Sunday, September 29, 2013

Who's going to College in America?

On Twitter I came across an Atlantic article entitled, "Who's Going to (and Graduating from) College in Seven Charts."  I liked the idea instantly, but was disappointed by the charts.  Go ahead and see if you can discern the story from the charts on the article.  I suspect not; I couldn't, even though I knew some of this data already.  To be fair to the author, part of the problem is due to embedding a visualization in a space that's too small; part of it is conceptual design.

But I did notice a few of the charts were in Tableau, so I downloaded the workbooks and gave them a quick, 45-minute makeover. (The others are really bad screen shots of god-knows-what, and will see if I can find that data later).

I'm interested in whether you think these views are better or not: Do they tell a story about Higher Education?

Here is the first, showing graduates by ethnicity and gender:


Here is the second, showing who's enrolled in two-and four-year colleges:

Thursday, September 26, 2013

Show Me the Money

Recently, Harvard announced a little fund-raising campaign.  Since Harvard is known as the universe's most highly endowed university already, I thought it might be interesting to look at the world of college and university endowments.

This map shows 833 colleges or universities listed in the Chronicle of Higher Education's 2013 Almanac.  The size of the square shows the relative amount of endowment dollars in comparison to the selections; the size changes if you take out Canada, or filter by rank, or only look at certain states, but the actual amount does not.

The color shows the percentage change since the 2011 year-end. Hover over a square for details.



Wednesday, September 25, 2013

Education and Income

Many people argue you should get more education to earn more money.  Others think it goes beyond that.  And still others believe that income causes education more than education causes income.  And the income that causes education is the type that then causes income.  Got it?  Good.

This map shows US Counties in 1970 (top)  and 2011 (bottom), for all intents and purposes*, and the percentage of adults who had a high school diploma or less in those years.  You can hover over a county to see the data in a popup box.

Use the filter at the top right to select counties by their 1970 rate to see if they've made progress.  You'll probably find that almost every county has; those that used to be dark blue are now lighter blue, or even orange.  Select the highly educated counties in 1970 (any small city with a Big 10 University would do) and you'll see they're still up there, as you would expect.  So that's good.

But after doing that, use the second filter to select counties by per capita income in 2011.  What do you see? Anything we can hypothesize about what this might tell us?

*The data are county level data from The Rural Atlas (education)  and Census Bureau (income).  Data listed as 2011 are actually from the American Community Survey, 2007--2011.



Thursday, September 19, 2013

How much do colleges and universities spend on athletics?

Seems like a simple question.  And if you had to guess, you might be surprised.  You might also be surprised by who awards the most scholarships (it's Stanford) and which university has the most coaches (it's Harvard).

Take a look; follow the instructions in the box on the lower right.  The default view goes to expenses, but there's an awful lot of data here.  And if you prefer your data visualizations with the data pre-drawn, you can try here.



Saturday, September 14, 2013

USNWR and the Pressure to Cheat (even just a little)

As the annual US News and World Report hype returns, as it does every year at this time, it's interesting to look at the spillover effects of it can be.

Colleges, many of whom are like hamsters on a treadmill when it comes to chasing the elusive "prestige" believe students are as obsessed with it as much as they are. And they have been known to fudge numbers to make themselves look better. Frequently one person takes the fall for this, but even if only one person is responsible, and I think that's highly unlikely, the pressure to move up is palatable at many institutions.

The case of Claremont McKenna was interesting because it was the highest-ranked institution to be exposed, and because actual data was available for analysis. I did so, and found the results almost laughable: Not because cheating is right, not because these inaccuracies didn't have some effect on the rankings (it's believed CMC was clinging to its number 10 spot and felt some heat not to lose it), not that the number 10 only seems meaningful because we use base-10 to count, but rather because the differences are so slight as to be essentially insignificant to anyone except people who don't know better. See if you agree.


Friday, September 13, 2013

What's All The Fuss About?

This is the week when people go crazy. The rankings come out, led by USNWR and their list of the best colleges. We all know it's a fool's game to try to rank colleges. Malcolm Gladwell's piece sheds some light on it in his usual excellent, conversational style.

But if you want to see how we as a society obsess over the smallest sliver of the the market, and how the media focus on the elite colleges, it's important to see how that market is shaped. Here are four views of IPEDS data, 2012, breaking institutions into selectivity categories. I broke selectivity into five bands: To be in the "most selective" category, you need an admission rate of 12.5% or lower. The bands then cut at 25%, 50%, 75%, and everything over 75%. The blue bars, representing the most selective institutions, are but a tiny fraction of the higher education industry. Which, of course, is why people are so fascinated with them. 

This set includes only those four-year, not-for-profit institutions that admit freshmen. You can see how many colleges are in each category, how big the total enrollment is at those institutions, how many high school seniors applied, and how many eventually enrolled. Use the filters to narrow it down to region or Carnegie Classification, if you'd like.

Thursday, September 12, 2013

Access and Institutional Wealth

This data visualization shows the relationship between draw rate (yield/admit) on the x-axis (it's a measure of market position) and percent of students with Pell Grants on the y-axis.

The size of the bubble is institutional wealth (larger is wealthier) and the color shows the spread between the percentage of students with no institutional aid (aka full-pay) and those with Pell. (Pell grants are awarded to students with the most financial need, usually coming from families that make less than $60,000 per year.) Red shows a dominance of full-pay, wealthier students, and green shows the opposite. This is interactive, so you can play with it if you'd like.

Comments and questions below, please. This shows about 1.200 Private, not-for-profit institutions. All Data is 2012 IPEDS, which is not adjudicated. Responsibility for accuracy lies with the institution.