields. It’s not going to come close to solving it, but it can put a dent in it, where it’s often very costly to change something like that. Changing a climate is expensive and difficult.

Rebecca: I think I would have been one of those students that would have moved. I’m a person who does a lot of coding and things and could easily have been in computer science. But I’m not, I’m in graphic design, and it’s a field that is much more populated by women. I got much higher grades in the things that I did, and I was very sensitive to grades. So as a student, if it was something that I didn’t do well in, then I thought I couldn’t do well in. And I had a very fixed mindset when I was earlier on in my career. Obviously, it’s much different now. I do much different things and explore other things. But at that time, I think I was incredibly sensitive to that. If you came in as an A student and then it’s like, all of a sudden you’re not getting A’s, it’s like “Woah…”

Peter: Yeah.

Rebecca: “…maybe college isn’t for me, even.”

Peter: Yeah. And that’s just really unfortunate that decisions will get based on nominal differences in these grades, that don’t really have content outside of something that is relative, relative to whatever the professor’s curving around. Another advantage to doing something like this, too, is potentially getting spillovers. Because once you change the gender composition a little bit, then that makes the environment more comfortable. And then that can have multiplying effects.

John: So, why don’t departments want to do this to improve the gender composition of their classes and to get more students in their classes and persisting to graduation in their majors?

Peter: Well, this is sort of where the equilibrium aspect comes in. Because even if you want to do this, everybody ends up ratcheting up their grades, because some departments really need more students. And in order to attract students, you’re going to have to do something if the students don’t have an innate demand for your field. And people have been talking about this for some time, that demand for the humanities has been falling and what to do about it. So, I think that the pressure’s sort of on both sides, even if the STEM departments decided, “Look, we want to have our grades be a bit higher for this reason,” that’s going to lead to response by non-STEM professors as well. So I think it’s hard to get rid of the grading differences across fields, absent doing something that’s pretty draconian.

John: By imposing a grading standard that’s equal across departments.

Peter: Exactly.

Rebecca: I wonder if initiating conversations between the humanities and STEM fields about the interdependence there actually are on those two fields and that people in the humanities should learn things about STEM and people in STEM should learn things from humanities and that there could be more of a relationship there that would actually help start to equalize some of those needs, because there could be some referrals to certain classes or other things that might help boost some of those relationships. I don’t know.

Peter: That’s right. So right now we’re playing sort of a non-cooperative game and move it more towards a cooperative one. Yeah.

Rebecca: Yeah, it’s competitive rather than thinking about “Why are we at a liberal arts institution? And what does an environment like that or general education offer?”

John: Well think about what would happen, though, if grading standards were relaxed in the STEM disciplines. More students would major in them, enrollments would tend to fall in the non-STEM field, and that would tend to lead to fewer jobs for non-STEM faculty. And maybe that’s not something that departments would react well to, and they face the same incentive to just inflate grades.

Rebecca: I don’t think my suggestion was necessarily to do something about the grades directly, but rather to start helping to increase some demand for humanities and helping out some of those…

John: You mean, more interdisciplinary programs?

Rebecca: Yeah, I’ve been thinking a lot about, even for my students, the benefits of them exploring a lot of different disciplines to be more effective in their own, and understanding how other people problem solve, and think through problems, and explore the world.

Peter: I really like that too because if you end up providing these skills that are then rewarded in the marketplace through taking some of the STEM stuff, but combining it with the humanities, that will increase demand. And then that will naturally even things out a bit more.

John: And you could move more from STEM to STEAM.

Rebecca: Hey, I didn’t say it, but I’m totally a STEAM gal. [LAUGHTER]

John: Well I was expecting that at any moment. And just in case some of our listeners don’t know what STEAM is, would you like to…?

Rebecca: Yeah, included the A …it’s for the arts, [LAUGHTER] which I am a part of.

Peter: Yeah, they’re doing that at my kids’ high school, you know, having a STEAM program to try to get that integration happening a bit more.

Rebecca: There’s some research happening about the need for these creative outlets to help with innovation and things in STEM fields. So, there’s certainly some research suggesting this interdisciplinary approach to all of our disciplines could actually improve the work that we do.

Peter: Makes sense.

John: And many STEM students tend to specialize as much as they can in their field. So, that broader training could be useful in the future in making them more creative perhaps.

So, we always end with the question: “What are you doing next?”

Peter: Well, it’s been interesting, nothing quite related to this. I got derailed because I’m actually an expert witness in the Asian American discrimination case against Harvard, and that is taking up all of my time. We released a couple of papers on legacy and athlete preferences at Harvard, and we’re going to be writing some more out of my reports for that case. That’s what’s keeping me busy.

John: Very good.

Rebecca: Well, thank you so much for joining us, this was really interesting, and I hope that it launches more conversations across disciplines.

Peter: Thanks for having me, I really appreciate it.

John: Thank you.

[MUSIC]

John: If you’ve enjoyed this podcast, please subscribe and leave a review on iTunes or your favorite podcast service. To continue the conversation, join us on our Tea for Teaching Facebook page.

Rebecca: You can find show notes, transcripts and other materials on teaforteaching.com. Music by Michael Gary Brewer.

John: Editing assistance provided by Savannah Norton.

[MUSIC]