Grade Inflation Again

For some reason, grade inflation seems to be (once again) a hot topic.  Larry Summers weighs in, with a brief headlined "If we really valued excellence, we would single it out."  I've already suggested a couple of reasons why I think that people who are concerned about grade inflation might be a bit overwrought.  So what does Summers add to the discussion?

I think that the pervasiveness of top grades in American higher education is shameful. How can a society that inflates the grades of its students and assigns the top standard to average performance be surprised when its corporate leaders inflate their earnings, its generals inflate their body counts, or its political leaders inflate their achievements?

That's an argument?  Evidence that grade inflation exists?  Frankly, I read it as an assumption that grade inflation--defined as students receiving higher grades for equivalent work than they would have received in the past--exists.  I don't see any attempt to present evidence.

Consider the equivalent in any other sector of the economy...say, health care.  In a world in which production processes change, in which we learn more about how to do things, in a world in which we observe that there are some things that we need to focus on, would we be content with a hospital sector that produced the same results it did 20 or 30 or 40 or 50 years ago?  Of course not.  So when hospitals report that they have reduced patient infection rates during hospital stays, or reduced patient readmissions to hospitals (because the patients don't need to be readmitted), do we go "Ooooh...bad hospitals...how could things be getting better like that?"

Well, look at Harvard, since Summers has essentially spent his academic life there.  What has happened to the quality of the incoming students with which Harvard has to work to produce quality educational outcomes? This has:

Fifty years ago, Harvard sent acceptance letters to 20 percent of roughly 6,700 applicants to the Class of 1969. By 2006, that figure had dropped to 9.7 percent for the class of 2010 and since then has continued on a downward trend overall, reaching a record-low 5.3 percent acceptance rate earlier this month.

So someone applying for admission to Harvard when I was about to enter college  would have had a 20% chance of admission.  Today, it's 5%.  In addition, the entering class at Harvard is about the same size as it was in 1965, despite a 60% increase in the US population and a vastly larger pool of applicants from outside the US.  Do you think Harvard attracts a less qualified pool of applicants today than it did 50 years ago?  Or would you guess that is uses stricter admissions criteria?  I know which end of that I'd take.  The quality of Harvard's entering students is higher today than if was in the past.  How much higher?  I don't know.  But it's probably substantial.  (Yale's acceptance rate is 6.3%; Dartmouth is 10%...all of these are much lower than in the past.)

With a more talented, better-prepared group of admitted students, wouldn't we expect students at Harvard to perform better?  If not, why not?  Harvard has a Bureau of Study Counsel that provides assistance to students who need it.  I'd be willing to bet that no such thing existed 50 (or 40 or 30) years ago.  Go look at that.  Guides to writing papers in 29 different courses/situations.  Guides for studying in introductory science courses.  Do you think that first year students in 1965 (or 1975, or 1985) had anything like that?  Wouldn't we expect that such additional support would lead to better performance. 

Harvard has a large and diversified faculty development program, which includes development of teaching skills.  Wouldn't you be surprised to discover that these programs existed in 1965 (or 1975...maybe by 1985)?  In the sciences, Harvard faculty have been leaders in developing new methods of pedagogy.  Wouldn't we expect that this additional support for faculty would lead to increased learning?

And what has happened at Harvard has happened at institutions across the country.  Highly selective institutions have become more selective.  My undergraduate institution (a fairly selective, private, liberal arts college) enrolled about 2,400 students in 1965; it now has about 2,200.  And it also has expanded its recruiting efforts overseas.  Its potential student base has probably at least doubled.  But it is smaller.  More, not less selective.  Why wouldn't students there perform better?

If all this is right, student performance is rising, not constant, at least at selective institutions.  The only way to have maintained the grade distribution that existed 50 (40, 30, 20, 10) years ago is to redefine the criteria upward.  I do not necessarily object to that.  But if that's what we want to do, we should be open about it.  And unless our standards have changed, we should not be surprised that, at least at selective colleges, student performance has improved.