Some Thoughts on Grades
More accurately, some thoughts on commentary about higher ed.
Mark Oppenheimer, a professor at Yale, writes the following for WaPo: There’s nothing wrong with grade inflation which is as much a critique of grades as it is about grade inflation, per se.
My general reaction to the piece is that grades are the worst system, except for the alternatives, but that wasn’t what really struck me in the piece. As a professor with a bit over twenty years teaching experience, roughly 18 of which with my current institution, the following jumped out at me:
I’ve taught humanities subjects for 15 years, and I still can’t say very well what separates a B from an A. What’s more, I never see the kind of incompetence or impudence that would merit a D or an F.
I can see the difficulties in the A/B distinction, although I don’t think it is quite as mysterious as he makes it out to be (but I cannot deny that there is a certain amount of subjectivity and art in evaluating essay responses, or even short answers for that matter). Still, I have to also say that, yes, I have had plenty of experience with the one, two punch of incompetence and impudence that merits a D or an F. This usually occurs either from blatant error or, more likely than not, simply not doing the work. This is striking not because I wanted to share divergent grading experiences, but because of the additional bit of information: “…I have taught at Stanford, Wellesley, New York University, Boston College and Yale…”
That, in an of itself, explains a lot. If you are teaching at an elite institution, you have (by definition) elite students. Now, even some elite students fail to do the work, but the odds are pretty high that if someone can get into Stanford that they can do, at least, C level work and likely can do better than that on a fairly consistent basis.
This is striking to me, however, because of that obvious point (elite institutions have elite students) nor am I denigrating the students I have taught for most of my career. Rather, this column underscores a problem that a lot of higher education discussion seems to suffer from, especially in places like WaPo. They far too often attempt to extrapolate from a very narrow place: the POV of the elite schools. Most universities are not Yale, so it would be far more useful to look at these issues from a broader perspective.
(And at the moment, getting rid of grades sounds awesome, as part of why I am writing this post is that I am procrastinating from, well, grading).
I’ve assigned letter grades, and in one instance, written evaluations. I much prefer the latter. Evaluations are a lot of work, but what you end up with is a much more honest and thorough appraisal of the student’s assets and liabilities.
If the elite schools are really as rigorous as they claim, why wouldn’t they give out just as many Ds and Fs as lesser institutions?
Mike
@CSK: In the abstract I can certainly see the value in written evaluations. The question becomes, however, one of practicality. A faculty member teaching at an institution such as mine is teaching between 4 and 6 classes a semester (assuming some overload teaching, but 4 at a minimum) and has somewhere between 60 and a 150 students (rough estimates) during that time period and has service and research obligations in addition to teaching. Finding the time for comprehensive written evals at the end of the semester instead of letter grades would be onerous (because one assume one would still give assignments that would also need evaluation).
@MBunge: Partly it is an issue of what you think grades are. If you think grades are truly supposed to be a measure of distribution (a C is average and therefore most students should get Cs) or whether you think grades are assessments of an individual piece of work.
Put another way: if one gives a math exam and everyone gets 80% or more correct, should they not all get Bs or should we curve the grades to reflect a normal distribution?
@Steven L. Taylor:
Yes. Class size does make a huge difference. I too had committee and other service obligations as well as my own writing to keep up with, but my classes were workshop or seminar-sized.
Still, I always felt as if I’d done a good day’s work after having written the evals, because they were a much more accurate assessment than a letter grade. I do know that prospective employers and grad school admissions committees found them very helpful. You can,for example, commend a student’s willingness and ability to work hard even if his or her mastery of the subject isn’t complete.
My experience in grad school has led me to believe that grades at the masters level are largely a function of one’s ability to boost the instructor’s ego. Regurgitation was preferable to critical analysis. And God forbid if your thesis adviser doesn’t agree with you.
@Pch101: Grades in grad school tend to be A, B, and C with C=pretty much failing (and, indeed, too many Cs usually leads to academic suspension).
My overall experience was that this was not the case, but certainly did encounter some faculty who behaved that way.
This is true.
@Steven L. Taylor:
The curve is the system used at the Harvard Business School, and it’s rigidly enforced. It never really struck me–from a teaching standpoint–as fair, but it may have existed in part to remind the students that life is often unfair–and arbitrary.
Steven: “That, in an of itself, explains a lot. If you are teaching at an elite institution, you have (by definition) elite students. Now, even some elite students fail to do the work, but the odds are pretty high that if someone can get into Stanford that they can do, at least, C level work and likely can do better than that on a fairly consistent basis.”
There was a letter to the WSJ a long while back, about what a professor at Stanford called ‘grade deflation’. He had gone there as an undergrad, left to get his Ph.D., taught there for a while, went into administration, and then returned to teaching.
He claimed that in the time since he had been an undergraduate there, the student body had improved radically; Stanford went from being a coastal CA school to a world school, drawing from a far larger student body. He felt that most of the work there was A work.
@Pch101: “My experience in grad school has led me to believe that grades at the masters level are largely a function of one’s ability to boost the instructor’s ego. Regurgitation was preferable to critical analysis. And God forbid if your thesis adviser doesn’t agree with you.”
My experience was 100% the opposite.
“If the elite schools are really as rigorous as they claim, why wouldn’t they give out just as many Ds and Fs as lesser institutions?”
Depends upon what you want it to mean. My son took a math seminar where 3 of the kids in the class had placed in the top 100 on the Putnam in multiple years. Brilliant kids. If you grade on a curve, those kids get the A grades. However, most of the kids in the class had placed somewhere on the Putnam at least once. and would likely be stars at a lower tier school. They mastered the material, just weren’t as bright as those three. So, should they get lower grades because of the curve, or get As because they really did have an outstanding mastery of the topic? I don’t really think there is a right or wrong answer here, it mostly comes down to your philosophical preference.
And of course, that doesn’t even begin to broach the topic of how they get those grades. In my day, I am old, graded homework was rare. Mid-terms and finals, sink or swim. Today, homework counts more than exams for a lot of courses. I think that has changed the meaning of grades more than grade inflation. As an employer, it makes me a little more suspicious of GPAs.
Steve
@steve:
That varies quite bit still. I cannot speak to, say, math, but in political science the likelihood is probably that there are 3 or 4 major grades (i.e., exams or research papers) and that’s it. I would be surprised that if “homework counts more than exams for a lot of courses” (that isn’t even the case for my high school age kids).
My favorite class was a grading perspective was an upper-level ecology class. It was mid-size (60 students). The grade was totally what I would call performance based. There were three projects (including the take home final). If you did all three you got an A. If you did two, you got a B, and one was a C. I think none was an F. The kicker was that the each project had to be A level work but you got three chances to get that project to the A level. It was probably a lot of work for the professor and his grad assistants but I really liked it. I liked knowing how to get an A. I liked learning the process of getting a project to an A level. And I liked the effort of the professor and assistants.
My wife went to a small college where all evaluations were written. She really liked that and was tempermentally suited to that system.
@Pch101: Been there, done that. When I was getting my MA, my adviser, having read the first paragraph of the rough draft of my last chapter, declared,”I’m simply not going to endorse the exploration that you wish to make in this final part. Find a new adviser or dump your argument.” Fortunately, the department chair assisted me in finding a new adviser and committee to finish out my thesis.
@MBunge: elite and highly selective institutions shouldn’t be admitting students that can’t do college level work in the first place, so by definition you would expect very few F’s and D’s.
@Scott:
For my take on the question, I see your statement above as one of the key then/now differences. Where I graduated from university and “back in the day” (late 60s early 70s) the syllabus for most of my classes was, literally, written in the upper left hand corner of the chalkboard on the first day of class. It fit comfortably on an index card for those so inclined. I stopped teaching in the states in 2005 and my last department chair suggested to me that my syllabus (at 10 pages) was about 20 pages too short and needed to include the lesson plans for each class session, including a detailed description for the students of how I would evaluate their success in each lesson activity, among other details. Where I had to guess how my professors would score my papers, my syllabus provided a detailed scoring rubric defining all of the elements of evaluation, what represented superior, good, fair, poor performance in each element and how each element would be weighted. If most professors provide such transparency, it is difficult for me to understand how students DON’T get A grades.
And then I went to Korea, where in my class of 35 students, I awarded 3 As, 5 Bs, and the rest would get Cs regardless of their percentage scores or abilities. One of my last classes before I returned to the states had the dividing line between B and C at 87.314% (no, I am not making this up!).
Students who were discipline problems were usually given D grades (not passing and the course cannot be repeated), but students who failed to do the work were given Failing grades and removed from the curve altogether–which sometimes lowered the number of As and Bs that could be awarded
@Steven L. Taylor: “if one gives a math exam and everyone gets 80% or more correct, should they not all get Bs or should we curve the”
In college, if you give a test and everyone gets 80% or more correct, the test is too easy.
Class work should’t be like the Kobayashi Maru where you try and fail people. But if your students aren’t occasionally getting Cs, you’re not challenging them enough. And every so often, a student is going to be too busy, too lazy or too drunk and that C is going to be a D.
On the other hand, a system where a C is considered failing is also ridiculous.
At an elite institution with more restrictive admissions, you shouldn’t be giving out the same proportion of Ds and Fs as ol’ State U because you are dealing with students that have a higher level of talent, greater ambition or at least superior educational backgrounds. But an educational environment where no one gets a D or an F is either not as challenging as it claims or is at minimum giving students an inaccurate view of their own accomplishments.
Mike
@Just ‘nutha ig’rant cracker:
In my case, I was able to use to (subtly) engineer a squeeze play involving the head of the program (who agreed with my argument and was pleased with the manner in which I presented it) so that my adviser had no choice to capitulate. I would like to think that logic and reason prevailed, but the process itself proved to be political.
@MBunge: Every discipline requires a different approach to grading and assessment.
I teach Principles of Accounting (sophomore level sequence). Either someone can take a narrative and turn it into a journal entry, or they can’t. Either they can calculate bond amortization, or they can’t.
If someone masters the ability to turn narrative descriptions of economic transactions into valid Accounting data and the ability to take source document information and use the rules and procedures of US GAAP to correctly quantify that information, they deserve to pass my class. If they don’t, they don’t.
CSK: I have 7 classes this semester. They range from 20 to 35 students each. Two are online. Two are face to face and 8 week. Three are standard face to face 16 week.
Any written evaluations I could provide would only be a meaningful assessment of the top 10 or so and the bottom 10 or so. Everyone else would end up with some sort of boilerplate assessment that really would boil down to “Did work worthy to earn an A/B/C/D/F.”
@MBunge:
That is simply not true as a blanket statement (and my students earn plenty of C, D, and Fs, so that isn’t the issue).
Back to the blanket statement: if you have a small class of say 12 students who are all exceptional, then it is possible that none of them deserve lower than a B. By the same token, one could have a class where no gets an A (but without more information than that you cannot assert that the test was too hard).
@Chris Lawrence: Indeed.
MBunge seems to be the only person to have picked up on the distinction between objective and subjective material.
I was a TA at an Ivy League university, in an engineering program. In theory, if everyone got the problems right they would all have gotten A’s. In practice, the performance was so bad that the tests were graded on curves and a lot of unacceptably poor performance got passing grades. I saw a lot of work that richly deserved a D or F.
I was a professor at a mid-rank public university. I had a few great students, a lot of mediocre students, and a fair raft of terrible students. The material I taught didn’t care about background or preparation or family or world view. Either you knew how to compute the limiting distribution of a Markov Chain, or you didn’t. Your program ran correctly, or it didn’t. You used the correct statistical test, or you didn’t. You implemented the algorithm correctly, or you didn’t.
Grade inflation is what happens when you give progressively more and more partial credit for work that, in the real world, would lead to failure. Bridges collapse, or they don’t. Telephone networks can handle the traffic, or they can’t. Assembly lines are profitable, or they aren’t.
I also have some liberal arts degrees, and grading in those fields is a wholly different matter. It’s not about whether you’re right or wrong; it’s about how well and eloquently you argue. I’m OK with that, but it makes it clear that grade inflation means something different in those fields.
@Steven L. Taylor:
Oh, Lord, yes. Now apply that same principle to a school with open admissions. Every single section is completely different. The most motivated, full time students fill up the late morning / early afternoon M/W and Tues/Thurs sections within hours of registration opening. Late afternoon and MWF 10 – 11 and 11 – 12 sections fill up next. Early morning sections and evening sections fill up the week before the semester starts, if ever. And the evening sections are a mix between eager adult learners and the kids who signed up two days before class started.
The same thing applies to the online sections. We open those one at a time, as they fill. Section 001 is all A and B students, and from there is shifts, quickly. The last section to open is almost always full of people who were enrolled in face-to-face sections that were closed due to insufficient minimum enrollment numbers and people who couldn’t find a section that fit into their schedule.
So if one is to curve, how to do you it? By section? By population? I don’t exaggerate when I say that the very worst performer in a M/W 12:00 – 1:15 course of mine might have a 78 and the very best performer in a T/Th 4:15 – 5:30 that got added two weeks before the semester opened might have an 85, based purely on raw scores.
@Steven L. Taylor: http://www.evergreen.edu/evaluations/
They have done it for years.
I find myself largely in agreement with Steve’s essential point, and his experience.
Separately, I think missing from the discussion is “playing to the level of competition.” Elite schools may have more resources or reputable profs, but rarely, in my humble opinion, decisively so. But as is well recognized in sports, a bigger, more talent filled stage requires upping the level of play. Grades or whatever be damned.
As for the notion of disagreeing with the prof, my thesis prof worked on the Manhattan Project and would just about take your head off not if you disagreed, but if your thinking was sloppy. Perhaps related to the relative objectivity of the physical sciences, or simply because of the stakes one might play for.
@Gromitt Gunn: Exactly–you cannot make judgments about any of this without more information than just grade distribution. And the nature of the institution matters substantially.
“They far too often attempt to extrapolate from a very narrow place: the POV of the elite schools. Most universities are not Yale, so it would be far more useful to look at these issues from a broader perspective.”
At the risk of sounding like a curmudgeon, if after a couple years in the workforce one is still talking about grades or where one graduated as opposed to what they can do then, well, something has gone very wrong.
@Tony W: Being familiar with Evergreen since it opened in the late 60s (one of my class mates transferred there in 1971), I’m not sure that it is fair to compare it with other institutions. The whole program was designed from the beginning for narrative evaluations in ways that other institutions are not. Also there is a significant enrollment difference from other schools. Evergreen has a total of about 4800 students. By comparison, Wikipedia tells me that Troy has just under 20,000.
@DrDaveT: You make a very good point. I agree that STEM courses should be able to rely on more-or-less strictly objective criteria. I teach history, where it seems that few issues are permanently settled. The ability to analyze information, frame a suitable response, and express it clearly and concisely is essential for my students. Alas, it’s very often lacking. The number of students who require remedial instruction is appalling.
@Just ‘nutha ig’rant cracker: Yep – I have some friends who attended there as well – to a person they give Evergreen credit for improving their writing and critical thinking skills to a degree that would not be possible in other university systems.
To your point I don’t really understand why 4,800 students is any different than 20,000. If an institution is committed to doing a quality job on evaluations they will spend the money on faculty & learning infrastructure instead of multi-million dollar head-coaches, high-tech sports arenas and food courts.
It’s essentially a student/teacher ratio problem, as well as one of priorities.
@Tony W: Indeed, the relevant issues would include: class size, teaching load, number of preps, service expectations, research expectations, the calendar, and other factors.
I do not know enough about Evergreen to comment intelligently.
@Tony W: The point on enrollment is specifically student teacher ratio. I also went to a small enrollment university. The largest class I ever took had 30 students in it because the school did not have the logistical ability to offer a second section of it. The others averaged 12-18. By comparison, the smallest class that one of my friends at the University of Washington (student population then, 19,000 underclass students) was 45.
When I attended William and Mary, one of the top issues was how W&M was going to address grade inflation at other institutions when it came to grad school acceptance. The thought being that W&M did not inflate grades relative to other institutions thus hurtING W&M undergrads when it came to grad school aceptance. I have no idea if that was true or not, but it was a big issue while I was there. I am sure that most people like to think their school doesnt inflate grades, its always the other place that does.
I went to a large public university (UC Berkeley), I was generally a social sciences-liberal arts student – I majored in economics and minored in statistics, took 2 years of calculus and linear algebra and differential analysis AND grading on or off the curve was all over the place. No method consistently applied.
A few of my economics courses were strictly curved and that seemed somewhat unfair, while some courses, particularly in mathematics, were wildly difficult. I vividly remember a Applied Differentials course where I once received 37 points out of 90 possible, and I was the 4th highest grade in a class of about 80. That test was too hard, I fail to see the value in a test where about 50% is a top score. However I think the professor was trying to both weed people out, and/or let us know that we were going to have to get with it to make it successfully in THAT course.
Some social science courses were pretty easy Bs, others, especially in History where a lot of writing and interpretation was in play, were very tough Bs.
I’m secular on this issue of grades. My university was a fast lane, and you had to get with it fast because, unlike places like Stanford that are small-ish, more collegial, and professors have more time to hand hold. My Berkeley (in economics, applied statistics, and math) was not a grade inflated place. I know that any one who slogs through ANY engineering program at Berkeley is not going to find grade inflation anywhere.
Anecdotal,
I had a prof who expressed his frustration by awarded only three grades: A, C, and F. “Your work is either Acceptable, Cwestionable, or F$@kedup to me. I don’t do nuance.”
It was in the humanities, needless to say. Very popular prof too. 😉