The College Graduation Gap

Paul Tough has a piece in the New York Times regarding the differential rates at which college students from different class backgrounds make it to graduation.

Graduation Rates

Tough claims "about a quarter of college freshmen born into the bottom half of the income distribution will manage to collect a bachelor’s degree by age 24, while almost 90 percent of freshmen born into families in the top income quartile will go on to finish their degree." Because it's a newspaper, this claim comes without a citation. It is curiously at odds with the chart he displays subsequently in the piece, which shows even students from the richest quartile with the highest SAT scores have only an 82 percent graduation rate.

I don't know how this graph can be true and it be the case that 90% of students from the richest quartile graduate college by 24, but maybe I am missing something.

In any case, it is true that rich college students have much higher graduation rates. The Dynarski-Bailey figures from the National Longitudinal Survey of Youth, for instance, places the graduation rate of students from the poorest quartile that attend college at 31 percent and the graduation rate of the richest quartile at 67.5 percent.

So the basic story of hugely lopsided graduation rates is true, but how Tough came up with a 90% graduation rate for the richest quartile is a mystery to me.

Controlling by Score

The most interesting part of the piece is the first graph above regarding graduation rates controlled by SAT scores. Here is the original graph with the full figures:

This coheres with other analyses of the same National Educational Longitudinal Survey data, including one that the New York Times displayed two years ago:

The upshot here is that even when you only compare rich and poor students with similar SAT scores (which is meant to be a proxy for their college aptitude), rich students are graduating at much higher rates.

The Trouble

The trouble with this kind of analysis, which is partially reflected in the remainder of Tough's piece, is that it can lead people to think that these kinds of things are the biggest impediments to the college completion of poor kids. We've seen this before with the Hoxby undermatching stuff as well where the focus is on making sure high-scoring poor kids get into appropriate schools.

To the extent that we can reduce undermatching and even out score-controlled graduation rates, that's a good thing to do. But it remains the case that these are fairly marginal gains up against the larger forces that sort lower income kids out of college completion.

For various reasons, most especially the difficulties of contending with poverty itself, poor kids are heavily concentrated on the bottom of performance metrics.

Even if you eliminated undermatching and closed score-controlled graduation rates, the overall graduation gap (percentage of all poor kids who graduate, not just those who enter into college) will still be massive. Fiddling around at the edges for some low-hanging fruit might make some strides, but not big ones.

Finally, for those who have broader economic justice and social mobility concerns in mind when discussing these topics, it deserves pointing out once again that the economic advantages rich kids have over poor kids persist even when poor kids graduate from college. Narrowing the graduation gap won't presumably change the fact that a rich kid that does not graduate from college is 2.5x more likely to wind up in the richest fifth than a poor kid that does graduate from college.