Those damn kids and their high G.P.A.’s

It seems like every year, some segment of the old geezer intelligentsia gets worked up about ‘grade inflation’. This cycle, it’s the TimesEconomix blog that kicks off the hand-wringing, with a new dig at private schools:

G.P.A.’s have risen from a national average of 2.52 in the 1950s to about 3.11 by the middle of the last decade. For the first half of the 20th century, grading at private schools and public schools rose more or less in tandem. But starting in the 1950s,  grading at public and private schools began to diverge. Students at private schools started receiving significantly higher grades than those received by their equally-qualified peers — based on SAT scores and other measures — at public schools.

My hunch is that grade inflation hawks have more in common with old men in rocking chairs yelling at ‘kids these days’ to get off their lawns than they do with, say, actual education policy experts. We’ll start by debunking the traditional inflation argument, and then take a look at the public-private disparity. First, the argument against traditional inflation hawkery:

1. There is nothing metaphysical about grading. Just as there is no inherent, metaphysical value of a dollar, there is no platonic ideal of an ‘A’ or an ‘F’ floating out there in the ether that we here on Earth should be compelled to perceive and imitate. Hence, we should all be very wary of arguments based on average GPAs across varying time periods, as if this information on its own were enough to win the argument. That’s because …

2. The real purpose of grading is to provide useful information. First, grading exists to help tell the student how s/he is doing during the class, so that the student can adjust his/her own study habits, effort, and expectations. Second, grading exists to help others objectively evaluate a student’s mastery of the subject matter, usually in the context of a competitive application for a job, internship, faculty position, or what-have-you. That’s it. If rising GPA’s hinder these two functions, then grade inflation really is a serious problem. If not, then there’s very little reason to take the grumpy old inflation hawks seriously — outside of the camp value of listening to how things were back in the day. And as it turns out …

3. Our current system of grading does, in fact, provide useful information. You can rank people just as easily on a 4-point GPA scale normed at 3.11 as you can on a 4-point GPA scale normed at 2.52. Now, a rising mean GPA certainly could threaten to compress the highest performers on the upper end and inhibit oridnal ranking, but I don’t see much evidence that this is actually happening. There isn’t really any good data on this, but my own experience in college indicated that GPAs even at elite private universities are concentrated heavily around the mean, with plenty of room left over on either end for the high and low performers to distinguish themselves.

4. Finally, you say ‘inflation’, I say ‘attainment’. It’s a big, bad world out there — bigger and badder than the world was in the comparatively sleepy 1950s. There’s a lot more to learn now than there was then, and the tempo of discovery and invention is only accelerating. As a result, we’ve done a lot of what my mother disparagingly refers to as ‘curricular cram-down’, but what I prefer to think of as ‘learning more’ — especially among top-tier students. It’s now expected that our top-flight students learn intro calculus and stats in high school, along with a hard science and history. What used to be considered ‘college-level’ courses (the AP program) are now seen as prerequisites for admission at the best universities. And it just keeps on going. What used to count as a graduate level seminar in comparative Latin American government is now given to juniors and seniors. It ought to surprise no one, then, that average educational attainment in college may have grown from 2.5 to 3.1 in the last 50 years. Just compare the standard mathematics classes taken by kids at MIT in 1950 and 2010. What some call ‘inflation’ could just as easily be seen as ‘appreciation’.

“That’s all well and good”, you might say, “But what about the disparity in mean GPA between private and public universities? Doesn’t that put the lie to your cute little story, JSC5?”

I’m glad you asked. In a word, no. There are two possible explanations for higher private school GPAs, neither of which should worry us that the sky is falling:

1. Poor study design. The study compares the GPAs of what it calls “equally qualified peers” in public and private universities  based on SAT scores alone. The problem here is that grades in college are awarded based on performance in college classes, while SAT scores come from a test taken in 12th grade that covers information normally covered by upper-track students by the end of 8th grade.  The study’s assumption seems to be that if 2 students scored equally on the SAT, then they should probably score equally in any given class in college. For this to be true, however, we’d have to assume that the SAT truly does measure scholastic aptitude instead of high speed recall and use of 8th grade knowledge. We would also have to assume that there is no difference between the quality of the education at public and private universities that may lead to higher educational attainment among private school students. We’d also have to assume that there are no confounding variables that might legitimately distinguish the student who opted for the private university from the one who went public. All of these assumptions strike me as problematic.

2. Actual grade inflation. But maybe private schools really do inflate GPAs above comparable public school students’ GPAs. Is this the smoking gun that inflation hawks might think? No. Remember that 1) grading is not metaphysical, and 2) grades are supposed to convey information. Let’s drop the issue of resentment for a second and look at practical effects. Everyone knows that grading at different schools is different. Without a rigid national rubric, it has to be. That’s why college admissions committees look at population averages and distributions of grades from different high schools to help them interpret the individual GPA of any given applicant in light of their educational background. Graduate schools do the same when interpreting college GPA. The goal of having grades providing useful information to help construct ordinal rankings can still be met even if some institutions have higher average grades than others. That’s why you employ recruitment professionals with this kind of industry knowledge, and it’s why you do population norming. Certainly private university GPA inflation, if it existed, would make ranking more difficult, but certainly not impossible.

The take-home lesson is that grade inflation is not the slam-dunk case most grade inflation hawks seem to think it is. It’s hard not to conclude that persistent grade inflation hawks are really just angry that kids these days have a higher GPA than they themselves did back in the day.

Advertisements

2 Responses to “Those damn kids and their high G.P.A.’s”


  1. 1 JSC7 April 21, 2010 at 12:24 pm

    I’m going to cry foul on your third point. In your second point you say grades serve as measures for the student to help them improve and as objective measures for outside people to judge. I’m with you on that, but in your third point you focus on how there’s still plenty of room for exceptionally performing students to outshine their peers, which kind of rushes past the issue of objective measures for outside people.

    A GPA scale normed on one level or another is irrelevant within a college, but outside a college it makes a big difference. If you have otherwise identical students from similarly ranked schools, and one has a 3.5 and the other has a 2.5, and they’re apply to the same jobs, the company either isn’t going to know that the former might be inflated, or, if they do know, they’re not going to know by how much. If colleges want to do whatever they want internally, that’s fine, but then they should publish averages and standard deviations of grades (although even this isn’t necessarily enough – it’s not certain the grades follow a bell curve. Schools could be shifting the middle 65% of students up a few notches without changing either end, which would essentially mean that the amount of effort it takes to increase your GPA by .1 changes as you go up the GPA scale). Otherwise, colleges have an incentive to inch their grades up, and it could become a race to the bottom, potentially to the point where it is sub-optimal in terms of measuring performance (and, remember, depending on how they inflate grades, looking at top performers may not tell you about differentiation problems. The clustering around a mean for middle-performers that you mentioned could be a symptom of upper middle-performers having trouble distinguishing themselves from bottom middle-performers). And, once that race has happened, it becomes hard for universities to try to change their systems without feeling like they’re hurting their students. And then what you get is a system where people just don’t really trust the value of GPAs.

    Think about intra-university grade inflation: what the hell does a 4.0 in Harvard anthropology or sociology mean? I have no idea, but I’m inclined to be distrustful of the number. I imagine that’s what its like for a lot of people looking at GPAs on resumes. Which sucks, because then the only tools you have to recalibrate with are broad sweeping statements like “Oh, well, sociology is easy” or “Oh, Harvard inflates grades, Princeton doesn’t.” And, there’s no real objective solution – you can’t have a non-subjective grading system across all colleges – but I think colleges and departments publishing their grading statistics would be a good step in terms of demystifying GPAs.

    I’ll make a more questionable point: I wonder if there’s a psychological issue for students. Does getting a grade closer to the top of the scale than to the bottom of the scale make you feel better about your performance regardless of the mean? If you get a B+, but the mean is a B+, do you still feel feel better than if you got a C when the mean is a C? No idea, but it would make a cool psych study.

    • 2 JSC5 April 21, 2010 at 4:12 pm

      I definitely agree that colleges, departments, and specific classes and professors should all publish information on the distribution of grades. I don’t see why we should stop at the mean and standard deviation — have everyone disclose the actual distribution itself. That’d be really useful.

      But I don’t see how your concerns are fundamentally tied to grade inflation itself. It seems to me that your concerns are all about how different schools, departments, profs, etc., all grade slightly differently rather than according to a standard rubric. I agree with you that this poses challenges for ranking people, but I guess I don’t see how they are caused by *inflation* per se.

      Also, these challenges can be mostly overcome. I mean, if you’re Goldman and you hire someone with a 4.0 in some concentration, then see his/her work output and decide “I guess a 4.0 in X from university Y isn’t as valuable as we thought” … and then you learn how to adjust. And this happens for every single hire for every employer. It’s basically why HR is considered a particular profession with defined skill sets and industry knowledge. It’s not a perfect system, but it seems to work out ok. And, most importantly, it’s a system that evolved to adapt to the fact that grading varies across all schools, departments, and professors … not in response to grade inflation.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




Subscribe by email, feedburner

Subscribe by e-mail

or subscribe with feedburner

This is a group blog. JSC5 currently writes from the US. JSC7 writes from behind the Great Firewall of China.

wordpress statistics

Categories and tags

Archives


%d bloggers like this: