Higher Ed Data Stories

I frequently hear that the interactive charts I publish are too confusing or time-consuming, and that it's hard to get the story out of them without some work. So today, I'm making it easier for you, for two reasons: First, this is real student data, not summaries: Each dot represents a student who applied for financial aid, so I'd never publish that data on the web; this is just a good, old-fashioned picture of a chart.  Second, in this case, one chart tells the whole story. The population here is all freshman financial aid applicants who completed a FAFSA but did not have need. Each column is one year, and each dot in that column represents a student; higher positions in the column show higher income, from zero to one million dollars in parental AGI (adjusted gross income).  This is arrayed in a box-and-whiskers, or box plot.  The yellow boxes show the limits of the middle 50% of the distribution (the "box") with the color break representing the median. ...

Ever since my first day in admissions, I've had a big problem with the concept of college application fees.  They just seem odd to me: You pay some amount of money for the privilege of being considered for admission, often not certain you'll attend if you are.  And if you're not admitted, you're out of luck. I understand those who support the concept, in concept: Students shouldn't apply to a lot of colleges, and they should be somewhat serious about the colleges they apply to.  Except we know that doesn't happen.  The counselor at my kids' school said a few years ago one student applied to 46, and the Fast Apps, Snap Apps, and VIP apps all encourage students to apply to places just because they can. I also realize that there are costs associated with processing applications, although those costs have dropped pretty dramatically in the past several years, especially when all the documents come in electronically. But all the costs of doing business are pa...

Another article appeared in my Facebook feed about college ROI, although it was called the 50 Best Private Colleges for Earning Your Degree on Time.    As is often the case, there is nothing really wrong with the facts of the article: You see a nice little table showing the 50 Colleges with the highest graduation rate. But it got me to thinking: What if high graduation rate wasn't enough?  What if a considerable portion of your freshman class that graduates takes longer than four years to do so? Is that a good deal?  Let's take some hypotheticals: College A: 1000 freshmen, 800 who graduate within four years, 900 who graduate in five, and 950 who graduate in six.  So the four-, five-, and six-year graduation rates are 80%, 90%, and 95%.  But of the 950 who eventually graduate, only 84.2% do so in four years. College B: 1000 freshmen, 750 who graduate within four years, 775 who graduate in five, and 800 who graduate in six.  So the four-, five-, a...

If you follow media following higher education, you know that for a while, many have been (somewhat gleefully) predicting the demise of the whole industry.  High costs, MOOCs, a weak job market, and shrinking confidence in the value of a college degree are all conspiring, they would say, to create a perfect storm that will be the end of us all. I'm not saying these people are wrong;  you can get in trouble arguing with self-proclaimed prophets, and until something either comes to fruition or it doesn't, all you have is a lot of heated discussion. Personally, I take exception to the smugness of some who seem to revel in their predictions. But that is, as they say, why they make chocolate and vanilla. The heat (if not the light) increased this week when Sweet Briar College in Virginia announced it was closing.  The pundits came out of the woodwork, proclaiming that this was just the first domino to fall, all the time apparently reveling in this presumptive proof of th...

While the endowment of a private university is not a big investment pot from which universities draw income to spend at their discretion (some portion of every endowment is restricted to certain use), it's a very good proxy for institutional wealth.  What's always been interesting is the enormous size of the top five or six institutions, always led by Harvard, in comparison to everyone else.  And yet Princeton, which enrolls fewer students, has the largest per-FTE endowment. This visualization shows two things.  On the top chart, it's a tree map, or what I like to call a sheet cake map.  Think of all the money in all the endowments as one big bowl of batter baked into a cake, and then, once baked, sliced up into pieces.  The size of the piece is that institution's endowment as a part of the whole. The bottom chart shows ten years of endowments, measured at the start of the fiscal year shown, so you can see the hits in 2008--2009 and the overall growth over t...

I love the Internet.  Thirty years ago, I couldn't have imagined being able to look up several years of tax returns for the Ivy League Colleges and Universities (let alone being interested in them.)  But Guidestar (a great site you should check out, in case you don't know it) comes to the rescue.  The documents are pdf, unfortunately, but you learn a lot by inputting the data manually into a spreadsheet. For your information: By law, all universities that receive Title IV funding must make tax returns available to the public, so there is nothing clandestine about this. The tax returns can show you, albeit at a very high level, at how the Ivy League Institutions generate revenue, and how they spend it. To no one's surprise, salaries and benefits dominate at almost all colleges and universities, and if you're really curious, the returns list in detail how much the officers and highest paid non-officers make. But as I once suggested , the most interesting thing is th...

Unless you live under a rock, you probably know that colleges are, in general, interested in increasing the number of students who apply for admission.  There are a couple reasons for this, but they're all mostly based on the way things used to be: That is, before colleges started trying to intentionally increase applications.  The good old days, some might say. In general, increasing applications used to mean a) you could select better students, who would be easier to teach, and who might reflect well on your college, or b) you as an admissions director could sleep a little better, because you were more certain you could fill the class, or c) your admission rate would go down, which is generally considered a sign of prestige.  After all, the best colleges have low admission rates, right? Well, yes, one does have to admit that the colleges that spring to mind when one says "excellent" all tend to have low admission rates.  Lots of people want to go there, and thu...

Frankly, the data are a little boring when you first try to visualize them.  When you're looking at the number of degrees awarded by discipline and by state, California, Texas, and New York win pretty much everything.  That's no surprise, of course, as they're the largest states with the most college students. So I broke it into regions, thinking there must be some differences in the degrees awarded in different parts of the country.  Nope.  The Middle Atlantic wins.  That's where the people are. Finally, I looked at each state by the percentage of degrees in certain fields, and voila! Something interesting. Different states award different types of degrees in dramatically different proportions. Some of this can be answered easily: A high percentage of business and computer science degrees in Arizona is driven by the University of Phoenix, but others are not so obvious. Why is there such disparity when you look at humanities, engineering, or health professi...

There are a lot of bad infographics floating around the Internet.  When they concern things like the difference between cats and dogs, or how many hot dogs and hamburgers Americans eat over the 4th of July, it's no big deal. But this blog is about higher education data, and when I see bad infographics on that topic, I feel compelled to respond.  This one is so bad it's almost in the "I can't even," category.  It takes very interesting and compelling data--The graduation rates of Black male athletes--and compares it to overall graduation rates at several big football schools in the nation.  Here it is: For starters, this chart appears to stack bars when they shouldn't be stacked: A graduation rate of 40% for one group and 40% for another group shouldn't add up to 80%.  The effect is that it distorts much of what your brain tries to figure out.  For instance, look at the overall rates (longer bars) for Georgia Tech and Pittsburgh;  Georgia Tech ...

I would have written a shorter letter, but I did not have time. --Pascal And sometimes it's that way with data visualization, too.  What starts out as a simple project--one you think will take a few minutes--gets slightly more complicated.  This morning, I came across this interesting Chronicle of Higher Education story , showing Pell Institute Data on Economic Diversity. If you don't want to look at the article, here is a screen grab of the chart.  Click on it to enlarge. It's not a bad chart, but I found myself taking more time than I thought I should to figure out the story, which is that for-profit institutions enroll far higher percentages of low-income students than public institutions.  (Another problem with this that I can't fix is that the data are not complete; for instance, there is no data on private Baccalaureate or Master's Institutions included in the set.) Additionally, notice the subtle changes in color when you move from the overall cat...

Last night, I gave a little demo of Tableau Software to graduate class, and tried to make the point that big, long, detailed spreadsheet reports are like a teenage daughter: The information you get sometimes seems like it's only given to meet the minimum requirement of reporting, not to allow you to extract any insight. This seems to be true with the people at NACUBO, too, who each year release a study of endowment values and one-year performance .  I'd encourage you to click on that link to see exactly what they provide.  Not only is the document lacking any insight about trends of the shape of the market, it's boring.  Most people will look at the top 15 or 20, and then go down the list to find institutions they know and make some comparisons. Extracting the data is difficult, and even when you do, it's laden with characters that should be stripped off or cleaned up due to footnotes and other caveats. Moreover, there is no ID number attached to the colleges (like ...

I've worked at a wide range of colleges and universities in my career: From a tiny little college with lots of adults and transfers and commuters, to a classic liberal arts college, to one of the country's best known and wealthiest colleges, to a place just coming out of financial exigency, to one of the largest private universities in the country.  Money, in case you didn't know, makes a difference. At one of those places, I was docked 18 cents on my first expense report reimbursement, because I had rounded up a tip, making it more than the 15% allowed by college policy.  But it was at one of the most heavily endowed colleges in the nation, at least on a per student basis.  I remember telling this to my mother, who only remarked, "Well, I guess now you know how they got it." So today's Chronicle of Higher Education article about the " Huge Explosion of Wealth " and the resultant $37.5 Billion in contributions to colleges and universities last ye...