Is grade inflation really that bad?

Great Debates
21

Well, not really; it’s in essence a statistical impossibility, and as such doesn’t really demonstrate anything about the practicalities of a grade curve. The whole point of a grade curve is that it represents reality. You simply don’t get the situation you describe unless the class sizes are so low that we’re talking about cases in which I’ve already agreed curves to be inappropriate.

I just don’t agree with this definition of objectivity. Surely if anything it’s more subjective to attempt to grade a student against some arbitrarily defined standard of difficulty, when the standard of one’s peers is so much more constant?

I do take your point, but I feel that what we’re both doing is saying “well what about outliers?” You object that outliers in the case of extraordinarily intelligent classes will be disadvantaged by a curve, and indeed they will. I on the other hand feel that students faced with an outlier in the form of a poorly-set test will be disadvantaged by strictly score-based grading. I just strongly feel that, presuming the class size is large enough, the latter is the vastly more likely case. Any system designed to cope with thousands of students is inevitably going to produce unfair results for some. I’d hope that an entire class producing Einstein levels of work would not go unnoticed, in the astoundingly unlikely event that it happens.

I also sympathise with your POV about being judged relative to your peer group, but again I feel it’s the least bad option. I’d love it if an employer took an interest in my work, and I’d certainly like to discuss my thesis with anyone who took an interest. The fact of life is that employers don’t have the resources to do this on a wide scale, however, and I will inevitably be judged relative to my peers in some manner. Unlike you, I don’t find the proliferation of assessment days encouraging. I spent 4 years gaining a Master’s; at the end of this, I am judged on 4 stages of the most asinine crap imaginable. A personality test? “Team-building” exercises, where I’m expected to plan an escape from a desert island, then present my findings to near-comatose HR types? I would much rather be judged on the results I spent 4 years attaining, in the field of my choice.

(sorry, I realise that last bit’s gone a bit ranty, those assessment centres just really get my goat :))

Oh and re: your clarification, I think this is a very pertinent point, and is probably the main reason why Princeton is making the move it is. All inter-university grade comparisons are meaningless anyway, regardless of grading strategy; any decent employer will take into account the institution, and the relative value of its course and grades. If a Princeton degree is perceived to be devalued, it will become discriminated against. Curbing grade inflation simply has to be done to protect their main product, and some form of normalisation is the simplest way to achieve this.

22

Laigle, I think you made the point that I was was trying to convey in the OP. If someone really has their heart set on medical school or law school or anything else where GPA is important then they should stay away from schools like Georgia Tech, unless of course they are confident that they can get whatever grades they need there anyway.

I think this is unfortunate. When the time comes if my girls are interested in something like this I’d hate to have to tell them to aim for the easiest school they can live with. I’d feel much better if I could say go to the best school that will have you.

Maybe a resolution to the above would be for some organization to put out a yearly multiplier for each school, based on relative performance on GREs, LSATs etc. So, for Princeton or GT, we might multiply your GPA by 1.3 and at Florida (only since you mention it - I personally have no idea and am not looking for enemies) we might multiply by 0.85.

23

Based on my experiences, I would highly recommend a good in-state school for undergrad. If you’re planning on going to a graduate program, what you need is contact with professors, research experience, and personal referrals. You just don’t get that at the big name universities because the profs are in the lab doing advanced research all the time. It’s a lot easier to work your way up the collegiate ladder than to jump in high and stay there.

24

This is the essence of the issue. “Average” is supposed to mean “average,” god damnit. I’ll still respect someone who graduates from Princeton with a B- GPA more than someone who graduates from Craptown State University with an A. On a relative scale, simply getting into Princeton should give you an automatic A without any further effort. O_o The point is, that they have HIGHER STANDARDS because they are a BETTER SCHOOL. A term paper that might get you a B+ at Craptown State University could probably rate you a D at Princeton. THAT is the point. Now, if Princeton started giving everyone A’s, such an achievement would no longer mean as much. The hypothetical medical school in the above example would look at that “A” as slightly less important than an “A” from, say, Harvard.

Since the subject question is quite different from your OP, I will address it on its own.

“Is grade inflation really that bad”

No, it is great, if you are a sub-par student. They best place to look at this effect is public high schools. The conditions there are absolutely laughable. Even a high GPA from a public high school means very little. Schools let so many students graduate when they don’t really meet the standards it is sad. Look at it this way: I took AP/IB courses in high school. AP US History got me a solid B. My jackoff friends took CP US History durign summer school and all got easy A’s. Which of us do you think actually learned anything about history? The ones who got the A’s? Well, one of them ranked 7th in our school GPA-wise, graduated with honors, and scored 890 on the SATs, earning his way into some local crappy college.

25

This is a weak argument. By this standard, we should give everyone an “A”, in the hopes it will create more cooperation.

26

This is an interesting idea. :slight_smile: The question is, who would create and maintain these multipliers?

A better way to do this might be a simple requirement that every school release a breakdown of how many students fit into each grade range, yearly. Then somebody would likely create a company to accumulate and reduce this data so that it could perhaps be used as some sort of year by year relative comparator.

27

There is already a system in place for law schools to compare grades from undergrad schools. On each person’s LSDAS report there is a chart that shows what percentage of student applying to law school from a particular school had what grades. (There is also a chart that shows the percentage of students from a school that scored in each percentile range on the LSAT.) This allows law schools to see if a student comes from a school that graduates 70% of students with a GPA over 3.75 or only 5%.

Now, to the subject of grade inflation. I started a thread on this about a week ago. Here is a quote from that thread:

Grades are used to judge the performance of students. There is a very strong belief in students and many professors and administrators that students should receive high marks if they just work hard. This is ridiculous and damaging to education. If an admissions office is doing its job they should create a class of students that have the abilities necessary to succeed. However, there are going to be students who excel and students that do not. Grades allow outsiders (grad schools, employers) to see who is excelling and who is not.

For more on curve see the other thread:

28

Educational Marxism. Neato.

Keep in mind, some would say that teaching the kiddies competition is more important than teaching cooperation. To these people I say, “duck!” then kick them in the face when they do.

How about we just fund music and art instead of football? It would take a generation or two, but eventually people would be clapping for a well performed student-run musical instead of screaming wildly over a bunch of guys grinding each other to meat. Bleh. A guy can dream, can’t he?

29

Well, at least Princeton has the balls to do something about it…

Harvard has used the exact argument described above to explain why well over half of the grades assigned by the University are A’s. “If our students went to community college, they’d all be getting A’s!!!” Combined with admissions policies that include legacies, that’s why I have trouble understanding why people have so much respect for some Ivy League schools.

Really, the most important point of controlling grade inflation is that it needs to be done on a consistent, university wide basis. If some schools like Haahvad want to stick their heads in the sand and not allow graduate programs any differentiation between students, fine, screw em. But as long as information about the average GPA’s and relative grading difficulty is available to employers or graduate programs, universities that are consistent within themselves will be just fine.

I go to a pretty well regarded public university, recent controversies aside, and have taken some pretty challenging classes there. The only thing that annoys me is knowing that my GPA after taking Honors Chem from the chief editor of Chemical Reviews and a NAS member (he used a pretty similar scale to the one Kel Varnsen described) might be compared to someone that loads up on Contemporary Literature Through Film. That’s why it might be cool to see some more consistent grading policies throughout the university, but I cannot complain too much. And I doubt that it will serve as any pressing concern to the administration distracted by a single-minded reverence to maintaining the competitive standards of the football program or squeezing more funds from the state.

30

Oh, and one more thing. Thank god for truly standardized measures of student performance, e.g. the MCAT’s, GRE’s, etc.

31

The trouble is, speaking with my UK job hat on, these geniuses are coming in with their A grades so woefully under-equipped that Physics and Maths degrees are shifting to 4 years to make up for what they should but don’t know.

32

I used to teach in a college. I confronted grade inflation in many different forms.

Before I taught college, I taught 2 years of high school (before I went to grad school). I quickly discovered that the average student performs at a much lower level than you possibly expected. I imagine this is quite a shock for most beginning teachers.

The problem is that if students do bad it is your fault. I graded ‘fairly’ but soon parents were asking loaded questions, I was getting ‘visits’ from the school administration etc. What to do? Well, I could work very hard getting my students work up to snuff even if they weren’t all that willing…or…

I took the low road and gave easier tests and grades. Soon everyone was happy. Hell, I was even district #1 beginning teacher of the year and have a plaque to show for it. If I had stuck to my guns then my quality of life at work would have been much lower and I even could have lost my job.

There just isn’t any reason for a high school teacher to give lower grades.

When I started teaching college I was involved in admissions and scholarships. Believe it or not, there is a HUGE amount of 4.0 high school grads out there. However when you look at the SAT scores…many of them score poorly.

I thought grade inflation would end at college. My experience had been that HS teachers may have to do it because if HS students do bad it is the teachers fault but in college it is the students fault. This turned out to be true.

—but—

Huge amounts of grade inflation.

I didn’t notice it for two years but I would get student complaints. I would ask colleagues and they said it was normal but when one very persuasive student presented a ‘report’ to me on his research into grades I took a look. Grades are posted outside office doors for students to look at after final exams. I walked around and recorded grades. I had opportunities to do the same at other colleges.

The same result. Over half A’s. Many times closer to 2 out of 3. I averaged about 20%.

This was so irritating and stupid. The grading system as freewheeling as it is…is meaningless. Was I supposed to ignore the way of the world and stick to my guns? Was this ‘fair’?

I was so incensed by this that I pushed it in meetings and with the administration and even with state officials. I proposed a curve system much like talked about in this thread. I even proposed a faculty (or division) weight so that if an instructor (or division) gave lower grades that the GPA was inflated higher and easier graders weighted lower. A ‘B’ in Dr. Hard’s class might have a higher GPA than an ‘A’ in Dr. Easy’s class.

The last idea didn’t go over well though the curve idea seemed to have merit. In the end, it required too much effort to change and people didn’t want to do it. If they had a model to follow…???..and hopefully Princeton will lead on this issue.

Why do college faculty grade easy when the pressure that HS teachers have to face isn’t there? I asked…and it still is ‘easier’ on the faculty. Students love you and you are known as a good teacher. People like to be giver of good news and not bad. Faculty too scared of student evaluations…less complaints and so on.

Many people when faced with upholding the integrity of the system vs. the job being easier on them will take the latter.

Something needed to be done then (10 years ago) and it was getting worse. By now I shudder to think of what it is like. Step to the plate Princeton!

33

Good post. Thanks!

It’s been quite a while since I took my SAT’s but I’ve read that over the years, the test has been made easier. Perhaps they face the same pressures from parents and teachers? I’m sure some parents write them nastygrams when their 4.0 perfect students wind up with 800 or 900 score. And I’ve always wondered why they chose to start the SAT grading at 200 points?

34

I think a person can get a score lower than 200 for a division (if the raw score is a negative number), but 200 is just the lowest reported score.

35

This attitude is exactly the problem with educational reform today - the idea that there have to be “winners” and “losers” or the whole exercise is meaningless. Maybe I’m just old-fashioned, but I thought that learning and understanding had something to do with it.

I remember taking college chemistry classes in which a 42% on a test was enough to score a “B+” because of the curve. 42% is failing - either the teacher is failing to teach, or the student is failing to learn. Given that no one achieved above a 60% on a test in that class - teacher failure was the most probable cause. But the mandated curve ensured that even people who had no freaking clue (like yours truly) could escape with at least a B, because I understood the material 10% better than the bulk of the class. It was stupid, since I clearly didn’t understand the material at all.

Grades should not be given relative to anything other than the understanding of the material. If a teacher is such a great teacher or has such motivated and bright students that his or her class understands the material being taught at a 90% level or above - they should all get A’s. Assuming that the material must be too “easy” and that the difficulty level must be kicked up a notch to ensure that there is an adequate “sorting” of people has the potential to penalize the students and the teacher for doing well. In a class with mandated grade distributions, the teacher has a lower motivation to bring up the students in the middle, and even less to bring up the students on the end. Why bother if, at the end of the day, someone has to lose. A teacher should want all of their students to learn and understand the material. They may not expect it, and most of the time, it’s probably not realistic to expect it. But they should want it.

Grade inflation is not necessarily a positive, as some grades (witness my chemistry grades) are clearly unearned. But mandating a curve simply reduces the incentive to try and teach the material so that students understand it.

36

The test has only been made easier once - it was “recentered” in 1996 or 1997, which adjusted the grade that they would give to the “center” (50th percentile) students. I took it on the turnover year (and I still can’t remember whether or not I got recentered).

Interestingly enough, the standardized tests work wonderfully for detecting high school grade inflation but are generally ignored when a high school with grade deflation (or just a tough curriculum) is encountered. I nearly aced my SATs, and every school I interviewed at asked the same question:

“Do you think your SAT score, when compared to your high-B/low-A average indicates that you might not be applying yourself to your full potential?”

I was at a boarding school with a vicious curriculum, though. You can’t say “My curriculum is tougher than 95% of your other applicants’ work, so figure out my rough position based on my SATs and then you can judge what kind of potential I’ve got.”

…because if you answer like that in a college interview, you’re not getting in. So you smile and nod and say “School’s tough, and I work hard. If you don’t think that’s reflected in my grades, I guess I need to work harder. I hope you’ll take my heavy class load under consideration.”

37

The problem here is your college’s administration. Either they truly had a professor who could not teach or more likely they were admitting students in into this class who were not qualified. This is why there needs to admissions standards and prerequisites. There is no way a 42% should equal a B+ in an undergrad chemistry course and any student who was prepared to take this class should be able to receive a 42% from just reading the course material.

Also, a curve shouldn’t create a minimum number of students who receive a grade, it should create a maximum. All students in a class should be people who are qualified to be in class and able to succeed.

38

**Is grade inflation really that bad? **
Yes, anyone should be able to figure out the a 2.5 GPA from Princeton is worth way more than a 3.8 GPA from some college like XSX …

39

I don’t think it should create either. My idea of a perfect grading system would be that a D- would be the grade given when a student just meets acceptable standards for passing the class. Then a heirarchy ranging from barely acceptable to outstanding would be given from D- to A. I would create a department wide guide to which grade should be given for what knowledge. For example if a student knows x y and z that is barely acceptable so they would get a D-. If on the other hand a student knows a b c d e f … x y z then they would be outstanding and given an A. If a certain class didn’t have anyone that reached the outstanding level then so be it no one gets an A. If half the class reaches an outstanding level thats fine half the class gets an A.

I think I would also remove the sort of mystery surrounding what is important to learn in class and what won’t be tested. Every student would have access to the departments guide to grades for each class. They would then it would be up to them to study and meet expectations at which grade they wanted. I would make test questions exactly the same as an assigned homework question with different numbers. This way every grade can be defended, if someone comes and begs for a higher grade I can pull out their test and the departments guide and show exactly why they are at their level. It would also allow the college to publish this list for potential employers. Say I am working at acme company and I have certain requirements of an employee I can look at their grade in a class and the grading guidelines to see exactly what skills they have.

40

jeevmon, I think you’re taking the wrong lesson away here. If you’re going to curve a test, the best thing you can do is write one which has a range of difficulty, and is longer than you think the students can finish. Why do I say this? Because it gets you information about all of the students. If you write a test that a significant percentage of people do very well on, you can’t rank those people.

If half the class understood the material as presented, and is able to do the problems posed, they all get the same score, but they certainly didn’t know the same amount of stuff. That’s why you should have so-called “genius catchers,” questions which are very difficult and which you don’t expect people to answer. In this way, if there’s someone who really gets it, they’ll solve the really hard problem, or at least make a good stab at it. Most people don’t even get to that problem, but so what? You’re curving.

This may not be what your prof had in mind, but its certainly a possibility. Since you say he intended to curve, he would be insane if he gave an easy test and was then faced with the unhappy prospect of trying to assign different grades to people who all had the same score. Also, if you give a test which is long, then the students get to work on it for the whole hour. Why should a very smart student finish in 20 minutes and then go home. They come to class every day and sit through the same lectures. When I took tests which were easy, I felt a bit insulted, the implication being that this is what the prof thinks the students can do. To hell with that, people come to college, in part, to be challenged. What better way than by working hard and producing something worthwhile for the exam? If the test is easy enough that an average student can get an A on it, you’re telling the above average students that they shouldn’t apply themselves to this course, and that their time would be better spent at the beach.

Kel, I’m afraid I’m going to have to disagree with your detective work on this one. There is an explanation for why a 42% should be a B+, and it doesn’t imply incompetence in any direction. Also, I take issue with the implications of the alst sentence I’ve quoted here.

First of all, I disagree with the implicit assumption that having all of the knowledge and skills necessary to pass the class, or even excel in it, is the same as scoring 100% on the test. If this were the case, how would you know who the Norman Einsteins are, and, even if you know them, why should they bother to apply themselves?

Secondly, just reading the course material should be the bare minimum effort required. Learning isn’t something you do by casually riffling through some photocopies of others’ work. It requires hard work and at some nonempty combination of blood, sweat and tears. If someone were taking this class and explained

would you think they had done sufficient work to pass? Had they learned anything?

Anyway, this is getting long, so I’ll pop off.

Tenebras