Higher Ed Data Stories is a lovely blog run by Jon Boeckenstedt, an administrator at DePaul University. On the blog, Boeckenstedt produces interesting, relatively simply interactive visualizations using large national data sets. This week, he’s focusing on graduation rates using data from the NCES. His most recent post looks at graduation rates by institutional selectivity. His main takeaway is that selectivity and graduation rate are highly related, probably because selective universities have the luxury of picking students who are best suited to graduate:
People in higher ed, and especially in government, talk a lot about graduation rates, and the presumption is this: That graduation rates are something we credit or blame on the colleges; that is, something a particular college does determines whether or not its graduation rate is high.
But choosing a college because of its graduation rate is backwards: The college will select you based on your propensity to graduate.
Straight-forward, if depressing, enough. We already believe that much of the “effect” of education is about signaling and credentialing, and the link between graduation rate and selectivity follows naturally enough. After the brief introduction, Boeckenstedt presents a simply visualization of graduation rate by selectivity and gender, with the option to select by institution type or race. Here are two charts that struck my eye, breaking down the data for public institutions by race (graduation rates for white students vs. black students):
For white students, the graph is more or less what we expect: a gentle slope downwards from very high graduation rates at the most selective public institutions (84%) to much lower graduation rates at not selective institutions (50%), with each grade of selectivity correlating with a decline in graduation rate.
Now here’s the graph for black students:
Notice anything different? The first thing that jumped out at me was the steep jump between most selective and highly selective universities – a 25 point drop for black students as compared to a 14 point drop for white students. I would be curious to know what’s going on here (there’s probably some more detailed research in the higher ed world). At a glance anyway, it looks like getting into a “Most Selective” public is far more important for black students than white students. Is it an issue of differential resources to help students? It it that black students at most selective universities are more different from black students at highly selective universities than white students? Etc.
The next surprise was the nearly equal graduation rates for black students between Highly Selective and Selective universities, with Selective universities having a slight edge overall. It’s easy to over-interpret small differences, but the chart still troubles me given the usual counternarrative to stories of “overmatching.” To summarize: some argue that it’s a big problem when poor or minority students are admitted to more competitive universities for which they are not academically qualified. The usual counter argument is that more selective universities have much higher graduation rates (see first chart, and top of the second chart), and so overmatching isn’t actually a problem – the more elite a university you get into, the better, in terms of graduation rate. But does this narrative hold at every level of selectivity? Perhaps not.
Finally, and to end on an even more depressing note, comparing across the charts we see that white students attending Not Selective universities graduate at a rate higher than black students attending even Highly Selective universities (50% to 44%).
As Armstrong and Hamilton, among others, have shown, succeeding at college requires access to family resources that are highly unequally distributed. Black students, on average, lack these resources (see the incredible racial wealth gap). It’s interesting to think about how that plays out across the spectrum of colleges and universities, and what that suggests in terms of responses by individual institutions and policy more generally.