OkTrends

For OkCupid‘s inaugural blog post, I’ve picked a few match questions and will be showing you some cool graphs. Graphs you’ll never find elsewhere.

But first, a reminder: OkCupid match questions are written by OkCupid users, not by staff. The community writes the questions, and our software simply asks them. Good questions climb to the top, and new users are asked to answer these first. By “good” I mean people (1) disagree over them and (2) feel strongly about them. God and sex are hot topics, as you’d expect. So are dating expectations, personal politics, and habits.

And a word about statistical validity: the best questions on OkCupid have been answered over a million times. Therefore we have unique insights into the American mindset. A quick comparison:

Old media could only get 3,050 people to answer a poll about Obama. And it was enough to call the election with confidence.

OkCupid, on the other hand, can ask the world’s most personal questions and get hundreds of thousands of answers. For example, here’s a provocative match question posed by a user:

300,000 people have answered that question in 3 parts, and there are thousands more questions with as large or larger data sets.

Yesterday, I wrote a Python jam that consolidates the data by state and plots the trends using Google’s chart API. I plotted a handful of questions, and what follows are a few of the more exciting results.

Would you consider role-playing out a RAPE FANTASY with partner who asked you to?

data set: 340,000 people answered

The few states skewing green said “Yes” more than the national average (hello, Nevada!), and the reddish states said “No” more. Perfectly yellow states (like Virginia and Tennessee) answered “Yes” and “No” in the same proportion as the nation as a whole. It’s worth pointing out that because there are only a handful of greenish states and yet roughly a dozen reddish ones on the other side of the mean, those few green ones came down very strongly in favor of rape role-playing compared to the rest of the country. It’s also worth pointing out that cattle outnumber women 26:1 in Wyoming.

Here’s the same question, plotted on Europe:

As you can see, the original England skewed the same as the brand New England. And there’s a strong interest in consent play in Lithuania, perhaps because of repeated Russian invasions. Sadly, countries in white have too little data for conclusions; maybe in a year we’ll be able to look deeper in those places.

Would you date someone just for the sex?

data set: 448,000 people answered

Westward, ho! I guess I didn’t know what I was expecting from this question; maybe that the more “metropolitan” places would be greener and the more “rural” ones redder, and while that turned out somewhat true (for New York at least), the overall geographic continuity of this plot was a big surprise. I guess things are literally more wide-open in the West, and it looks like the depression is hitting more than just pocketbooks in the near Midwest. Is the Rust Belt now the Chastity Belt?

How often do you bathe or shower?

data set: 261,000 people answered

This is a good example of results that seems surprising at first, but make sense after some thought: it stands to reason people in hot and/or humid states must shower more regularly. I’d be very interested to hear from any New Mexicans why their state bucks this logic; I suppose there are a lot of mountains there, so maybe it should have more in common with Colorado than with its other neighbors. Those of us up North skip once in a while, probably in the Winter. Vermont and Oregon are earthy as hell.

The scale for this map is slightly mislabeled, because this match question has multiple answer options. The first, “At least once a day,” is followed by “Most days. I skip some,” and goes all the way down to “Rarely.” This graph represents the fraction who chose “At least once a day,” and I labeled the scale as I did, because I think it most accurately reflects my query.

Should burning your nation’s flag be illegal?

Attempts to chart this question state-by-state yielded indifferent results, which makes sense since a person’s answer is fairly closely tied to his general political views and only a few states show extreme (far-off-the-norm) tendencies one way or the other on that. Digging deeper, I plotted this question by latitude and longitude. And now the results really show what’s going on: only people in cities believe flag burning should be legal. And even then, many conservative cities are opposed. Look at Texas: Austin, a liberal college town, is surrounded by Dallas, Houston, and San Antonio. It’s interesting that Florida, even though it went blue in the last election, is almost entirely red here.

These are just a few of the thousands of questions we have in our database. Are there any other you’d like us to analyze? I’d be happy to; just drop me a line. Either way, I hope you enjoy these graphs and the discussion. I had a lot of fun programming them. We’re going to be posting similar back-end and data discussions on this blog every week, so see you next time. Also, just so you know: though we plan to discuss and manipulate user data on this blog, it will always be anonymized.

OkCupid is totally free, so if you’ve stumbled on this blog entry, check out www.okcupid.com. If you’re an existing OkCupid user, invite your friends! Become a pixel in our graphs!