endowments and economic diversity

The NYT has an interesting story this morning about how top colleges do (or do not) use their large endowments to increase economic diversity.* Along with the article, the NYT also offers up the data themselves in a linked spreadsheet.** The selection criteria were a bit problematic for some purposes – they only include universities with graduation rates over 75% – but this is just a blog post, so let’s play around.

In particular, I was interested in the claim made in the article that endowment doesn’t especially predict economic diversity. The article uses a pretty scatterplot to motivate its claim that economic diversity is a choice: “The biggest theme to emerge from our analysis is that otherwise similar colleges often have very different levels of commitment to economic diversity. In this area, endowment is not destiny, and prestige is not destiny.”

Here’s a screenshot of their spreadsheet:

NYT 2014 Endowment vs Pell Grant

Looks pretty noisy! But what’s interesting to me is that although the NYT captured multiple variables, the plot picked out just one, Pell Grants, to measure economic diversity. But I wondered if the relationship with other variables might be a bit different. Specifically, the NYT also collected data about “net price” for students who come from households with incomes of $30,000 and $48,000/year. Here’s a much uglier scatterplot of endowments vs. net price:

Endowment vs Net Price

While there are still some schools with smaller endowments that have low prices, all of the schools with a high net price (above $15,000/year) have smaller than average endowments (<$350,000/student).***

So, what do you make of these divergent trends between endowments and Pell Grant rates (basically noise) vs. endowments and net price for middle income students (curvy, but distinguishable, relationship)?

EDIT: Had a bit more time to play around. Here's a graph of net price vs. logged endowment. The relationship is now a pretty strong linear one (the correlation is about -.7).

Logged Endowment vs Net Price

* H/T Beth Berman for the link.
** I puzzled over how to scrape this for a few minutes before trying to just cut and paste it into Excel, which worked fine.
*** Almost all of these schools have very large endowments, relative to the population of universities. The median endowment in the sample is $200,000/student and the average endowment is $360,000/student. Note that I dropped four universities for which the NYT did not provide endowment data.

Author: Dan Hirschman

I am a sociologist interested in the use of numbers in organizations, markets, and policy. For more info, see here.

4 thoughts on “endowments and economic diversity”

  1. I agree that net price is a far better indicator of the real cost of the student which is where extra financial aid would come in to play (though I went to a lowly state school so I have no clue whether select colleges determine net price before or after pell grant awards). In any case, Pell Grants don’t track state median income at all (and has a negative correlation): http://i.imgur.com/8Bjcm16.png. Perhaps tolerablely , but not strongly tracks poverty percent for all persons: http://i.imgur.com/cSfnMI0.png.


  2. So the implication is that colleges actually use endowment money well, right? That is, that more endowment money tends to mean better low- and mid-income access? As you noted on Twitter, UNC’s endowment figures are missing. If I’m reading http://bit.ly/YtheK6 right, our $3.5b endowment, or roughly $121K/student or $184K/undergrad for FY13, combined with a net price of $7600, would have to put us in the far bottom left corner or your distribution, right?

    Liked by 1 person

    1. I’m guessing a lot of the public ivies (basically none of which are included in this chart, often because their graduation rates didn’t quite make the 75% threshold) would end up in the bottom left quadrant.

      I’m still not sure what to make of it, but it is consistent with an actual effort to charge students from middle-income households a bit less at these elite universities (at least for the past few years). Of course, the same schools with low net prices might be the ones with low Pell Grant-eligible populations! Actually, let’s run that… the correlation between net price for middle income students and 2012 Pell Grant eligible % is about .14. So, slightly positive. Hmm!


  3. This is interesting because it seems that many of the schools that are “dedicated to economic diversity” are really creating high amounts of student debt for the students that attend. Also, with high prestige schools, I don’t know what percentage of the applicants from low socio-economic homes who apply would qualify to attend.


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.