The Washington Post

We asked you to Name That Data and you kind of didn’t

Okay, so this week's challenge was a doozy. I told you to "do your worst," and that's basically what you did. Productivity was lost:

Offices were torn asunder:

Some of you gave in to despair:

Many respondents either ignored the hint or took a very broad interpretation of "it has to do with money."

Reader Clark Ridge e-mailed in to guess "percent Dutch ancestry," based on the high numbers in Iowa and Pennsylvania. Several tweeters said the same.

Two of you settled on an oddly specific answer:


Commenter "whiplashomega" came very close to the mark: "percentage of people over the age of 25 with at least a masters degree, from the US Census." The education angle here is correct, but it's still missing money. In the end only one respondent got the answer right, and another came close enough to receive partial credit. The correct answer is percentage of tax filers claiming the student loan interest deduction.

This is one of those indicators you'd probably expect to track with general income. But you'll notice that the numbers aren't terribly high in expensive urban areas like San Francisco and New York. Wealthier students are less likely to take out student loans (for obvious reasons), but more to the point, the student loan interest deduction phases out after a certain income threshold.

The geographic pattern generally tracks the share of graduates with debt in each state. State spending on higher education also likely plays a role.

Once again the good folks at PolicyMap nailed the answer, and within like five minutes of when I posted the challenge. Since they are technically ineligible to win the Right on the Internet prize, having won it last week, I award them the Early Bird Prize for Prompt and Expeditious Accuracy.

That brings us to commenter RP_McMurphy. In a tour-de-force of logic and observation he concluded that A) the correlation between the mapped quantity and income/poverty was low; B) that "the stark contrast between some states and the counties abutting them -- Iowa and Pennsylvania in particular -- leads me to believe that ___ is in some way influenced by state-level policy"; and finally C) that "it has something to do with education spending, probably post-secondary."

While not 100 percent correct, this is closer than anyone else came, at least without the aid of whatever dark arts PolicyMap used to derive its answer, so I award RP_McMurphy the title of Data Wizard/Ninja/Unicorn/Whatevs, and pronounce him/her Less Wrong on the Internet than Just About Everyone Else.

Thanks everyone for playing, and we'll do it again this Friday!

Christopher Ingraham writes about politics, drug policy and all things data. He previously worked at the Brookings Institution and the Pew Research Center.
Show Comments
Most Read
DJIA -1.29%
NASDAQ -3.25%
Last Update: 02/07/2016(DJIA&NASDAQ)



Success! Check your inbox for details.

See all newsletters

Close video player
Now Playing

To keep reading, please enter your email address.

You’ll also receive from The Washington Post:
  • A free 6-week digital subscription
  • Our daily newsletter in your inbox

Please enter a valid email address

I have read and agree to the Terms of Service and Privacy Policy.

Please indicate agreement.

Thank you.

Check your inbox. We’ve sent an email explaining how to set up an account and activate your free digital subscription.