Is Grade Inflation Real?

If the productivity of college graduates has risen, that would be evidence that college graduates are now capable of better work.

Of course, you would need some way to get a consistent estimate of the productivity gain caused by going to college (then and now). This might not be easy.

If pre-college education is better now, then even if the gain from going to college is unchanged, there would be rising grades without grade inflation.

One way to check this relatively carefully would be to look at subjects like math and (theoretical) physics where there can (a) be fairly similar tests across the years and (b) a way to compare whether the same number of right and wrong answers is correlated with the same letter grade as before. The results of an experiment like that would be interesting.

I would look at the relative grades in different subjects. I think that average grades in the humanities have gone up more than average grades in the sciences; that isn’t the pattern I would expect if the difference is better preparation or harder-working students. In my observation, students who are looking to coast are far more likely to avoid the sciences, and high school preparation in the sciences and especially in math has gotten much better over the last 30 years, while English preparation has if anything gotten worse.

This is a tangent, but:

In my observation, students who are looking to coast are far more likely to avoid the sciences.

That’s certainly true, but doesn’t mean such students end up in the humantities. My imrpession is that at better schools slacker students end up disproportionately concentrated in economics and psychology, while at worse schools they’re getting bachelors degrees in things like communications and business administration.

bb thinks we need a time machine, but that’s not the only way to compare past and present. We can’t look at yesterday’s students in today’s classes, but we could look at today’s students on yesterday’s exams, i.e., use experimental rather than historical methods.

Plenty of university archives will have syllabi and exams from 20, 30, 50, a hundred years ago. Give your predecessor’s exam and compare your students’ answers to their predecessors.

Not a perfect method, but it could give some nice samples to compare.

Hmm – a few thoughts:

1. To the extent that professors are grading students compared to their contemporary cohort, shouldn’t some of that already be factored out? If B (or C, or whatever) is supposed to be “average,” then it doesn’t really matter that a B student in 2005 performs better than a B student in 1965, because that’s not the point of comparison. The question is whether, in any given cohort, professors have raised their standard for what letter grade they assign an average student.

2. Individual professors certainly have anecdotes to that effect–ie, of raising the grade they assign to what they consider average work. Others tell stories of assigning higher grades to the same quality work today than they would have in the past. That evidence is more slippery to intepret, but cumulatively does seem important.

3. Your theory about the improving quality of students would also need to take into account the changing slopes of the “mean grade over time” line. It rose sharply in the late 60s-early 70s; roughly plateaued for more than 10 yrs; and then began rising, gradually but steadily again, in the late 80s. At least some of those changes seem more consistent with changes in grading policy than in the underlying quality of students.

I think there are a number of effects here. One effect is that at the best schools, the students are better. Yale when Bush went there didn’t accept women, now it does. The students almost have to be better if there are now twice as many applying. It’s also true that at the best schools, the size of the incoming class has not increased with population and increasing access to higher education. As we widen the pool of who gets to go to college, this means that the competition is fiercer.

It’s also true that academic scholarships and legacies throw a wrench in this analysis. After all, it’s clear GW Bush didn’t get into Yale on merit, and I doubt his daughter did either. Anyone have any statistics on whether legacies have increased or decreased as a portion of the admitting classes?

The analysis gets even weirder as you move down the academic food chain.

The two comments of Jim Harrison and Tim are related in a quirky sort of way. Sure, there is more biology (or any science) to learn than in the past. But Tim’s suggestion that we experimentally enquire of students skills using 20-50 year old tests misses the nature of developing fields and the sloughing of old skills. In my discipline (chemistry and biochemistry) students today would fail most qualitative tests for chemical analysis – we have machines that do it (better?) today. By learning a new technology the old becomes redundant. Sadly, with the loss of some of these skills, so too goes critical thinking. Not always, but often.

This isn’t directly on topic, but I think it’s still relevant.

When I was employed in academic administration at a large public university system a few years back, I recall working on a project which established that, net of just about every individual and classroom level control you can think of, being in a class taught by an adjunct–instead of a tenure tracker–raised a student’s grade considerably. (This report was for internal purposes and therefore wasn’t published, but I’m sure someone must have done this is the published lit.)

It seems pretty plausible to me to hypothesize that this “adjunct effect” holds more generally–whether you like to look at the world in terms of incentives, norms, or what have you.

We do know that the use of adjuncts and assorted part-timers has exploded over the last 3 decades.

So, I agree entirely that we don’t have compelling evidence of rampant grade inflation. I guess I’m proposing a mechanism for a phenomenon I can’t document.

It’s hard to believe that there is no variation even in a highly select group of students. If students are entering classes better prepared they need to be challenged to a higher standard.

Resolution of GPA between students is not an issue, for the same reason that all students at the 99.99% and above still represent a very large variation, but grade compression near one end in a particular course may indicatae that the brighter students are not being challenged enough.
Evidence for grade inflation may be suspect, especially as people like Harvey Mansfield use it as evidence of affirmative grading.

Re #9: Many universities will also have scattered students papers (especially from the humanities) in their archives, or in the papers of their professors. One could have modern instructors grade these old papers interspersed with new ones (blindly, of course) and see what happened. Advances in the field, of course, would cause problems, but I’m sure you could write a decent methods section that would explain it away.

In sports, no one blinks an eye at the fact that records are continually broken and what was before exceptional is now commonplace. If athletes get better with passing time, why does it surprise us that students should?

There is some grade inflation, just sitting in exam boards watching people work out to recalculate grades to produce higher marks tells me that, but in addition, the recruitment pool has changed. My partner entered Oxford in 1965 with grades that wouldn’t get you into a Poly now. He could do that because only five per cent of the population applied.

The problem is that the rising grades are observed within institutions; and as john and sebastian suggest they tend to correlate with rising SAT scores. The humanities/natural sciences split may be explained by two things: the fact that a number of natural science courses act as gatekeepers to lucrative professions, and… (now I’m going to offend some people) that quality of instruction in the natural sciences and math in HIGH SCHOOL has declined relative to instruction in the humanities. In the UK, where I know the high school situation better, all the experts think that the latter has happened (and that it will get worse). So prior achievement, the best predictor of grades in a course, has risen less in the sciences/math than in the humanities. Though, how you control for changed knowledge base, goodness knows.

I see no reason to assume constant achievement. We know, for example, that adult illiteracy rates have fallen dramatically in the past 200 years. Why assume constant achievement?

“that quality of instruction in the natural sciences and math in HIGH SCHOOL has declined relative to instruction in the humanities”

That it has decreased in the sciences and math, I have little doubt. On the other hand, I know of no reason why it has improved in the humanities.

Economics as a slacker course. Erk! At least in the economics courses I took – and in those my son is currently taking – you often needed advanced calculus, statistics, etc. It sometimes seemed the field was submerged in math.

When I got my degree twenty-plus years ago at a state university I had occasion to compare notes with friends attending private colleges. They definitely got better grades for the same work. Likewise, bell curves were more often utilized in my classes instead of grading based on an absolute mark on the test. This still seems pretty common in my son’s classes. I’m skeptical that grade inflation is so much of a problem in public universities.

My son was also much better prepared in high-school – both in sciences and humanities – for university than was I. It seems common for public high-schools in middle-class communities to have AP courses of all sorts.

Don N.

Being currently enrolled in an Ivy Leauge school, I should have something to say on this topic. I don’t really have a posistion so i’ll just put forth my observations.

the classes I have taken where i feel the grades were easiest were two philsophy classes (200 level which is the lowest because there is only one 100 level) but on the other hand alot of people in those classes seemed very confused so they wern’t all that skewed.

In my intro physics classes they intentionaly set the mean about a half grade to a grade higher because they were the honors sequence and did not want to deter people from taking the honors classes for the reasons of grades (there was a complicated process which involved control questions which were on both the honors and normal exams and such)

the math classes here do not have much inflation

the social sciences and the business people (particualry the hotel school) are thought of as much easier by the hard science people and engineers.

In chiense there wasn’t really inflation in terms of shifting the middle forward but they did pull the bottom up to the middle (i think this was to let you give up gracefully, they take it rather seriously, not designed for the casually interested student at all)(I also think this was inpart to make the course fairer for people who did not gorw up around chinese or already spoke one dialect (i took mandrian, alot of the class already spoke canto))

My father is also a physist and he says the stuff I am doing now is similar to what he did towards the end his undergrad (i have finised 4 semesters) and when i graduate i will be about where he was when they gave him his masters in terms of course work.

There is also a question of how to compare grades between institutions. Our TA this semster commented that he didn’t see anything on the level of what we were doing as sophmores until he got to grad school. (he went to UC school, i think, he wasn’t very clear on that), so does that mean a c here is better than an a someplace else? If you let the schools decide them selves then you get grade inflation at top tier schools, or do they grade harshly and hope that employers and grad schools will take in to account where they are from.

how do you deal with brown where you can take every course pass/fail?

as for the claims that high school level math/science is going down, it depends on where you look. If you look at math/science magnet schools there is no way you can say it is going down. even in my normal public school i took math up through calc and college level chem.
Not that i think that even matters past freshmen year because you are doing material you didn’t even really know existed in high school. maybe this effect could be seen in courses for non-majors, but not in major. show me a time when it was common to cover anything past algebraic newtonain physics (it doesn’t get any easier), maybe some calc (i would be really nice if they brought back the draconian trig and geometry classes though, one of our profs was picking on us for not knowing them and they would have been useful to know before i needed them) or the amount of chemistry that is covered in a semester of college chem?

watching SAT scores is a silly thing to do. They track nothing but your ability to memorize words and lean questioning patterns. now this may be a useful skill in some fields i would not conflate it with demonstrating intelegence, critical thinking ability, or better work over all. this might be constant with an increase in grades due to an increase in people who are good at identifying what the profesors want and giving that answer to them and are willing to do absurd things (like just memorize 600 words) for grades.

i’m going to stop typing now because i am ranting, and most likly starting to offend people

One piece of anecdotal evidence: A couple of years ago I taught some revision courses for ‘A’-level students (high school students) in philosophy and, in the process of setting sample essay questions, had occasion to read through about 20 or more years of past exam papers.

I definitely didn’t get the sense that they were any easier now than they were 20 years ago — although it was the case that the subject matter looked to have changed a little. The later papers had rather less strictly historical material and rather more broad ‘conceptual’ questions (not necessarily a bad thing, either).

The fact that exam papers are broadly comparable doesn’t necessarily mean that grade inflation isn’t going on, as the mark required to achieve an A on the exam may be lower, but the papers themselves certainly didn’t seem to be dumbed down.

[the] quality of instruction in the natural sciences and math in HIGH SCHOOL has declined relative to instruction in the humanities.

I have to take issue with that one, I think. That certainly wasn’t my experience. As someone who always did much better in humanities than natural sciences, I found high school to be incredibly frustrating. For those who had an aptitude in math or science, there was a great deal of room for advancement and a challenging curriculum. If your aptitude lay in the humanities, there was really nowhere for you to go. Even at the AP level, I found the humanities courses to be completely unchallenging. The school just didn’t seem interested in developing that program, though they were excellent at turning out math and science whizzes.
There’s no way to extrapolate from one high school’s curricula, of course, but I found my experience borne out at college. I worked in the school’s writing center as an undergraduate and was absolutely amazed but what I saw. Students would come in for help for help with papers (particularly humanities papers) with no idea about even the most basic principles. Keep in mind these were people in bioengineering, materials science and pre-med programs (in which our university consistently ranks in the top ten, often top five). But they had no idea where to start when assigned a paper about a novel. It seemed odd that they had never been taught how to argue a point in writing or how to structure a paper. Similarly, basic familiarity with literary criticism, the history of literature, etc. was completely lacking, despite an amazing aptitude in the natural sciences. In most cases, I don’t believe it was any fault on their part; they were quite frustrated by the experience but had simply not been exposed to this stuff before. And it wasn’t just people in math and the hard sciences: humanities, econ and psych majors had the same difficulties. Admittedly, the experience one would have working in a university writing center is biased towards any student with this problem, but it also seemed to be the case in all of the lower-level humanities courses I took.

Anecdote: my wife and I both received our first degrees in South Africa – Math/Philosophy/Comp Sci in my case, English/Comp Sci in my wife’s. We’ve taken a number of courses at USA colleges/universities, mostly Comp Sci but some psychology, and it’s notable that work for which we expect C or B grades gets A or better. I know I’m not putting my best efforts into the coursework: my best efforts go into my job and raising kids: so it’s almost demoralizing to get these grades. The student as consumer ?

Another anecdotal data point: I attended Hightly Regarded Big State for my bachelor’s degree in engineering, graduating in 1980. I’m doing a radical change of career, gone back to school to study geology at Not-Quite-So-Highly Regarded State, so of course I had to take all the upper division geology classes.

In the ’70s I was a B/B+ student. Now I’m a straight-A student (A+ isn’t given) in the undergraduate courses. I get much the same score on exams as I did as an undergraduate, maybe a little higher. I certainly get much better grades on writing assignments, but I’m also a much better writer after a couple of decades of practice. Based on my own evaluation of my work, in 1980 at HRBS I would have been a B+/A- student.

So, is Highly Regarded Big State that much better than my current school? Or is it grade inflation? Perhaps a little of both.

One of my pet conjectures is that the growing quantity of AP credit that students have results in slightly higher grades, since those students are often able to avoid difficult classes (Calc 2 is the most failed class at my college, and a lot of people simply place out of it), or to take fewer classes while still graduating on time. Similarly, there are websites that provide information on the grading distributions of different professors or classes. Finally I’ve heard anecdotal evidence that dropping courses was much rarer in my parents’ generation (college in the late ’60s).

So perhaps some of the observed changes in grade distribution can be explained by the fact that students these days are much more concerned (conniving?) about their grades and have quite a few more tools they can use to make sure that they get good grades. And none of this requires any major changes in grading policies. (Even if these are real factors, I can’t imagine they could come close to explaining historical differences in the grades students have received, I just want to point out that the typical inference from “higher average grades” to “oh no the professors have ruined everything” is far too hasty).

“I’m sure the students are better, but why should that affect the distribution of grades? The distribution should stay the same.”

No, they shouldn’t. GPA is an extremely important metric for grad school applications; why should any elite school want to penalize their students? Podunk U has 5 outstanding students, 30 good students, and 25 average students. Elite U has 50 outstanding students and 10 average students. Why should a student at Elite U be penalized for being in the bottom half of its outstanding students by receiving a C (or even a B) when his work is almost certainly superior to most students at Podunk U?

As for math instruction declining over the years, that’s almost certainly not true. All the college admissions tests reflect increasing math scores over the years, and verbal scores are consistently behind. English Comp AP classes have a far lower reputation than AP Calculus and AP Stats (although both of these courses are criticized for other reasons).

While everyone focuses on college grade inflation, I see a real problem with grade *deflation* in the top level high schools. These schools want good test scores and AP tests, have great teachers that they can pay for, and the best way to attract parents is to tout the great scores AND the tough grading.

As a result, the elite rank of juniors and seniors are working at courses that are often harder than they’ll ever see in college, and getting Bs for work of a difficulty and quality that 80-90% of high school students will never approach. While colleges give some weight to harder courses, they usually limit the weight. They also use AP grades but not AP scores in admissions. As a result, students who receive a B+ in an AP course and a 5 on the test (a very common situation) could very conceivably lose out to students who get an A in an AP course and a 1 on the test (or who didn’t bother to take it at all).

The best way to address this for highschool students would be to use SAT Subject and AP test scores as a proxy for grades–or at least as a balancing mechanism, adding and subtracting from the GPA based on scores. And the only reasonable way to address college grade inflation would likewise include some standardized solution.

Barring that, there’s no real solution. Given that the best schools are turning out high quality students, I’m not sure there’s a real need for a college solution.