Thursday, February 27, 2014

A Deeper Dive on Financial Health

Federal Student Aid, a division of the Department of Education, releases its list of colleges and universities based on Financial Responsibility (sometimes called Financial Health, which I think is more accurate.)  In general, to continue to be eligible to award your students federal financial aid, you need a score of 1.5 or above (on a three-point scale.)  The scores go down to -1.

Most presentations of the data are pretty static: A table with the data, for instance.  But I think there is a bigger story here: Where are colleges in trouble located? How many are not-for-profit? How many students do they enroll? What about Pell grant recipients and students of color? Do are they more likely to enroll in colleges with failing financial health?

So I merged some 2011 IPEDS data into the mix.  Some of the results surprise me; neither Harvard nor MIT are a 3, for instance; the Franciscan School of Theology in California, however, is.  Results like this have caused some consternation among accountants, CFOs and their professional organizations, causing them to question whether there is any value at all to these rankings.

What do you think?  There are several views of the data here, including one that allows you to look at individual institutions.  For viewing, I've rolled all the scores into 1/2 point bands, and color-coded them. Enjoy.



Monday, February 24, 2014

Are Graduation Rates an Input or an Output?

It seems obvious: Students come in, and students go out. The type of students who enter your university are measured on lots of things, like test scores, GPA, ethnicity, and parental income, to name just a few. Universities are measured too, on lots of those same things, as well as others, including the graduation rate. Lots of people think the graduation rate is a function of what the university does or does not do, and in some sense, of course, they are correct: If you don't care about your students, or their progress, or you're not challenging their minds, they may leave.

But in another sense, it's also possible to think of outputs as a function of inputs. It's been suggested before by Malcolm Gladwell, for instance, that we often confuse selection effects with treatment effects:

Social scientists distinguish between what are known as treatment effects and selection effects. The Marine Corps, for instance, is largely a treatment-effect institution. It doesn’t have an enormous admissions office grading applicants along four separate dimensions of toughness and intelligence. It’s confident that the experience of undergoing Marine Corps basic training will turn you into a formidable soldier. A modelling agency, by contrast, is a selection-effect institution. You don’t become beautiful by signing up with an agency. You get signed up by an agency because you’re beautiful.
At the heart of the American obsession with the Ivy League is the belief that schools like Harvard provide the social and intellectual equivalent of Marine Corps basic training—that being taught by all those brilliant professors and meeting all those other motivated students and getting a degree with that powerful name on it will confer advantages that no local state university can provide. Fuelling the treatment-effect idea are studies showing that if you take two students with the same S.A.T. scores and grades, one of whom goes to a school like Harvard and one of whom goes to a less selective college, the Ivy Leaguer will make far more money ten or twenty years down the road.
The extraordinary emphasis the Ivy League places on admissions policies, though, makes it seem more like a modelling agency than like the Marine Corps, and, sure enough, the studies based on those two apparently equivalent students turn out to be flawed. How do we know that two students who have the same S.A.T. scores and grades really are equivalent? It’s quite possible that the student who goes to Harvard is more ambitious and energetic and personable than the student who wasn’t let in, and that those same intangibles are what account for his better career success. To assess the effect of the Ivies, it makes more sense to compare the student who got into a top school with the student who got into that same school but chose to go to a less selective one. Three years ago, the economists Alan Krueger and Stacy Dale published just such a study. And they found that when you compare apples and apples the income bonus from selective schools disappears.
Take a look at the interactive graph below, and see how strongly certain input characteristics are associated with graduation outputs.  What do you think?  You might also want to look at research from HERI at UCLA, especially the table on the bottom of page 23 in this publication.





Sunday, February 9, 2014

Trends in Federal Student Loans to Undergraduates

Since I was at the College Board Midwestern Regional Forum today doing a presentation, I thought I'd show some of their data in a visualization.  It's one I've been working on for a while, because it's trickier than it looks.

This shows loan volume for undergraduates borrowing Subsidized Stafford, Unsubsidized Stafford, and PLUS Loans.  In case you're not familiar with the parlance, you can read about it here.

Up until 2010-2011, colleges and universities could particpate in either The Direct Loan Program (FDSLP on these charts) or Federal Family Education Loan Program (FFELP) which was administered through financial institutions.

I think the trends are interesting, but you decide: I recommend when looking at volume you use constant (inflation-adjusted) dollars to see the trends unless you really want to see the growth in nominal dollars.

In order to stack the bars to show total volume, I had to do this on two charts, as it doesn't make sense to stack two averages for different types of programs.  Note that when you choose "Number of Borrrowers" and "Number of Loans" the numbers are not the same because some students do either Unsubsidized or Subsidized or both; a student is counted only once.  And the bottom chart only shows the average per type of loan; it's possible a student/family could take out all three types of loans, so it does not show student indebtedness.  That's a viz for another day.


Tuesday, February 4, 2014

Where did Doctoral Recipients go to College?

Don't get freaked by how busy this looks. It's starting off showing all the data, but you'll want to filter it down to make it easier to view.

This shows the college or university that awarded the bachelor's degree to doctoral recipients in 2011. I'd recommend you start by choosing just one of the large academic rollups, or even specific programs.

It's no surprise, of course, that larger institutions with more students produce more students who go on to a doctorate, so you can choose "Highest Degree Awarded" to look at different types of colleges. If you want to look at your specific favorite college, you can use the text box, or the state filter to narrow it down.

On all dropdown filters, make your selections and click "apply" to activate the filter.

The data came without the IPEDS ID number attached to it, so I can't (yet) merge this data with enrollment; but at some point I'd like to be able to convert the display to show percentage of graduates earning doctorates.  Obviously, it would be more meaningful, but I can't get to it yet.

Note: Large numbers of doctoral recipients came from foreign or unknown institutions; I've excluded them in the first view, but you can use the filter on the bottom right to add them if you wish.