Writing Proficiency by Religion on OkCupid.com

by Luke Muehlhauser on September 8, 2010 in Funny

OKCupid is a dating site that mines their database of users to come up with fascinating statistics on dating. One recent finding that is totally not surprising:

Previous post:

Next post:

{ 26 comments… read them below or add one }

Reginald Selkirk September 8, 2010 at 9:27 am

So what’s the conclusion you draw from this? That atheists, even if they are highly educated, still need to use a dating site to find suitable partners?


Adito September 8, 2010 at 9:32 am

That’s not much of a difference. 1.2 grades or so. I’m not sure how they’re measuring proficiency though… I remember learning a lot of completely useless grammar rules about then but nothing about how to communicate complex ideas. Hopefully adults do pretty well in the latter even if they don’t in the former.


Thrasymachus September 8, 2010 at 9:40 am

Isn’t like 9th Grade 14 years old or so? And they’re meant to be grown-ups?


Blair T September 8, 2010 at 9:51 am

I checked out the link and the second table on this was even more pointed – showing that the less serious people were about religion, the better their writing proficiency.

Reginald – I am one of those highly educated atheists who used a dating site to find my wife. It worked great for me – I found someone who shares my values and tastes who I would not have crossed paths with otherwise.


bossmanham September 8, 2010 at 10:48 am

I think what’s more surprising is that the average reading level for everyone on the site doesn’t go above 9th grade.


stamati September 8, 2010 at 10:56 am

Not serious Buddhists scored higher than everyone. Nice.


Hermes September 8, 2010 at 11:05 am

Is any of the raw data available^ so that other calculations beyond the average can be performed?


^. Minus personally identifying bits, or limited to age, proficiency, and religious affiliation.


CrossTrained September 8, 2010 at 11:19 am

Jezus rocks, guyz. You need to lern that he rulez and if you dont except him that you will be lost forever! He’s always waitin for your to cum home!


lukeprog September 8, 2010 at 11:33 am

hahaha lol


Derrida September 8, 2010 at 11:36 am

Interestingly, writing proficiency increases for religious people when they take their beliefs less seriously. But for atheists and agnostics, writing proficiency increases when they take their beliefs more seriously.


Jeff H September 8, 2010 at 1:00 pm

Just to note: It’s likely that the 9th grade writing proficiency is because these people are not writing treatises on metaphysics or creating new theories of quantum mechanics. They’re describing themselves for a dating site. So a) don’t conclude that apparently nobody passed high school, and b) realize that there may be other issues at play here. For example, since they are using the Coleman-Liau Index which counts the number of characters in a word as a measure of writing level, it may simply be that describing one’s atheism requires longer words than describing religious beliefs. “God” versus “metaphysical naturalism”, perhaps.


Daniel September 8, 2010 at 1:21 pm

I’m not so sure about this – Catholic schools messed me up in a lot of ways, but I’ve always felt like one of the few useful things I did get out of it was an ability to write well.

I would also like to know the methodology here. Having used that site, I can say that well over half the people I met there seemed to me to be above a 9th-grade level … not that I tested them on it, or anything … ha ha.

Also, on their website, under religion, there is a category for “Other” … where did all those people go? I’d like to see their results too.


Ben September 8, 2010 at 6:49 pm

The graph seems visually misleading. The difference between Protestants and Atheists is about 1 grade level, everything else is pretty even. If you did the graph on a k-12 scale it probably wouldn’t be as provocative looking.

How much can we really infer from a one grade level difference when those grade levels involve being either 13 or 14 years old? It just seems like at OKCupid, people probably write poorly because 1) they are on OKCupid and 2) not writing their honors thesis.


svenjamin September 8, 2010 at 7:00 pm

Okcupid is run by a bunch of Ivy League math majors. They know what the fuck they are doing. I would like to know the variance on those grade reading level scores though.

And I’m dating a pretty cool girl I may or may not have met on okcupid.


Hermes September 8, 2010 at 7:03 pm

Well, there’s an additional complication. Writing complex sentences isn’t an indication of having greater intelligence. It’s just a demonstration that you can write complex sentences.

Ideally, shades of Occam, the sentences should only be as complex as necessary but no more complex. The old rule was to aim for a 5th grade level of complexity even if the words used are well beyond that low mark. Revel in the words and the meaning, not in the arrangement of those words in an effort to require intense concentration from your audience.


lukeprog September 8, 2010 at 7:22 pm


Indeed. On my past site I actually refused to publish anything I wrote that was measured at higher than a certain grade level. Simplicity is king, if it can be achieved. Often in science and philosophy, it cannot. At least not by me. At least, not without a million words.


Hermes September 8, 2010 at 8:01 pm

Luke, agreed. Imposing an artificial metric as you did can be useful as a learning tool. In general, some tools allow you to improve your writing without having to wait to have the text cool off. It also mutes your own ego somewhat, allowing you to slash through what the ego tends to protect (legitimately or not). That said, nothing beats 2-3 days of not looking at the text before editing it.

It is hard work to write simple sentences and it’s not always possible to just apply force to get the optimal results. In the sciences, that is why Carl Sagan was such a good presenter and (to a lesser degree) Neil deGrasse Tyson is taking his place but Richard Dawkins is not as skilled in the same way that Sagan was.

As an example, I’ve spent the past year working on my ability to rhyme on the fly and I’m just now getting good enough that I can do it at the level that a young child can understand what I’m saying. Moving beyond that will probably take a few more years though I may not be able to get very far.


Thomas Lantern September 9, 2010 at 8:56 am

Proficient writing does not a good person make.


Hermes September 9, 2010 at 9:11 am

It is an indication of both aptitude and intelligence, though. People who know more and are able to do something with it are less likely to make bad decisions or to harbor ill-informed prejudices.


Jugglable September 9, 2010 at 8:52 pm

Posting this seems smug. And seeing as how atheists are approximately 1 grade level higher on this graph also makes it seem petty.


Hermes September 10, 2010 at 1:30 am

Yes, it is smug, just as comments about religion being necessary for ethics are smug. Discussing them, though, actually fleshes out the facts or identifies where there are gaps that need investigation.


Jugglable September 10, 2010 at 8:40 am

Hermes — First of all, if it WERE smug to say religion is required for ethics, it wouldn’t make this smug posting any better. And it’s not smug to say God’s existence is required as a ground for ethics. Because to say so says nothing about whether religious people are actually morally superior to non-believers. And, the idea that God is necessary for morality has actually been defended by atheists.


lukeprog September 10, 2010 at 5:03 pm

Comments? Are you working?


lukeprog September 10, 2010 at 5:05 pm

How about after upgrading Akisment to? Comments? Working?


lukeprog September 10, 2010 at 5:05 pm

THIS IS A COMMENT WRITTEN BY HERMES (that got posted into the far future because WordPress comments got fucked up, so I deleted it and am reposting it here)


Jugglable, it is smug to say that religion is required for ethics precisely because it is a claim that has not been examined. I don’t begrudge the claim, though, but I do begrudge the shallowness of the claimant in not actually examining the claim openly. I mention that specific claim precisely because it is brought up frequently and as a truism. That I think it is demonstratively false does not impact that it is a smug claim.

That it is false only intensifies the need to actually discuss it with people who have an interest in having as many of their claims grounded in reality and not fiction or bias as possible.

That attitude — wanting to get it right factually and critically — is neither shallow nor smug.

Now, back to the topic of this thread. OkCupid’s data. You are correct that the mere mention of the results implies smug superiority. Note that what Luke comments on directly is not the chart, but a list of what self-identified white people like (in contrast to other groups that self-identify themselves as something else). With a sample size of over 1/2 a million, it’s probably worth taking a look at even if it is flawed.

If you look at the responses to this thread, note that quite a few of them were not smug responses. Go ahead and count them up. What ones showed smugness and what ones were curious about the data.

Did you take a look? Here’s my count, not including your comments or my responses to you so far. See if you agree with it;

Smug = 3 (1 atheist, 1 Christian, and 1 unknown)

Analytical/Critical = 13 ( judgment call: 7 primarily analytical, 6 primarily critical)

Neither = 11

Humorous = 3

Now, if all 3 of the humorous are added to the atheist smug pile, the results of Luke bringing this up are primarily still either neutral or questioning.

My conclusion is that, smug or not, bringing up the survey increased interest in the facts and investigating them not solidifying the surface conclusions.


Victor December 7, 2010 at 8:37 am

All this chart shows is that personality types (MBTI) that have a propensity towards skepticism and knowledge are more careful about grammar and spelling. Typically religious people tend to be more extroverted and feeling. Not saying it’s exclusively based on personality type, but you’ll find plenty of ENFJs in church. I’m a Christian INTP (a personality type typically associated with skepticism) so I take note to be as clear as possible in my communication, though I may not catch typos since I do not have an eye for detail. Personality types normally associated with Atheism tend to pride themselves on exactness and competence. Following Foucault’s theories on power and knowledge, they elevate things such as mathematics, philosophy and grammar (things in which they happen to excel) above more characteristically emotional traits. For the Jewish bar of the graph, I would have to speculate that the tradition of literacy in Judaism (going back thousands of years) would be a boon to writing proficiency. I would imagine as well that the sample size for every religion to the right of Catholic is much too small (about 5% percent of the US population) to fairly compare it to the nominally Christian denominations (around 80%). The claims made on the basis of this chart also assume that the demographic of OKCupid users is comparable to that of the US population, which it is not. As Mark Twain famously said, “There are three kinds of lies: lies, damned lies, and statistics.”


Leave a Comment