the persistence of the old regime

Yesterday afternoon I ended up reading this Vox story about an effort to rank US Universities and Colleges carried out in 1911 by a man named Kendric Charles Babcock. On Twitter, Robert Kelchen remarks that the report was "squashed by Taft" (an unpleasant fate), and he links to the report itself, which is terrific. Babcock divided schools into four Classes, beginning with Class I:

And descending all the way to Class IV:

Babcock’s discussion of his methods is admirably brief (the snippet above hints at the one sampling problem that possibly troubled him), so I recommend you read the report yourself.

University reputations are extremely sticky, the conventional wisdom goes. I was interested to see whether Babcock’s report bore that out. I grabbed the US News and World Report National University Rankings and National Liberal Arts College Rankings and made a quick pass through them, coding their 1911 Babcock Class. The question is whether Mr Babcock, should he return to us from the grave, would be satisfied with how his rankings had held up—more than a century of massive educational expansion and alleged disruption notwithstanding.

It turns out that he would be quite pleased with himself.

Here is a dotplot of the 2014 USNWR National University Ranking, where the dots are color-coded for Babcock Class. There are two panels, one on the left for Private Universities, and one on the right for Public Universities. USNWR’s highest-ranked school at the moment is Princeton, and it is at the top of the dotplot. You read down the ranking from there.

You can get a larger image or a PDF version of the figure if you want a closer look at it.

As you can see, for private universities, especially, the 1911 Babcock Classification tracks prestige in 2014 very well indeed. The top fifteen or so USNWR Universities that were around in 1911 were regarded as Class 1 by Babcock. Class 2 Privates and a few Class 1 stragglers make up the next chunk of the list. The only serious outliers are the Stevens Institute of Technology and the Catholic University of America.

The situation for public universities is also interesting. The Babcock Class 1 Public Schools have not done as well as their private peers. Berkeley (or “The University of California” as was) is the highest-ranked Class I public in 2014, with UVa and Michigan close behind. Babcock sniffily rated UNC a Class II school. I have no comment about that, other than to say he was obviously right. Other great state flagships like Madison, Urbana, Washington, Ohio State, Austin, Minnesota, Purdue, Indiana, Kansas, and Iowa are much lower-ranked today than their Class I designation by Babcock in 1911 would have led you to believe. Conversely, one or two Class 4 publics—notably Georgia Tech—are much higher ranked today than Babcock would have guessed. So rankings are sticky, but only as long as you’re not public.

I also did the same figure for Liberal Arts Colleges, almost all of which are private, so this time there’s just the one panel:

You can get a larger image or a PDF version of the figure if you want a closer look at it.

Again, there is a substantial degree of stability over the course of the century. Here we see a bit more evidence of some movement up by colleges that Babcock put in Class II—Swarthmore, for example, as well as Middlebury and Pomona. The Class I schools that seem to have fallen from favor most are Knox, Lake Forest, and Goucher colleges.

Now, some caveats. First, because I was more or less coding this stuff while eating my lunch, I have not attempted to connect schools which Babcock did rate with their current institutional descendants. So, for example, some technical, liberal arts, or agricultural schools that he classified grew into or were absorbed by major state universities in the 20th century. These are not on the charts above. We are only looking at schools that existed under their current name (more or less—there are one or two exceptions) in 1911 and now. Second, higher Education in the U.S. really has changed a lot since 1911. In particular the postwar expansion of public education introduced many new and excellent public universities, and over the course of the twentieth century even some decent private ones emerged and came to prominence (such as my own, which competes with a nearby Class II school). This biases things in favor of the seeming stability of the rankings, because in his own data Babcock had the luxury of not having to classify schools that did not yet exist.

We can add these in a final, rather large, chart for the National University data.

You can get a larger image or a PDF version of the figure if you want a closer look at it.

Now the coding includes the pink "None" category, which adds universities that appear in the USNWR rankings but which are not in Babcock, either because they did not exist at all in 1911, or had not yet taken their present names. In fairness to him, the new additions still leave Babcock’s classification looking pretty good. On the private side, Duke, Caltech, and Rice are added to the upper end of the list, and a number of new private schools further down.

Meanwhile on the public side you can see the appearance of the 20th century schools, most notably the whole California system. The UC System is an astonishing achievement, when you look at it, as it propelled five of its campuses into the upper third of the table to join Berkeley. But the status ordering that was—take your pick; these data can’t settle the question—observed, intuited, or invented by Babcock a century ago remains remarkably resilient. The old regime persists.