Grade Inflation

by Harry on August 18, 2008

You might want to check out my colleague Lester Hunt’s excellent new edited collection on Grade Inflation: Academic Standards in Higher Education, which is just out. It originated in a rather well-thought-out conference Lester organized back in 2004. My own contribution arose because he asked me to comment on Valen Johnson’s talk, based on his book Grade Inflation: A Crisis in College Education, and then sneakily inveigled me to contribute a self-standing chapter. The collection is great: genuinely diverse and thoughtful contributions from Clifford Adelman, David T. Beito, Mary Biggs, Richard Kamber, Alfie Kohn, Charles W. Nuckolls, Francis K. Schrag, John D. Wiley and Lester and me. Recommend it to your library, and to your Deans!

In the course of writing my own paper several things happened. I started off assuming (with no real evidence) that grade inflation was real and believing (for no real reasons) that it was bad; I discovered that there is no evidence of grade inflation (which doesn’t, of course, mean that it doesn’t exist) and that the reasons for thinking it would be bad if it did exist are pretty weak. Commenting on Johnson’s book, in other words, convinced me that his subtitle is entirely wrong (even though the book is, actually, terrifically good). It’s not the first time that I have changed my mind as the result of writing a paper, but it is the first that I’ve changed it quite so radically.

I developed, mainly through reading Valen Johnson’s book, a conviction that student evaluations are next to worthless for evaluating teachers. His book also convinced me that grade variation within departments exists and is bad, though not that there is much we can or should do about it.. Finally, I became more and more irritated with Harvey Mansfield’s piece in the Chronicle. So, below the fold, here’s a taster of the book, adapted from my chapter, and arguing specifically against Mansfield:

In 2001 Harvey Mansfield published an article in the Chronicle of Higher Education criticizing his colleagues at elite universities for giving their students high grades. To be fair he targets himself with the same criticism; he believes that he, too, participates in this bad practice, which he dubs, in the title of his piece “Grade Inflation”. This might expose him to charges of hypocrisy. But he is not necessarily wrong to give out higher grades than he thinks are generally proper. Because the assignment of grades has psychological and emotional effects on students, and those effects are influenced by the prevailing culture of grading, we should all, to some extent, make our grading practices somewhat sensitive to those that prevail around us. In my first few years of teaching I did the same as Mansfield did, and we were both right to do so.

Mansfield says that grade inflation exists, and is a serious problem. But he gives no evidence for it existing, and his reasons for thinking it a problem are poor. Let’s start with his reasons for opposing it:


Grade inflation compresses all grades at the top, making it difficult to discriminate the best from the very good, the very good from the good, the good from the mediocre. Surely a teacher wants to mark the few best students with a grade that distinguishes them from all the rest in the top quarter, but at Harvard that’s not possible.

Why is it so important to hive off a handful of students for special recognition, if the whole top quarter is doing work that is very good? In nearly 20 years of teaching in research universities I regularly—in just about every class—come across students who are smarter than I am and more promising than I was at their age, but there have only been 4 or 5 students whose work placed them unambiguously well above the rest of the top quarter, and only one whose work stunned me. Reserving an A (or A+ or A++) for them takes grades too seriously. How could the one stunning student know that he was being rewarded with a stunning grade? And why should he care? (The student in question, who is a regular CT reader, I know, would have found the very idea of reserving a grade for him absurd, laughable, arrogant, and vain). A professor can reward, or ‘mark’, those students’ work much more effectively with verbal or written praise, or with a request to meet to discuss the paper, or with frank admiration of a thought in the public forum of the classroom. Only a student unhealthily obsessed with their grades would be more motivated by a special grade than by alternative forms of recognition. I have not yet come across a student whose work is extremely good and who is sufficiently grade-obsessed that adding a reserved high grade would motivate or reward them at all in the presence of any of the alternatives I have mentioned.

There’s something else going on here. Mansfield reveals that it is not, in fact, grade inflation that he is bothered by:

Some say that Harvard students are better these days and deserve higher grades. But if they are in some measures better, the proper response is to raise our standards and demand more of our students. Cars are better-made now than they used to be. So when buying a car would you be satisfied with one that was as good as they used to be?

Mansfield, of course, does not think that the students are better these days. But his response is interesting because it specifies circumstances in which he thinks that grade deflation ought to occur. Given any starting point, he thinks that we should make high grades harder to attain as the students improve. He does not, though, seem to think that standards should be lowered as the students get worse. In other words, it is not grade inflation, but the very existence of nominally high grades, however good the work, that he objects to.

It is hard to come up with a rationale for this asymmetry. For example, a single institution’s unilaterally raising standards for the allocation of grades does nothing to help grades inform prospective employers or students themselves. It might help with the pedagogical function of grades – the way that grading prompts a student to want to learn more, or triggers the self-confidence that helps them to go deeper into a subject – but it might not; it will all depend on the students themselves and the culture they inhabit. In fact, Mansfield seems unhealthily obsessed with his own function as evaluator of students. He want to feel that he is tough and hard headed, just as we all do, and that is certainly valuable for us. But presumably the point of a grading system is not to make faculty feel good.

Ok, onto the lack of evidence for grade inflation. Mansfield offers none. He does offer anecdotes about how it arose, but nothing more. When he first raised the issue of grade inflation, he said that

when grade inflation got started, in the late 60’s and early 70’s, white professors, imbibing the spirit of affirmative action, stopped giving low or average grades to black students and, to justify or conceal it stopped giving those grades to white students as well

He excuses his lack of evidence for grade inflation and its initial triggers by pleading lack of access to the facts and figures. But his comments about what he lacks access to reveals, worryingly, that he doesn’t understand what would constitute evidence of grade inflation. This is what he says:

Because I have no access to the figures, I have to rely on what I saw and heard at the time. Although it is not so now, it was then utterly commonplace for white professors to overgrade black students…..Of course, it is better to have facts and figures when one speaks, but I am not going to be silenced by people [referring to a dean and the then-president, of Harvard] who have them, but refuse to make them available.

The problem here is that the figures he calls for tell us nothing about whether there is grade inflation, and that nobody in the administration has the facts that would tell us (or, if they do, Harvard is a more remarkable than even its greatest enthusiasts believe). Regular inflation isn’t just an increase in price, but an increase in price relative to the quality of the goods on offer: for the analogy to hold grade inflation occurs when, over time, mean grades increase faster than the mean quality of work produced by students, or mean grades decrease more slowly than the mean quality of work produced by students.

Grades probably have increased at Harvard, and if so Harvard is not alone. Stuart Rojstaczer has compiled data from selections of private and public Universities, showing mean GPAs increased by .15 in both groups between 1991 and 2001. Using data from the same universities he finds an increase in mean GPA of .68 in private universities and .5 in public universities from 1967 to 2002. But this is simply no evidence at all of grade inflation. Why?

Suppose that the mean grade has increased dramatically during the time in which Mansfield has taught at Harvard. Does this mean that students are now being given higher grades for the same quality of work, or the same grades for lower quality work, than before? No. The improved grades may reflect improved work. Perhaps Harvard now hires better teachers than it did when it hired Mansfield. Perhaps the students are more talented, or harder working, or better prepared on entry, or all three. If the first of those possibilities sounds unlikely think again: it is hard to imagine even a legacy student as weak as now-President George W Bush was at the time gaining admission to Harvard or Yale or any other elite college today, as he did (to Yale, admittedly) in 1964. In the period we are discussing the mean number of children born into elite families has declined, enabling those families to invest more in each of child; Harvard has also expanded dramatically the talent pool from which it draws, by admitting women (although the first woman was admitted in 1950, women were admitted on an equal basis with men only in 1973), and hiring professors in a much more open labour market than when Mansfield was hired; and more of those women have been socialized to be ambitious in academic and career terms over that time. The “Gentleman’s C” which both the 2004 Presidential candidates were awarded by Yale in the 1960’s is reputedly a thing of the past, and so are the gentlemen at Harvard and Yale who were awarded them.

How could we know whether there was grade inflation? To know it would require a large year-by-year database of the actual work done by Harvard students (including, presumably, evidence of their classroom participation), and the matching grades. I’d be very surprised if the administrators have such a database, access to which they are jealously guarding. Mansfield ought to know: if one existed he would have been asked to contribute to it. But he does not even seem aware that that is what would be needed.

So, onto my final quibble. Mansfield attributes putative grade inflation to the response of professors to affirmative action and the self-esteem movement. Just as he offers no evidence for the existence of grade inflation, he offers no evidence for attributing it to these factors; it appears to be mere prejudice. Quite apart from that fact that if we include legacies, athletes, and the male students who would not have been admitted if they had competed on an equal basis with women as affirmative action admissions, as we should, Harvard has in fact reduced its reliance on affirmative action in admissions over the period Mansfield has worked there, and was doing so pretty dramatically in the period when he thinks grade inflation began.

{ 114 comments }

1

Tom T. 08.18.08 at 2:51 am

One data point, albeit not from Harvard: My law school grades to an average, and that average was raised from B to B+ in the last few years.

2

John Quiggin 08.18.08 at 3:26 am

You haven’t spelt out why you think grade inflation wouldn’t be bad. I’ll toss up one of the standard reasons, and then give my own answer
(1) Unfair to earlier cohorts of students. This is most obvious where the inflation consists of awarding a qualification (for example, at the Masters level) for diminishing levels of achievement, as has certainly happened in Australia. The Masters degree I hold would now be called “Masters Honours” to distinguish it from the much easier course now given for a Masters. Fortunately, I don’t rely heavily on the Masters in job-hunting, and maybe its true that the half-life of grades is short in relation to the inflation rate.

My own answer, given in a post a while back,
http://crookedtimber.org/2008/06/29/marks/
is that the whole grading business is one of the more dubious aspects of the university system, and that I’m fairly relaxed about it being undermined.

3

Caleb D'Anvers 08.18.08 at 3:52 am

Another data point:

In the early ’90s, an American exchange student arrived in my 7th form (read: high school senior class) in Auckland, New Zealand. He was appalled when he received his first New Zealand grade — 74% on an art history essay. ‘That’s, like, a fail‘, he said, and of course, it was, from his perspective. He told the rest of us that, if he submitted the same essay in Kansas, he would expect at the very least a 90. Yet that 74% placed him comfortably in the top third of the class.

None of us ever regularly got 90s, even though we were the top class in the top state secondary school in the country. 88%, say, was considered a rare achievement, and because that was the horizon of our expectations, we accepted it.

So what’s behind the discrepancy between that NZ 74 and its 90+ equivalent in Clayton, KS? I very much doubt that it reflects any real difference in student aptitude or achievement. Ergo: US high school and college grades are grossly inflated, when compared to those earned in other First World countries.

4

Righteous Bubba 08.18.08 at 4:02 am

Ergo: US high school and college grades are grossly inflated, when compared to those earned in other First World countries.

No, it means the numbers are used differently, not that they are inflated.

Filipino schools, for instance, use a grade scale in which everything under 75% is a fail. Their transcripts look pretty good for a sec until you remember.

5

arthur 08.18.08 at 4:36 am

I think grade inflation is bad because it lowers the variance of scores and increases the effect of randomness. A loss of 10 marks because of an arithmetic error/preparing for the wrong question is going to count for less if the marks are distributed over 40 – 90, but it will probably put someone in the bottom of the class if the marks are distributed over 70 – 95. As such, grades will be less informative after taking into account of “random shocks” such as these.

6

Walt 08.18.08 at 6:20 am

Like most Americans, I only turn to Harvey Mansfield for advice on how to be manly, not for advice on how to grade.

7

Z 08.18.08 at 6:45 am

Grade inflation has always seemed to me to be the standard “times ain’t like they used to be” complaint, academia style. According to most objectives measure I know (unchanged standardized tests administered to generation of students for instance), the level of students is improving. However, some perfectly careful specialized studies do show some grade inflation from times to times: a famous example in France would be the way spelling is graded.

Based on that example, I would say that grade inflation can be detrimental in the sense that it shows a probable discrepancy between the demands of society and that of the school system. This discrepancy, as any other non explicit rules, tend to give an unfair advantage to some segments of the society (those who know that having a perfect spelling will ensure their CV will be accepted) on others (those who think that their A in French will be enough to convince a recruiter) .

8

Martin Wisse 08.18.08 at 7:17 am


when grade inflation got started, in the late 60’s and early 70’s, white professors, imbibing the spirit of affirmative action, stopped giving low or average grades to black students and, to justify or conceal it stopped giving those grades to white students as well

And here we see that once again, racism is hid behind a veneer of reactionary, “common sense” drivel.

9

Martin Wisse 08.18.08 at 7:20 am

Btw Harry, next time it would be nice if you made it clear from the start that you’re talking in an US educational context…

10

Nancy Lebovitz 08.18.08 at 7:46 am

Just checking on something I’d heard, which was that there was grade inflation during the Viet Nam war to keep young men from being drafted. Anyone know if that was happening?

11

agm 08.18.08 at 7:55 am

I’d quibble with the details such as legacies at the elite US schools, but overall, you sound as if you’ve had quite a run of unique students. Hopefully that will keep retarding the jaded view it is so easy to develop.

A professor can reward, or ‘mark’, those students’ work much more effectively with verbal or written praise, or with a request to meet to discuss the paper, or with frank admiration of a thought in the public forum of the classroom. Only a student unhealthily obsessed with their grades would be more motivated by a special grade than by alternative forms of recognition. I have not yet come across a student whose work is extremely good and who is sufficiently grade-obsessed that adding a reserved high grade would motivate or reward them at all in the presence of any of the alternatives I have mentioned.

You’re f’ing kidding me, right? Just how often do you deal with pre-meds… Shortest term, yeah, it’s nice but all they care about is whether this or that grade will hurt their final grade. On a semester time scale they care about whether the grade they get for the class is going to hurt their shot at getting in. Long-term they don’t give a damn about your praise unless it’s in a letter designed to outshine all the other people’s letters, even after taking into account the fact that there will be lots of other letters written to outshine the letters of the competition.

And in a technical field, it’s just not that hard to justify whether or not someone deserved a grade higher than someone else. Student A finishes problem sets because they did the work to develop the mental framework needed to understand what the homework asks them to do. Student B only cracks the book to hunt for a plug-and-chug example and therefore can’t always figure which numbers to put in which parts of which equations. Student C doesn’t even attempt every problem in an assignment. Student D is just trying to squeek through and check off a class on the degree plan. E never bothered to show up, while F skips exams in addition to class and homework assignments. Negative reinforcement does worlds of damage, but positive reinforcement of the sort you like doesn’t serve the purpose you think it does. And that’s before you get to the fact that you may be violating all sorts of privacy laws if aren’t careful enough in discussing someone’s work in class.

12

Stephen Kinsella 08.18.08 at 8:00 am

As a product of both an Irish and a US grad school program, I found the differences in grading policies to be only skin deep. What mattered was where you ended up in the stack, not the index of the stack itself. So, in Ireland, if a goodly proportion of the class gets a 2.1 (second class honours, first class) and a few receive a first, then it’s all good, you can determine the high flyers from the merely good. Similarly, in the US, if everyone got an A minus, and a select few got an A, then it was also all good. No one in my graduate class got less than a B+ in the entire time I was there, but it didn’t make any difference really. For me, the role of a grade is distributional ranking. The units are irrelevant, unless you think a B+ from the US correlates to a 2.1 in Ireland. Which everyone, after your book has been published, knows isn’t the case. So once there is a standard enough translation of grades (from Oxford to Trinity College Dublin to Harvard to Florida State, say), then we’re all good.

13

Steve P 08.18.08 at 8:10 am

I am a business professor who has taught in Australia, the US, Singapore, Hong Kong, Spain and Italy. I can confirm the NZ observation that US grades are uniformly higher than Australia and New Zealand. So an average class grade in Australia would be 65-70% and in the US about 85% for similar work.

Does this matter? I think it does on one point that the author of the piece almost acknowledges. In Australia, an A is a “stretch” grade that causes top students to strive to go beyond simple mastery of the material.

US students expect an A if they are able to master (or regurgitate) the material. This gives the impression that mastery is the highest level of learning. However, Bloom’s taxonomy of knowledge (see http://www.esf.edu/erfeg/endreny/courses/LevelsofKnowledge.htm) places comprehension at Level 2. To earn an A in an Australian university would require Level 5 or 6 knowledge.

This is not to say that US students are incapable of developing Level 5 or 6 knowledge. They just don’t have to develop it because they don’t have to – there is no incentive. My typical grade distribution is 1/3 As, 1/3Bs and 1/3Cs, which would correspond to Level 3, 2 and 1 (I just finished a bunch of grading and basically applied this principle). MBAs students expect 50% As and 50% Bs – making fine distinctions even more problematic.

So, bottom line, the thing that is lost to grade inflation is the incentive to develop Level 5 and 6 knowledge.

14

Alex 08.18.08 at 8:26 am

I feel the time has come to take a stand on the pressing issue of Olympic inflation. Throughout the history of organised sport, indicators of so-called achievement have been rising steadily. Clearly, our standards are slipping. For example; the first known marathoner actually died immediately after the race. For many years subsequently, there is no good data, but when accurate record keeping resumed in the late 19th century, the trend towards marathon inflation rapidly becomes evident. The damning evidence: this trend is visible across a whole range of sports. Surely, if today’s runners were in any sense “better”, this would be a factor specific to the marathon? Clearly, it’s past time that more miles were added to the marathon in order to restore standards.

But the phenomenon is visible across the whole of our declining western society. For example, even fundamental economic data is affected. Average output per hour worked, to take the most sinister example, has “risen” with astonishing speed since some sort of triggering event in the 18th century – no doubt part of the intellectual so-called enlightenment brought about by the Red Terror. Clearly, work is becoming easier.

As a result, I propose that we introduce measures to make it tougher. Manual workers could carry a weighted belt. Computer programmers could have a simple application installed on their desktops that constantly opens arbitrary dialog boxes in order to test their concentration….

/snark. Seriously, who imagined that a declinist narrative might turn out to be a bunch of reactionary bullshit?

15

Z 08.18.08 at 8:55 am

The damning evidence: this trend is visible across a whole range of sports.

And as in the case of academia, it all started when affirmative action forced organizers to accept black athletes to competitions. Starting in Berlin 1936, judges, imbibing the spirit of affirmative action, took the habit of giving away gold medals to black athletes.

16

sanbikinoraion 08.18.08 at 9:33 am

#10 is quite beautiful :)

17

Greatzamfir 08.18.08 at 9:37 am

Computer programmers could have a simple application installed on their desktops that constantly opens arbitrary dialog boxes in order to test their concentration

Like MSN?

18

Tracy W 08.18.08 at 9:38 am

In the early ‘90s, an American exchange student arrived in my 7th form (read: high school senior class) in Auckland, New Zealand. He was appalled when he received his first New Zealand grade—74% on an art history essay.

I understand USA marking practices have a higher starting point. An “A” is anything above 90, while in my NZ high school an “A” was anything above 80. With the exception of the Bursary exams, where an “A” was above 65%. I never figured out the reason for that difference.

Personally it seems silly to me to not use the full range that the marking system allows.

None of us ever regularly got 90s, even though we were the top class in the top state secondary school in the country.

Ah come on, how can an Auckland school be top? :)

19

Alex 08.18.08 at 10:24 am

It’s an interesting consequence of the letter grading system that the underlying marks cluster around significant boundaries. Further, the notion of “passing” has been meaningless for years because grades below C are fail grades for practical purposes (and in fact, the government’s statistics recognise this with the concept of X per cent grades A to C). Also, the bands are narrower between C and A than anywhere else. 55 – mediocre, 70 – excellent!

Personally, I’d go for just using the numerical marks, which would provide the finer grained measurement exam whingers always say they want. It would also, I think, help to anchor the system psychologically.

20

Peter 08.18.08 at 11:11 am

Another underlying justification for the dangers of grade inflation, particularly about the compression at the top, is that a main reason for grades is not for students, but for external actors (work, professional and graduate schools) to be able to reliably evaluate students. I would suggest that this is the source of lots of hand-wringing about grade increases irregardless of the grade-to-work-quality ratio. Professors and colleges are under some external, not often acknowledged, pressure to sort students for the workforce and beyond.

This is something to raise and then dismiss as well, though I would be interested in varied rationales for dismissing it.

21

Dave 08.18.08 at 11:18 am

The comparison with athletics is interesting, because it goes to the heart of what is being discussed – viz. what is the point of this?

Athletes have improved, mostly, because coaching is much better than it used to be, ‘natural ability’ is now only the foundation for scientifically-sculpted musculature and nutrition, every microsecond of a sprinter’s or swimmer’s performance can be video-analysed and used to micro-manage improvement.

This is all fine, because the goal is simple: faster, higher, stronger. And the stopwatch and the tape-measure don’t lie.

By analogy, we may say that students are now much better ‘coached’. This is certainly the case with UK school students, who almost never fail anything any more. And this is pretty much what they expect to happen in HE. Disinclination to tell students exactly [and I mean *exactly*] what to do is a route to disaffection, alas.

Now, the problem is, while the goal of sprinting may be to get across the line faster, the goal of higher education, at least in the humanities, used to be to make a person a better, more rounded individual. Good luck with measuring that, of course; and so it has largely been replaced with the goal of measurable improvement in achievement. Or “grades”.

Thus, if you think students are like athletes, you can be happy grades keep going up [or is it staying up?], but that’s only one kind of analogy. Suppose, for example, we took the analogy [equally far-fetched] between HE and religious belief. Should we be pleased to note that people were going to church more often, could we “count” their spiritual development by seeing how much money they gave to the church [or how many hours they spent on their knees]? Would anything “measurable” actually tell us about the state of their souls [assuming the existence of such a thing for the sake of the argument]? I suspect both atheists and theologians would be united [for their different reasons] in saying “No”.

Therefore, insofar as “Higher Education” is about more/other than just “grades”, grades are meaningless in discerning the health of the enterprise. Focus on them, indeed, is itself detrimental to that other thing which is the purpose of education.

22

Slocum 08.18.08 at 11:33 am

Suppose that the mean grade has increased dramatically during the time in which Mansfield has taught at Harvard. Does this mean that students are now being given higher grades for the same quality of work, or the same grades for lower quality work, than before?

But that raises the obvious question of whether a grade is supposed to reflect an absolute or relative standard of work.

If grades are relative to the institution, then students and prospective employers alike should be very happy with a ‘C’ average from a Harvard graduate (as, supposedly, they once were back in the day). After all, Harvard admits only those with the highest high-school GPAs and SAT scores, so average for a Harvard grad is still excellent (as in, say, 2-3 standard deviations above the national mean).

But if grades are to be an absolute standard reflecting, say, how the student would be rated if all university students were, hypothetically, graded using standardized national exam like the SATs (and the GREs, LSATs and MCATs to follow), then nearly all Ivy League students really should graduate with identical 4.0 GPAs, just like high-school.

In practice, it’s a mess — with different institutions, departments, and individual professors using different combinations of absolute and relative thinking.

23

Matt McIrvin 08.18.08 at 11:37 am

I’ve heard that ex-military employees of private corporations are often enraged when they get their first performance evaluation, because in the US military, evaluations of officers are considered unsatisfactory (and insufficient for promotion) if the maximum possible value hasn’t been awarded in every category, whereas in the private sector this is usually not the case. So does the military have unconscionable evaluation inflation? I’m not sure I know the answer.

24

Z 08.18.08 at 11:54 am

Matt McIrvin makes a good point. Grading is a social activity, and so involves the grader as well as the graded. Depending on the norm of social interactions between the grader and the graded, it can happen that the grader feels obliged to give the maximum possible value (or else to justify her choice at length). For instance, french aviation controllers were always graded at the maximum possible value until the french state asked the grader to give a certain proportion of grades below the maximum value. What happened is that grader started to distribute uniformly lower grades (if you had a B last time, you get an A, if you had an A, you get a B). The reason is that when french aviation controller are angry, they get on strike an paralyze the country, so everyone is careful to deal with them. Anecdotally, I was once given an undeserved top professional evaluation by my employer in Japan, presumably because he felt he had no authority to grade me (either because I was a foreigner or because I worked there on behalf of a more reputable institution than his).

I wonder if this phenomenon could play a role in american higher education (perhaps because higher tuition fees have modified the relations between grader and graded). It certainly does in french élite institution.

25

Dave 08.18.08 at 12:01 pm

Or, to put it another way, truth and power rarely mix well.

26

Eszter Hargittai 08.18.08 at 12:05 pm

Harry, do you know Jordan Ellenberg (occasional CT reader and commenter), a colleague of yours in the Math Dept? He wrote about this topic – also addressing Mansfield – back in 2002: Don’t Worry About Grade Inflation.

Regarding grading on a curve, I’ve never been a fan. If many of the students taking my class in any given quarter happen to do all the work well then I will grade them all accordingly.

27

engels 08.18.08 at 12:08 pm

A+++++ Great communication, prompt payment, all round excellent eBayer, would deal with again!!!

28

Preachy Preach 08.18.08 at 12:26 pm

My last place graded staff on a rigidly-applied ‘bell’ curve – 50% were to to get a 3 rating, 30% were to get 4, and 10% 5. (The remaining 10% were for flex, and for those unlucky enough to be given the ‘unsatisfactory performance’ gradings of 1 & 2).

Some bright spark had the great idea of enforcing this grade curve by means of roundtable discussions involving the manager population and above, in which everybody’s ratings were fitted onto the mandated distribution. I leave it as an exercise for the reader’s imagination as to what happened when almost all the 4 & 5 ratings available to our department had been given out to the graduates and other junior levels…

29

David Velleman 08.18.08 at 1:21 pm

Harry —

I’d like to hear more about this:

His book also convinced me that grade variation within departments exists and is bad, though not that there is much we can or should do about it.

When I have served on institutional committees that have discussed grade “inflation”, the main concern has been differences among departments and divisions, not the temporal trend that the word ‘inflation’ would suggest. In fact, a tendency to speak in term of a trend may be due to the fact that the disciplines whose grades are highest are also newer disciplines, such as area studies. (Don’t scream at me, readers, until you’ve checked the statistics for your own institutions.)

Disciplinary differences in grading have serious and very undesirable consequences — for example, the shockingly low enrollments in physics and math. Of course, the physicists and mathematicians have themselves to blame (or thank?) for those low enrollments, insofar as they insist on forcing a curve onto their grades — a senseless practice, in my view. But even if they stopped curving their grades, they could never approach the leniency of some disciplines in the social sciences and humanities.

I think that disciplinary differences in grading also account for some of the divisive attitudes among faculty themselves. When serving on inter-disciplinary committees that discuss these matters, I find that the view of the humanities held by scientists — that we are not “real” disciplines, that our courses are a form of “recreation” — is largely based on their perception of the grading differential.

So I like to hear more about why there “isn’t much we can or should do about” this aspect of the problem.

30

Nick 08.18.08 at 2:01 pm

Of course, the physicists and mathematicians have themselves to blame (or thank?) for those low enrollments, insofar as they insist on forcing a curve onto their grades—a senseless practice, in my view.

As I recall from my undergrad math and physics classes, the curve was almost always to the advantage of the students. The average score was set to “C”, but that average was almost always below 75%. So, students usually received a letter grade that was higher than that implied by the percentage of problems they answered correctly.
(That’s assuming that without a curve, F=0-59%, D=60-69%, C= 70-79%, B= 80-89%, A= 90-100%).

Removing the curve would make the grading less lenient. How would that help enrollment?

31

laura 08.18.08 at 2:06 pm

Harry, Why are student evals worthless?

32

Dave 08.18.08 at 2:10 pm

Because people lie. Haven’t you seen House?

33

bicycle Hussein paladin 08.18.08 at 2:13 pm

I’m with #5 on the random noise issue. If all the grades are within a small range of values, random shocks become a bigger deal. If you only need Bs to be seen as doing a great job, the odd C would hopefully be easier to correct for by getting an A somewhere than if you need As and the only way to correct the C is a half dozen more As.

I went to college with a lot of people who would avoid taking any challenging classes because their GPAs were so important to them for med school, the job market, or whatever. Departments like economics seemed to cater to this by giving much easier grades than the other departments. If you were an econ major shooting for a high GPA, you didn’t want to take public policy or anthro classes, nevermind a harder math or statistics class, if you could avoid it, because it would probably hurt your GPA.

34

harry b 08.18.08 at 2:19 pm

John — I agree with what’s in your post, and address the purposes of grading etc extensively in the chapter. Your post was too late for me to cite it, though!

David — I address all that at length in the chapter: others who want to know what I say about it should get the book!

Martin: I’ll figure out a way of tagging the US-specificity of education posts (most of my complaints come from me not tagging the UK-specificity of the UK-specific posts!).

agm — I’ve had grade-obssessed students, I promise! But none who were exceptional intellectually. In the humanities, just as in technical subjects, it is easy to justify different grades and not so differently from in the technical subjects. But reserving a grade for the truly exceptional? (which is what Mansfield wants). If you get a few truly exceptional students in each class, then maybe, but in that case are they really, truly, exceptional? And I baulk at the thought that the truly exceptional student is likely to care that much about having a grade reserved for them. If he does, he should grow up.

Caleb: I discuss the bizarreness of the American culture of grades in the chapter, along your lines. In my day 45% was a C (a passing grade) in O-levels; 65% was an A.

35

harry b 08.18.08 at 2:22 pm

Oh, and Jordan Ellenberg addresses the issues raised in 33 and 5, but only partially (in the article Eszter links to).

And Laura, I’ll put something up about student evals later. They are alarmingly easy to manipulate, and alarmingly unreflective. Valen Johnson’s book basically demolishes them. Welcome back from vacation!

36

Cougian 08.18.08 at 2:34 pm

My problem with grade inflation is not knowing where I stand. I just got a A- in one of my MBA classes. If there were a standard bell curve, I could be fairly sure I was in the top 10-15%. But I have no idea what my grade actually means. But if no one got lower than a B+, I may actually be in the lower half of the class, and many students are out-performing me.

37

mpowell 08.18.08 at 2:41 pm

13 makes an interesting point, but I don’t think it’s particularly relevant to this topic. Where I went to school, getting an A was quite difficult and corresponded to a pretty high level understanding of the material. Still, typically 1/3 of the students in an undergraduate course would get an A. In graduate courses it was 1/2, but graduate students, although sometimes less gifted than the undegrads, were more dedicated to their studies.

I would agree with others that the biggest concerns were probably inter department variations. Physics > Engineering > Business tends to creates certain incentives. I suppose you might consider them to be desirable, but they are probably ad hoc in that case.

38

praisegod barebones 08.18.08 at 3:00 pm

‘I’ll figure out a way of tagging the US-specificity of education posts (most of my complaints come from me not tagging the UK-specificity of the UK-specific posts!).’

Not sure that this one should be tagged as such: having taught philosophy in the UK and Turkey, I found it illuminating wrt to my experience in both countries.

One thought about whehter grade inflation is a bad thing or not (like a lot of people here, I’m inclined to think it’s perniciousness is highly overstated.)

Presumably for some purposes the value of a top grade is a feature of its rarity. But if so, then so is the power/right to give out such grades. So, other things being equal you might expect some academics to regret the disappearance of such rare grades.

Perhaps its overly cynical to think that this is what’s going on with Mansfield; but I’m inclined to think it’s one element in some of these discussions.

39

tps12 08.18.08 at 3:07 pm

Also, to give pre-meds a bit of a break, remember that in the US, they’re competing for a number of med school slots kept artificially much lower than the market would provide were it not for the AMA.

40

lemuel pitkin 08.18.08 at 3:31 pm

Walt (at #6) hits the key point ehre, which is that the proper response to something by Harvey manfield is not to evaluate its content, put to point and laugh derisively.

Personally, I find the whole topic a little depressing. No serious person would spend time seeking or giving out praise in these precise increments.

Harry, what do you think of John Q.’s argument that the only remotely justifiable basis for grades is to motivate students who would otherwise do no work at all (for which a simple pass/fail system would be enough)?

41

lemuel pitkin 08.18.08 at 3:33 pm

I baulk at the thought that the truly exceptional student is likely to care that much about having a grade reserved for them. If he does, he should grow up.

You don’t think this can be generalized beyond the exceptional students?

42

harry b 08.18.08 at 3:47 pm

40. Yes. Good!

39: I discuss the purposes of grading in the chapter. I think that grading has another legitimate purpose, which is providing a kind of simplistic feedback on performance, that enables the student to figure out how much she needs to work on things. Obviously, written and spoken comments are more useful for this purpose than an assigned grade, but the grade can do some work (eg, I sometimes give very critical feedback indiciating that student should do a lot of work on this or that, and then say, something like “I’m only sounding so critical because I think you can really excel; this is already very good indeed by normal standards” — when I do that, having an A or AB presumably gives that encouraging comment a bit more weight).

43

mpowell 08.18.08 at 3:48 pm

38: Yes, it’s true. Not only are the available spots limited, but the selection criteria are excessively anal. And I am convinced that it doesn’t lead to the best possible medical workforce. So fault the system, but that shouldn’t prevent you from making fun of the students who choose to subject themselves to it. After all, only a small portion are really doing it out of a genuine love for the social good of doing medicine.

44

jayann 08.18.08 at 3:50 pm

@14, 21; 21, yes indeed, but also

<IThus, if you think students are like athletes, you can be happy grades keep going up

not if you’re a cyclist from Ulan Bator watching Team GB’s array of lottery-funded coaches and cycles and cycling uniforms and so on (e.g.), and not if you go to a school whose students never get As, and not if you think school children and students and others are being conned by current grades. Indeed, it is worse for you as As become more common but you can’t get one, up to a point, anyway (there’s research on this, way back in The American Soldier, I think). (Yes I know I’ve strayed slightly but I think these considerations are relevant given that most people think grades actually mean something.)

@ Laura; I don’t think student evaluations are “worthless” in the sense that they tell us nothing at all. I do though agree — provisionally — with what Valen Johnson seems to be saying (!); so it’s the use to which evaluations can be put that’s probably the issue. Harry, I look forward to hearing more about Johnson but would also like to know more about the reasons for saying grade inflation hasn’t occurred (because I’m convinced it has).

45

Righteous Bubba 08.18.08 at 3:50 pm

Personally it seems silly to me to not use the full range that the marking system allows.

Downwards also. India and Pakistan often use 33% as their pass grade.

The numbers used are simply different.

46

lemuel pitkin 08.18.08 at 3:53 pm

41: Fair enough. This discussion would be very different (and IMO more productive) if it started from the premise that the purpose of grades is to provide information *to the student*.

47

Righteous Bubba 08.18.08 at 4:21 pm

Thursday was A Level results day in the UK. Grades are higher than ever and the Telegraph is alarmed on behalf of universities all over the UK.

48

geo 08.18.08 at 4:38 pm

Harry@41: Obviously, written and spoken comments are more useful … than an assigned grade

Yes, obviously; but they’re also a hell of a lot more work, and teaching is already substantially undercompensated.

Of course one wants to evaluate students as individually as possible, both for their own sakes and to help those who have responsibility for the distribution of educational opportunities. One wants both to give students what they need and to give society what it needs. If this were all grading were about, then no one would object to any reasonably uniform grading scheme.

But of course, that’s not what grading is about. Because the reward of good work in contemporary is so often money rather than the instrinsic satisfaction of doing good work, and because those money rewards are distributed so unequally against a background of general economic insecurity, grades serve to discipline and socialize students into pathological degrees of selfishness and competitiveness. Make the society more equal and cooperative, and the grading problem will disappear.

49

jayann 08.18.08 at 4:47 pm

@46

I’m no Telegraph reader but the inexorable yearly rise in As at A Level is worrying given some people don’t realise there’s grade inflation, and though you may not want to feel sorry for universities — and why should you? — the Cambridge man has a point.

@ 47

Make the society more equal and cooperative, and the grading problem will disappear.

I almost believe this. It is anyway what I would want to see.

50

christian h. 08.18.08 at 6:13 pm

I want to support what lemuel says: grading should be for the students. By all means grade their work during a course, but use pass/fail as the only reported grades.

51

Dirty Davey 08.18.08 at 6:40 pm

A few things…

First–I think it useful to extend the economic metaphor beyond the term “inflation”. The standard claim of “grade inflation” is equivalent to “workers are being paid more, so there has been inflation”. But wage increases can also result from increased employee productivity, and the opponents of “grade inflation” have yet to make a case that (a) increased student “productivity” does not explain higher grades or (b) that such increased productivity should not result in higher grades.

Second, re:

when grade inflation got started, in the late 60’s and early 70’s, white professors, imbibing the spirit of affirmative action, stopped giving low or average grades to black students

I think the late 1960s saw a reduction in failing grades, but what was far more important than affirmative action was Vietnam and the draft. In earlier times, a student who failed could go off and find a job; at the height of the Vietnam draft, flunking out could mean losing a student deferment and getting shipped off to the war. It’s easy to see a professor of that era believing that even extremely poor student performance should not be punishable with potential death in Southeast Asia, and bestowing a C rather than an F.

@5:

I think grade inflation is bad because it lowers the variance of scores and increases the effect of randomness

One cannot argue this, realistically, unless one is also willing to argue against strict 90-80-70 (or 92-84-76) A-B-C scales for grading. If increasing variance in GPAs (by awarding more Bs and Cs) is a good way to decrease the “effect of randomness”, then so is re-doing test scales and scoring so that the vast majority of students don’t fall into a tiny 25- or 30-point window on the much-larger 100-point scale.

Of course, the anti-grade-inflation folks would shoot steam from their ears at the thought of 80% representing an A-minus or 48% representing a solid C, even if the instructor can honestly claim to have designed the test that way.

52

Righteous Bubba 08.18.08 at 6:57 pm

One cannot argue this, realistically, unless one is also willing to argue against strict 90-80-70 (or 92-84-76) A-B-C scales for grading.

The numbers and letter grades, in my fantasy world, are different things. The former is a raw score, the other is a judgment.

ECTS grades at the Wikipedia. A definition is suggested along with a percentage of the class that should be meeting that definition.

This leads to things like Lund not noting the percentages and Tilburg not noting the definitions.

I favour Lund: if the student has done work that meets the definition that’s that.

53

elspi 08.18.08 at 7:05 pm

50 “Of course, the anti-grade-inflation folks would shoot steam from their ears at the thought of 80% representing an A-minus or 48% representing a solid C, even if the instructor can honestly claim to have designed the test that way.”

Being one of the anti-grade-inflation folks, and one of the “80% representing an A-minus” Instructors, I am calling you out as a LIAR.

54

student 08.18.08 at 8:35 pm

Dead wrong on a definitional level. Grade inflation does _not_ and never has meant an inflation of grades vs. work quality. Everyone knows from the get-go that calibrating an absolute measure of work quality is both impossible and useless.

Grades are normalized, and grade inflation refers to a shift in the normalization. Grade inflation is bad for the same reason that poor normalization is bad in any domain — because it makes the scale less useful. (Those who argue against grade normalization and for an absolute eg 90-80-70 scale are way off. The only way to calibrate a test that well in advance is to always give the same test, which leads to cheating. And it was, in fact, common and good in many of my classes for the raw scores to be in the 10s or 20s for students who did poorly.)

No one is arguing that only the best 4 or 5 students of all time should get As. The argument is that the average student in a given class should get a C, the very best students should get an A, the very worst an F. The absolute quality of the class or the institution is beside the point, and it’s pure self-serving snobbishness for Harvard to claim that their students are so good that they all deserve As. If they’re so good, let them be held to a higher standard so that we can tell who among them are the best. A few Cs aren’t going to hurt the brilliant grads of Harvard.

One other wrinkle caused by grade inflation: I’ve seen data numerous times showing a strong correlation between average grades at the departmental level and average reviews of teachers by students. In other words, there’s an apparent incentive for faculty to inflate their grades so that they’ll get better reviews from their students, leading to an all-A-all-the-time future.

This is exactly what normalization is for. Use it!

55

Lucas 08.18.08 at 8:43 pm

I m willing to bet (without data) that the current situation is one of “grade equilibrium.”

Individual institutions do not have many incentives to combat inflated grades, that is, to “go back to the good old days” when the majority of students earned C’s. What would it mean if, next year, a C from Harvard required the same amount of work as an A from Yale? If a single American university were to reverse grade inflation by issuing relatively lower grades, then its students would be handicapped in the marketplace. Additionally, the university would signal its own incompetence in failing to graduate students with higher GPA’s.

It is also reasonable to assume that a single institution can only inflate grades so far before its reputation would suffer. “Everyone gets an A at Happy University” (wink, wink). Lower grades still must be given for poor performance, and they do indeed mean something when they are given.

Most likely, there is an equilibrium. Universities actually have a set of recognizable incentives to inflate, and, as a result, employers (and admissions committees) generally realize that high grades alone do not signal a person’s competence/employability. But, if a University inflates grades too much, it devalues its “product.”

Although it isn’t the same as the good ol’ days, the current state of affairs isn’t necessarily inferior. An A is not guaranteed. Lower grades are undesirable. Universities incur a reputational cost if they inflate too much, so we can stop wringing our hands. Overall, its a system that everyone understands. Realistically, unless the incentives are radically (an uniformly) changed, the current system of grading is here to stay. As long as we all know what it means, then where’s the crisis?

56

Righteous Bubba 08.18.08 at 8:55 pm

If a single American university were to reverse grade inflation by issuing relatively lower grades, then its students would be handicapped in the marketplace.

See Princeton.

57

lemuel pitkin 08.18.08 at 8:56 pm

A data point: I’m just now coming back from an interview for a mid-level position in my corner of the New York City bureaucracy. The candidate, with a BA and Masters from a top state university, offered to send us his transcript. No need, we said. His resume didn’t give his GPA either — nobody cared.

As John Q. said in his piece, the importance of grades to employers is nil.

58

geo 08.18.08 at 9:18 pm

the importance of grades to employers is nil

But the importance of grades to the admissions committees of top-tier business, law, medical, engineering, and other professional schools is very great. And the importance of degrees from those schools to employers choosing from a large pool of applicants for plum entry-level jobs is also, I suspect, considerable.

59

sg 08.18.08 at 9:19 pm

I too have never had to present my qualifications at or after interview – even the recent one, done over the phone from Japan for a job in the UK.

I also remember the scaling at my university for physics was very helpful to students. I got a Distinction for a 64% mark, and I think in maths the pass mark was below 40% in my first year.

And my Masters degree also, after I finished, introduced a Masters (Honours) to separate the sheep from the goats. But I think this is largely because Australian Masters degrees are becoming increasingly worthless. I am amused to see that Mansfield attributes the reason for these problems to all the black people. Very novel.

60

Gulielmus Ferox 08.18.08 at 9:21 pm

## 14 & 15 are hilarious. I plan to steal these inights shamelessly. Well done.

61

Scott P. 08.18.08 at 9:58 pm

None of us ever regularly got 90s, even though we were the top class in the top state secondary school in the country. 88%, say, was considered a rare achievement, and because that was the horizon of our expectations, we accepted it.

So what’s behind the discrepancy between that NZ 74 and its 90+ equivalent in Clayton, KS? I very much doubt that it reflects any real difference in student aptitude or achievement. Ergo: US high school and college grades are grossly inflated, when compared to those earned in other First World countries.

The real question is: 74% of what? If the best student can only get 88%, that suggests to me that either the test was too difficult for the course level, or that the teacher was inadequate. Under my testing philosophy, 100% (or close to it), should be achievable by the best students (and I’m not talking Stephen Hawking-level geniuses here). When averaged over enough groups of students, there should be a general bell curve around the average. What the average is varies — in an introductory course, the object is to learn a body of material, which most students should be capable of — an average between 75 and 80 isn’t unusual. Upper-level classes are more difficult and grading is more rigorous, but then the student population is self-selected to include only the better students. I don’t grade ‘on a curve’ since I try to have clear expectations and set the test difficulty to those expectations. Since I teach in the humanities, there is little need or desire to use lower level classes to ‘weed out’ the students who can’t cut it, so insanely difficult exams and draconian grading are not necessary.

62

Bloix 08.18.08 at 10:00 pm

My experience as a TA in a reasonably good history department in the 1980’s was that the grading of undergraduate work was arbitrary. There were no departmental standards, there was no training in grading, and there was no quality control. In that environment, it would have been highly unfair to grade from A to F on a bell-curve. But as long as acceptable work got an A or B, poor work got a C, and only students who didn’t show up got D’s and E’s, the arbitrary nature of the grading system didn’t matter too much.

63

Kikuchiyo 08.18.08 at 10:02 pm

In this entire thread, the words “adjunct”, “part-time”, “contingent” or “teaching assistant” never appear. It’s all well and good to talk about Harvard and imagine most courses are taught by tenured faculty. But the facts are the facts and most courses in American universities are now taught by part-time faculty who are often only on the radar, so to speak, if they get low evaluations or complaints from students. I have no doubt that many part-time instructors somewhat justifiably offer higher grades as a way to remain in good stead with their employers.
Tenured faculty are for the most part almost entirely ignorant of the “facts” of the courses that go on all around them. The answer isn’t standardization but, rather, engagement. If faculty truly took on the responsibility of mentoring and engaging with junior and part-time faculty, we would see a much greater benefit than simply grade “fairness.” Maybe–for the first time in a long time–a major in a discipline would have some coherence.

64

virgil xenophon 08.18.08 at 11:44 pm

Those who say that the importance of grades (or a “well-rounded” education at all) to employers is nil are probable right. As my Father used to say apros po of such things: “Speak to me in French if you must–but make me a buck!” It seems to me however that there really two questions:(1) is grade inflation really happening and (2) Does it matter and why should anyone care?

The answer to (1) above has to be an unequivocal yes to anyone being honest with themselves. And while it indeed began during the Vietnam era as a response to the draft, what has maintained it besides inertia is the fact that the world, following America’s lead, has morphed into one, big, therapeutic facility for the propagation of egocentric, self-administered self-esteem bolstering, i.e., the Lake Woebegone Syndrome where everyone’s children are above average and awards for high accomplishment are deemed a social right once one has managed to vault the necessary hurdles to join the right academic “club.” Anyone trying to deny this fact of grade inflation and the grandiose sense of self-importance and feelings of automatically deserving high grades that exists among the last few generations is whistling past not only the academic graveyard but the sociocultural one as well. As to the answer of the second question?
Not so clear cut.

In answering the second part of my question I would say that most here are focusing on the wrong level of the educational totem pole.
Granting that inflation exists at the college level aside, it is hard to say that it means much to employers on one level of analysis because educational institutions vary so much. As a result, most employers have, over time, learned to use the jungle grapevine about which institutions produce what kind of candidates in order apply some “Kentucky windage” to their hiring practices.

What employers ARE alarmed about, however, is the quite obvious degradation of knowledge skills of the product of today’s undergraduate institutions. To find a graduate today that can string together three grammatically correct consecutive sentences is like discovering the Hope Diamond–let alone one capable of producing three consecutive logically coherent ones. And their conscious knowledge of, well, what makes the world go round, is equally lacking in the majority of graduates. Ever watch Lenno’s “Jay Walking?” Especially the college graduates and, god forbid, the TEACHERS that appear there? And how many surveys do we we need in which most have not even a nodding acquaintance with even the very basic aspects of science, economics, etc., i.e., what makes the world go round. How many horror story surveys does it take like the one in which over 50% of respondents place New York State in a State west of the Mississippi river and every State therein labeled as New York by at least one of the respondents? The problem, it seems to me is that today’s graduates are so lacking in basic knowledge about most things as to be measurably dysfunctional to one degree or another.
Which is why other nation’s are beginning to eat our proverbial lunch with better educated, and subsequently better trained workforces.

But the real problem lies in grades K-12. Today these rungs on the educational ladder are little more than indoctrination centers for PC., IMO. That this is so one neeed only look at SAT average scores which have been in free fall since 1963 when they peaked. And it does no good to say this is all due to the result of English-as-second-language types entering at the bottom and dragging down the averages; and that the system is otherwise OK. If this were so, one would expect the absolute numbers of those achieving perfect scores to continue upward unabated in line with population growth, even as averages fell. But this has not been the case. The absolute numbers of perfect scores are sinking like a stone as well. Such trends strongly suggest that the educational system is failing, and failing fast. Grade inflation only helps disguise that fact.

65

Righteous Bubba 08.19.08 at 12:00 am

(1) is grade inflation really happening and

Harry’s assertion is that there isn’t evidence so…

The answer to (1) above has to be an unequivocal yes to anyone being honest with themselves.

What evidence can you provide?

66

Hans 08.19.08 at 12:32 am

Thanks for the post. Most instructive. . . . How lovely that affirmative action is once more made a villain: laughable. Affirmative action for whom: Harvard legacies?! Larvae of the oligarchy? . . . . Has any0ne ever demonstrated a connection between grades and learning? If not, why worry about grade inflation? Contrasted with global warming, plagues, the Bush presidency, etc., grade inflation is of less concern than a single pellet from one of Cheney’s blasts of friendly fire during one of his grouse-hunts. I’m much more concerned with Cheney-inflation. Thanks again.

67

PHB 08.19.08 at 1:07 am

I don’t quite know how to parse the comments on legacies etc. at the end of the piece. What is Mansfield’s position here and how has Harvard’s position changed?

As a Brit who went to Southampton and Oxford University, the idea that ‘legacies’ could be seriously defended as an admissions practice at a purportedly ‘elite’ university is preposterous as far as I am concerned. The fact that some on the right simultaneously rail against ‘affirmative action’ while still defending these legacies of discrimination merely confirms the view that the US is still hidebound by the class based social attitudes that Britain manged to shed almost entirely by the mid 60s.

I tried to find out what the current legacies policy at Harvard was, but the Web site seems curiously silent on the topic. I assume that means that it is still in practice.

I don’t see why colleges that practice such discrimination should enjoy the tax advantages that they do in this state. The threat of taxation is probably the only way that this obnoxious form of discrimination, originally brought in to keep Jews and others out, will end.

68

harry b 08.19.08 at 1:48 am

virgil — if you have actual evidence of grade inflation that would be great. I’m not at all unpersuadable on this, I have just come across absolutely no evidence at all from anyone; and, as I say, lots of people seem not to know what would actually be evidence.

PHB — I’m sure that Harvey Mansfield is as strongly opposed to legacies as to other forms of affirmative action. I just haven’t found him saying it anywhere.

My understanding is that Harvard does still practice legacies, but the bar is much higher than it used to be (i.e. students have to be much more competitive on other grounds than they used to be before the thumb gets put on the scale). We may well have readers who can correct me on that.

69

PHB 08.19.08 at 2:15 am

harry b – when people get into fits of indignation about discrimination it is always rather interesting to see the types of discrimination that raise their ire and those that apparently do not unless a specific question is raised.

And while it is certainly the case that Harvard, Yale and the rest have become considerably more coy about the issue of legacies, none appears to have made an unequivocal rejection either.

Without transparency there is no way of knowing if things are getting better or worse. And I will continue to regard Harvard as less than world class as a result.

70

vivian 08.19.08 at 2:21 am

Hmm. Grade inflation wouldn’t be a problem since teachers do not have an obligation to the outside world to provide a ranked list of students. Teachers have obligations to their students and their employers, and as teacher-scholars, other obs to their scholarly communities.

And yet, I suspect that the thing that motivates Mansfield’s better nature is the problem of Too Many Really Well Qualified Applicants for Grad School and Tenure Track Jobs. Nowadays there are hundreds of applicants for one-year adjunct jobs paying nothing, miles from anywhere. Piles of applicants to grad school and med school too. People on admissions committees are desperate to make the piles smaller, and people with Really Great Students are running out of ways to turn the rhetoric up to eleven.

The gentleman’s C was for people who didn’t need anything beyond family ties to find success after school. They surely didn’t want to become philosophy professors.

71

vivian 08.19.08 at 2:27 am

PS I don’t doubt Mansfield overheard some colleague(s) in the 1970s saying something like “well of course they’re not as good but we have to compensate for those racists…” Using the anecdote for a causal mechanism would definitely get one soundly mocked at Harvard these days, afaik.

72

Matt McIrvin 08.19.08 at 2:45 am

Somebody has to mention the Flynn effect in this thread, so I will.

Is the Flynn effect relevant to this question? It’s not measuring college-level humanities skills, obviously, but it does call into question the idea that basic reasoning skills have been decreasing or stagnant.

73

John Quiggin 08.19.08 at 3:33 am

“To find a graduate today that can string together three grammatically correct consecutive sentences is like discovering the Hope Diamond—let alone one capable of producing three consecutive logically coherent ones.”

On the other hand, to find a sentiment like this expressed at any time in the last 50 years is easy thanks to the marvel of Google books. Here’s one dating the complaint to the 50s, and expressly linking it to the beginning of mass higher education

http://books.google.com.au/books?id=8rj_VA0Y4NYC&pg=PA287&lpg=PA287&dq=declining+standards+schools+1950&source=web&ots=9eDDLF1Cyv&sig=HsOOt2JuddNa9Zr6N4WBdLj11-w&hl=en&sa=X&oi=book_result&resnum=3&ct=result

74

LogicGuru 08.19.08 at 4:07 am

I teach at a midrange private college where we are continually pressured to keep a lid on grade inflation by administrators and gung-ho faculty who congratulate themselves on how tough they are because they’re low graders. I’ve heard that one department, not mine, has established a quota system–no more than 15% A’s.

I suppose I get it. Degrees from more prestigious colleges carry enough weight to get students decent jobs and admission to decent graduate programs. But a degree from a school like the one at which I teach, or the the one where I did my undergraduate work, carries little or no weight. My impression is that for students outside of the sciences and engineering, to get a job that requires a college degree–not a wonderful, highly paid job–they need at least a 3.5 GPA. Less than that and they’re competing on a more or less equal footing with applicants who have “some college” in the market for routine clerical work and the like, and may as well not have bothered sticking it out for 4 years.

This puts me and other faculty in a lousy position because while as a teacher I want to get all students to do as well as possible as an agent of a pre-employment screening program I have to rank students and see to it that what my university regards as a suitable percentage get (and presumably deserve) adverse grades. I wish there were a system of external examinations to do the employment and grad school pre-screening so that I could teach without worrying that I was doing too good a job at it.

75

virgil xenophon 08.19.08 at 4:32 am

John Quiggin: HA! Ya got me there. As soon as I punched send I thought of the Head of Tulane’s School of Education in the mid-fifties voicing much the same sentiments (can’t quite get my hands on it now–think he was complaining that majority of students majoring in ED. were in lowest quintile of their H.S. graduating class). My only answer is that, yes, this trend/impression has been around for a while,
but I would hold that a good longitudinal study would confirm my frankly anecdotal view that it has been intensifying geometrically/logarythmically over time and seems now to have burst finally into full flower. But what do I know? Maybe it’s like beauty or obscenity–all in the eye of the beholder.

76

Righteous Bubba 08.19.08 at 4:35 am

I would hold that a good longitudinal study would confirm my frankly anecdotal view that it has been intensifying geometrically/logarythmically over time and seems now to have burst finally into full flower.

What factors would be involved in this study?

77

John Quiggin 08.19.08 at 4:48 am

On zero evidence, a la Mansfield*, I suspect that a large part of the complaint about declining student quality rests on the evident fact that the graduates of the 1930s and earlier were nearly all gentlemen, while the vast majority of today’s graduates manifestly are not.

* I have to admit I’ve only seen him once, so my suspicion that he would be quite amenable to this kind of argument is based on similarly limited evidence.

78

giotto 08.19.08 at 5:22 am

PHB:
I don’t know about Harvard, but legacies still get a bit of an advantage at
Yale. As one might expect, that is particularly true “if the legacy comes from a donating family.”

In some cases at Yale the admissions process becomes base money-grubbing:

A family can significantly increase its child’s chances of getting into Yale by giving a large sum of donation money, the former admissions officer said. Yale’s development office has an official list of students whose parents or families make substantial donations. “Development kids,” as the admissions officer called them, are almost guaranteed admissions if their families are big enough donors.

“The development office has an A-list, a B-list and a C-list,” the officer said. “The A-list has kids whose parents, for example, are CEOs of Fortune 500 companies. The B-list has kids whose families are second-rate compared to the A-list, and the C-list is basically friends of really important and wealthy people.”

Sometimes, the former admissions officer said, he was not even required to read the development kids’ folders because there was no way to reject them.

“Year after year, there are kids who get into Yale and take the spots of more deserving students because of money,” the former officer said.

79

PHB 08.19.08 at 6:24 am

Giotto – quite true, only the weird aspect to the Yale scheme is that they track ability to donate, they have no idea whether the donors will actually cough up the goods or not. And its worse than that as they only know the reputed wealth, not actual. Set up your own 501c, name it after your grandfather, hire the offspring as an intern over the summer so they can put it on their resume.

Vivian – methinks that the obsession with quantifying the unquantifiable state of Higher Education is the reason that it has got into the sorry state it is in. Most of the academic work I see at conferences is unremittingly sterile. Boring silly little problems that are set up for the purpose of allowing cute mathematical solutions of the approved form and a minimum publishable unit.

At the last conference I went to I had to sit through an hour and a half of silly password ‘usability’ schemes designed to make it easier for the user to remember an unguessable password, completely ignoring the fact that today we all have fifty or more of the things and thus they all have to be the same if they are going to be usable. Anyone who wants serious security does not rely on a single factor solution. In the real world we solved this problem years ago and the users have made plain that they are not going to stand for password schemes that are made any more complex than at present. But people can make a cute problem out of it and measure it to death.

If Witgenstein applied for a job at Cambridge today he would probably be turned away. Fortunately he would still get in at Yale – but only in the hope that as the heir to the richest fortune in Austria he might leave them a handsome endowment!

80

abb1 08.19.08 at 6:55 am

‘Grade scaling’ – if that’s what it’s called – is what they do here in Geneva in public schools. The same number of correct answers in the same multiple choice test would get you a different grade next year, depending on how well the other students do. Distribution of grades remains constant, your grade signifies your relative position within the cohort. You success has to be judged by the combination of your grades and the year they were received, like wine by its name and the year. I don’t think they do it, but I suppose it wouldn’t be too difficult to assign a weight to each school year that could then be applied to your grade-average to produce the absolute value of it.

81

Dave 08.19.08 at 8:00 am

With regard to the employability of graduates, “inflated” or not, article in today’s Guardian [UK]:
http://www.guardian.co.uk/education/2008/aug/19/students.highereducation
Killer quote: “Last year, just one in 10 applications submitted to the Cooperative [by graduate applicants] was filled in properly. Most were either incomplete or failed to follow basic instructions. “
60% of leading recruiters say graduates lack “writing skills”.

And yet educational achievement has never been higher…

82

Stuart 08.19.08 at 8:45 am

Surely academic qualifications really only need to be comparable within a limited age range – once someone has a few years of experience any academic qualifications, and especially the exact grade achieved, becomes rapidly less important .

83

Z 08.19.08 at 8:50 am

On the other hand, to find a sentiment like this expressed at any time in the last 50 years…

Or indeed, in the last 2000. Cicero wrote a discourse about it, IIRC (and to think that a hundred years ago I could have quoted it, but the level in Latin is not what it used to be). My own contribution (rough translation) “The spelling of the literature students has become so faulty that the Sorbonne has had to ask for the creation of a new tenure position whose main occupation would be to correct the literature assignments of its students”. Dated 1899.

Anyone trying to deny this fact of grade inflation and the grandiose sense of self-importance and feelings of automatically deserving high grades that exists among the last few generations is whistling past not only the academic graveyard but the sociocultural one as well.

Except that most serious longitudinal study actually show that the level has risen, as I have already written. Ot to take the Alex way, I swim faster than Johnny Weissmüller. Do I deserve an A in sport? Most, but not all, spelling for instance has really become worse on absolute terms these last 20 years (in France).

84

Tracy W 08.19.08 at 9:21 am

Grade scaling’ – if that’s what it’s called – is what they do here in Geneva in public schools. The same number of correct answers in the same multiple choice test would get you a different grade next year, depending on how well the other students do. Distribution of grades remains constant, your grade signifies your relative position within the cohort. You success has to be judged by the combination of your grades and the year they were received, like wine by its name and the year.

In New Zealand this was traditionally done on the national exams at age 15/16 and 17/18. The exams were released after the examination, so couldn’t be used again next year, so the exam difficulty could vary from year-to-year so scaling was a way of dealing with that variation.
It got quite complex, eg more As were permitted in Latin than in typing based on the difference in the grades those students tended to get in the English exams.
It was abolished before I took School C, which was a bit nervewracking due to the risk of that year having a tough exam.
I don’t think there is any perfect solution to grading. Releasing exams may have caused some problems for inter-year comparisons but it was a good incentive for the exam-writers to write good quality questions.

85

Alex 08.19.08 at 9:28 am

“Grade scaling” appears to be what was done in the UK pre-1983(?). We called it “norm referencing” as opposed to criterion referencing. For some reason, decline fans refuse to learn from one year to the next that GRADES NEVER VARIED AT ALL BEFORE 1983 BECAUSE THEY WERE FIXED IN ADVANCE AND THEREFORE CONTAIN NO COMPARABLE INFORMATION WHATSOEVER. Excuse the caps; I get tired of pointing this out.

I assume rightwing people like this because it means however well you do, if others do well it doesn’t do you any good; it’s that David Frum scare-them-into-being-more-conservative thang. Why they refuse to remember that getting rid of it was the declared policy of the Thatcher government is another question.

86

Dave 08.19.08 at 10:08 am

This is all very well, but if everything isn’t getting worse, why is it so hard to find decent servants these days?

87

John Quiggin 08.19.08 at 10:11 am

#72 I just use my son’s names as passwords for ease of memorization. But for security we call him r!t45Lpg Hbn6@34 8Hrtöes and we change his name every 60 days.

88

Dave 08.19.08 at 10:15 am

Re. norm referencing. To point out the bleedin’ obvious, there’s nothing wrong with it – unless you assume, for example, that there are an unlimited number of places at elite institutions, that could be filled by a theoretically unlimited number of A-grade students. Since there aren’t, using grades as a marker of how good you are compared to everyone else applying that year, as opposed to how good you are against some theoretical absolute standard, made perfect sense, and would do again if revived. As noted above, actual grades rapidly become meaningless once a little life-experience has been accumulated. And not least of the advantages of a return to the older system would be to do away with the annual blather about pass-rates.

With that little conundrum out of the way, we could go back to arguing about whether the actual exams do get easier – only we could do it sensibly, by actually looking at the questions set and the answers expected, not leaping to conclusions about a 0.2% shift in awards here or there. We could even ask what the exams are supposed to be for. But that might be a stretch.

89

Kevin Donoghue 08.19.08 at 11:35 am

#72 I just use my son’s names as passwords for ease of memorization.

#72 is now #79, appropriately enough.

90

harry b 08.19.08 at 12:38 pm

Dave,

50 years ago nobody filled in applications to the Cooperative correctly. Because there were no written applications.

91

Greatzamfir 08.19.08 at 12:50 pm

What’s “The Cooperative” ? It sounds a bit scary, like the Borg.

The Borg only accepts beings of absolute perfection, and slaughters the rest. If the Cooperative is only slightly like them, spelling skills are indeed critical for your well-being.

92

harry b 08.19.08 at 1:40 pm

Its a big retail company. More scarily, it is the biggest provider of funeral services in the UK, so has some odd incentives built into its food quality control systems.

93

harry b 08.19.08 at 1:42 pm

Oh, and Dave (88), I agree with that. Alan Johnson said he was going to set up a database to make this possible, but he was reshuffled a few weeks later, so probably didn’t.

94

engels 08.19.08 at 1:51 pm

Surely academic qualifications really only need to be comparable within a limited age range – once someone has a few years of experience any academic qualifications, and especially the exact grade achieved, becomes rapidly less important .

Yes, that’s the solution! Because nobody ever has to start a career from scratch several years after graduating, after having, say, taken time away from the single-minded pursuit of The Corner Office (TM), in order to do something stupid like start a family, or care for a dependent relative, write a novel, recover from a serious illness, or just to pursue a completely unrelated occupation which turned out to be inappropriate. Nobody would be that crazy, surely?

95

Dave 08.19.08 at 2:01 pm

Nobody would manage to “start a career” after a significant multi-year break if all they were able to offer was an old exam transcript, regardless of whether their 66.7% equalled that year’s B+ or three years later’s A-.

96

engels 08.19.08 at 2:07 pm

Also, on the proposals by Lemuel and Christian above that universities shouldn’t give out final grades at all, of course I’d love to live in a world without grades (or at least one where people treated grades with the indifference they deserve) for similar reasons to those touched on by Geo. But in the world we live in, my guess is that the effect of this as regards employment would be to further advantage mediocre graduates of the most prestigious universities relative to brighter students from lower ranking institutions.

97

engels 08.19.08 at 2:10 pm

(I didn’t say that was ‘all they were able to offer’, did I?)

98

PHB 08.19.08 at 2:10 pm

John Quiggin: That’s good, you must let me steal it and use it in my security presentations.

99

engels 08.19.08 at 2:27 pm

(Also, the whole point is that many recruiters screen applicants based on grades, not transcripts. Anyway, as I have said before, Dave, I’m not really interested in trying to arguing with you about this or anything else. Please go and bother someone else.)

100

engels 08.19.08 at 2:33 pm

Also, none of the above should be taken as disagreement with Harry’s critique of Mansfield, which I thought was very well-taken, my only reservation (if it is one) being that seemed more patient and reasonable than the guy deserves.

101

Dave 08.19.08 at 2:49 pm

When you get to be one of the blog owners, you can stop me commenting, can’t you? I don’t see, myself, why your long paragraph of sarcasm was so superior as not to merit a short paragraph of sarcasm in return.

102

Righteous Bubba 08.19.08 at 3:59 pm

using grades as a marker of how good you are compared to everyone else applying that year, as opposed to how good you are against some theoretical absolute standard, made perfect sense, and would do again if revived.

I’m not sure if this is helpful to the student. X people in class are dumber/smarter than you seems to me to be less worthwhile than a signifier of how well you understand your material.

103

engels 08.19.08 at 4:01 pm

Please go and bother someone else, Dave.

104

harry b 08.19.08 at 7:04 pm

engels — patience and reasonableness to those who don’t deserve it is a lot of fun, when you can manage it….

105

engels 08.19.08 at 8:00 pm

I’ll happily take your word for it…

106

James Wimberley 08.19.08 at 9:27 pm

The last word on examinations must be the anecdote about the German prince in the Baroque era who advertised for a new court composer. He had a chamber orchestra and no choir, so the applicants were asked to submit a few suitable instrumental concerti. You have no doubt heard the submissions of one of the losing candidates – to, of course, the Elector of Brandenburg.

107

F 08.19.08 at 10:58 pm

Okay, it’s definitely not conclusive, but are you seriously suggesting that students got substantially better from 1967 to 1971, didn’t change at all from 1971 to 1988, and then have gotten monotonically better from 1988 until 2000? Because that’s what Rojstaczer’s data indicates.

108

Righteous Bubba 08.19.08 at 11:10 pm

are you seriously suggesting that students got substantially better

I believe the suggestion is that while grades went up there isn’t an identifiable cause to point to, so no.

109

John Quiggin 08.19.08 at 11:13 pm

#98 Feel free! I stole it from someone else.

110

harry b 08.19.08 at 11:41 pm

#109, it was obviously protected with an insecure password.

111

Benjamin Mako Hill 08.20.08 at 3:24 am

I find this discussion a bit silly. Grades are a very simple single dimensional scale that we use to summarize and communicate very complicated events. Of course these sorts of problems come up.

Meanwhile, there are colleges and universities who have been successfully been using written evaluations for decades in place of or in addition to grades. It’s not hard to call out that truly exceptional student in an written evaluation or the brilliant one who could have put more effort and done more than very good — you just say so.

112

Dave 08.21.08 at 11:23 am

@104, thanks, that’s what I try to offer to sad crypto-neo-Marxist rants. It’s a tough job, but someone’s got to do it.

113

dr ngo 08.23.08 at 7:17 am

Belatedly, FWIW, I wouldn’t blame the VN war too much for grade inflation:

1) At the time, when there was talk of professors giving higher grades to enable people to beat the draft, it was *men only*, and in most cases mentioned, the “inflation” was just enough to lift them from a failing grade (or D) to a passing one. The key concept, supposedly, was not to be responsible for their failing out of college and thus losing 2-S status and becoming draft eligible. There were, anecdotally, a very few profs who decided to give the entire class “A”s as a kind of protest, but I can’t imagine they had significant impact on academe as a whole – except as yet another angle from which to question the significance of grades in general.

2) As others have noted, grade inflation – or at least the perception of it – was not limited to the US. E.g., it was widely perceived at the University of Hong Kong, where I taught for many years, and certainly would have had nothing to do with draft exemption or (affirmative action!) there.

114

Chris Garos 08.24.08 at 4:12 am

The really funny thing about legacies at schools like Harvard and Yale is that those schools don’t even need the money. You can make a decent argument that if someone donates $1,000,000 to a financially-struggling state school or relatively young private institution, their kid ought to be admitted – their money benefits hundreds of other students and allows for more scholarships to low-income students, academic support, better facilities, etc. Taking the kid probably keeps the money flowing and builds a better future institution. But Harvard and Yale are at the level where they couldn’t spend down their endowments if they tried, and they’re STILL chasing more endowment money as if they’re William and Mary or Northeastern. What’s Harvard going to do: bulldoze Allston and build a city of gold?

Comments on this entry are closed.