Atheists score highest writing proficiency level on Okcupid.com
Posted on: November 5, 2010 - 11:23pm
Offline
Atheists score highest writing proficiency level on Okcupid.com
Okcupid.com scanned over 500,000 profiles to put together several charts profiling various ethnicities and beliefs. Here is the entire blog at okcupid.
Is there a Comic Sans version of the Bible? There really should be. We subdivided this chart further, by how serious each person was about their beliefs:
Note that for each of the faith-based belief systems I've listed, the people who are the least serious about them write at the highest level. On the other hand, the people who are most serious about not having faith (i.e. the "very serious" agnostics and atheists) score higher than any religious groups.
Vote for Democrats to save us all from the anti-American Republican party!
- Login to post comments
I would have anticipated as much, It's never a surprise to see a correlation between lazy thinking and the most brutal and oppressive religious beliefs.
Do you find it as interesting as I do to see the dabbling buddhist sitting out there so distant from the central values? I wonder how that works.
Theist badge qualifier : Gnostic/Philosophical Panentheist
www.mathematicianspictures.com
Wow, that's a great study- and of a size that's actually statistically significant.
It's largely not surprising; even the ordering of the groups.
I imagine if you went to an Islamic country centric site and did the same, Muslims would be lower than protestants-- that is to say, I suspect it's an element of "default religion of the masses" that makes the protestants less educated than the Muslims in that Western centric context rather than the content of the belief itself.
I'm not surprised to see that the people who are "very serious" about the religion come out above the "somewhat serious" ones- back to consideration of the default, since "very" indicates a divergence from the norm and may imply more exceptional study (the apologists of the group).
The very serious Hindus and Muslims diverge from this, but I suspect that is because some of these individuals represent first generation immigrants and more socially isolated second generation immigrants, with lower English ability, thus bringing down the average significantly.
The Jews and Buddhists being above the agnostics are not surprising; Jews being more culturally obsessed with education giving them a bump, and the Buddhists there being primarily progressive-liberal white people who were looking for a religion that wasn't as crazy as whatever they were coming from (strong educational correlation there).
What is a little surprising is that the "not serious" Buddhists come out above the "very serious" Atheists. These "not serious" Buddhists are generally atheists, but that the cop-out religious claim would correlate to better writing is unusual. It might be that there's an angsty subset of teenage "I hate god" atheists bringing down our writing level, whereas the "not serious" Buddhists are more mature on average.
The lazy thinking is correlated strongly with the most common beliefs. I have explained this in more detail.
I know many of these people. They're claiming "not serious" Buddhism to avoid claiming "not religious", despite for all purposes not being religious. In a secular atmosphere, one can extract an admission of atheism from these people- but they will claim "Buddhist" for social/political reasons when asked.
These are largely people who have been inundated in the social belief that religion is good, and that everybody needs to have one. It's a cop out. If you're going to lie to yourself, you make it as close to the truth as possible- and that's what these people have done. There's a huge difference between the people who are "not serious" and the people who are serious about Buddhism at all.
That's why there's such a large divergence there- they're largely different demographics. It's not surprising to see the "not serious" triangle floating off by itself for Buddhism- I would have expected the same for Judaism, though, since (likewise) there are many "not serious" Jews who are actually atheists since it is largely an ethnic heritage as much as a religion. You'll see the same in Catholicism, but to a slightly lesser degree.
Not much condescension there, eh Blake? But I beg to differ, you've no basis for saying any particlar beliefs are 'common' among the sample group, not from the blog anywhere i can see anyway. Do you have privy to some additional information on the statistics or something?
I think if that were so the degrees of atheism would rank nearer beside this group, which is the only sector of the sample population sitting entirely above, by some margin, the 9th grade level. It doesn't. The dabbling buddhists are a distinct group indicated as more sophisticated than atheists among the population, not less, a your generalisation might imply.
In my opinion it may be an anomaly of the population itself, ie looking for love on the internet, rather than the belief. My guess is it's the member Indian nationals, who are by and large demonstrating more literacy than the rest, identifying as weakly Buddhist. I would propose that they might be attempting to use their english sophistication to make themselves more attractive to other races, I would guess they are not in a position to believe weak english would win them any brownie points with the potential mate. I think there are other reasons besides ability to write well that one might make the effort on the internet, the breakout culture mindset that characterises people in the Indian provinces does seem like one of those to me.
Theist badge qualifier : Gnostic/Philosophical Panentheist
www.mathematicianspictures.com
Yes.
You're evidently not familiar with the site's demographics, so I'll leave you to it.
Orly?
Are you trying to say because its a US site it doesn't have Indian nationals using it and/or having their profile information mined by Oktrends?
Theist badge qualifier : Gnostic/Philosophical Panentheist
www.mathematicianspictures.com
Shooting down my own theory, a quick data mine of the "Buddhism and laughing about it" demographic shows it, by all apearances, to be overwhelmingly comprised of white americans.
http://www.google.com.au/images?hl=en&safe=off&client=firefox-a&hs=Qb4&rls=org.mozilla:en-GB:official&channel=s&um=1&ie=UTF-8&sour...
So I'm back to wondering how the uniqueness of this group's written prescence comes about.
Theist badge qualifier : Gnostic/Philosophical Panentheist
www.mathematicianspictures.com
In a profound act of generosity, I just did a search and reviewed 100 buddhists on O.K. Cupid
About 70% white people (the majority being older gay men), and about 20% asian- but only three nationals (One in each Hong kong, Thailand, and Singapore), 15 second or third generation asians in the states and an australian Asian. Then roughly five african Americans. There was one Hawaiian guy (native Hawaiian), and one (only one) guy who might be Indian but who lives in the states.
Maybe the older gay man correlation even has something to do with it?
I did another search based on new members who are Buddhist, and I got 45% white (many older gay guys again), 40% asian (Mostly university aged second or third generation Chinese Americans), and two who *might* be second generation Indian or middle-eastern (in the states and the UK), and an african american guy.
That was only a 20 person random sampling, though, possibly not as thorough- I encourage you to follow up yourself if you really care- quite frankly I'm tired of clicking for the night.
The larger number of ABCs in the sampling might be biasing language level, given the cultural impetus on education (like in the case of the Jews).
Mature gay white men and Chinese American university students- though it might mean more to compare with the demographics of the other groups.
There are Indian personals sites- they have special search features for family and social class- O.K. cupid is primarily used in English speaking countries, and more in the states than anywhere else. You will find profiles from around the world, but not nearly so many- the demographics of those in the states (and Canada) overwhelm those from anywhere else in the database to the point that all others are negligible in this kind of study.
Drat, post ninja'd. I should have used Google; would have been faster than clicking through and tallying those profiles.
See my post above.
A search of OK Cupid in the 250 mile radius of India's capital maxes the limit number of results pages. Using the most efficient method directly available to glean the numbers comparing any further to an American major city is out of the question since they both easi;ly max the limit of results (which is 1000) within the city itself. To get better figures I'd require a more lateral approach.
http://www.okcupid.com/match?fromWhoOnline=1&filter4=1%2c1&mygender=&filter1=0%2c63&low=991&sort_type=0&locid=690995&filter3=3%2c2...
http://www.okcupid.com/match?fromWhoOnline=1&filter4=1%2c1&mygender=&filter1=0%2c63&low=991&sort_type=0&locid=4335338&filter3=3%2c...
On the whole I think the Indian population is well represented by these figures, well enough to generate a small anomaly in the stats, I'd believe, but it doesn't seem to be the case here anyway.
Ok, mature gay white men definitely rings a bell, a demographic which is both generationally and socioculturally characterised by 'properness', it makes some sense that we might find this group's written manner outlying in a netspeak population.
Theist badge qualifier : Gnostic/Philosophical Panentheist
www.mathematicianspictures.com
O.K. Cupid is quite enormous in scope, so maxing out search results in any area isn't terribly odd.
However:
http://www.giveyourhandabreak.com/sites/okcupid/
The site is mostly white people, followed by blacks, then Latinos, then Asians, and trailing behind them Indians and the rest. Indians do come out the most cultured of the races for writing level, but not by enough of a margin to make up for the smaller number (and the average there is still well under a ninth grade level).
The American born Asians might bring up the base-line a bit (being somewhere between 20% - 40%), but I think the mature gay white man effect is probably the best bet for explaining what actually raises the level. My guess is that an educated gay protestant or Catholic is likely to turn to Buddhism if not become an atheist (due to social pressure to have a religion). On the West coast Buddhism seems particularly common among the gay culture from my limited experience- it's like a socially acceptable way of leaving religion (or, at least leaving a religion you're dissatisfied with).
Now, as to why gays are more cultured? *shrug* I don't think sexual attraction has any real bearing on intellect, so my guess is that the stupid ones stay in the closet and become those ultra-macho or evangelical homophobes, so the only visible examples have been refined to the most intelligent, educated, and progressive of the batch- but that's just a guess, and I don't have any statistics that would back that up right now.
lol wut?
You're superior to Christians, that's what.
Vote for Democrats to save us all from the anti-American Republican party!
Please become a Patron of Brian Sapient
I want to become a God's children.
Yes, I want to be Saved.
Is anyone surprised by this? I'm not.
Try [ctrl-s]
While the liittle bit that got posted here is interesting there are some other interesting things that showed up. I mean I take this whole "study" with a huge grain of salt but it is interesting.
It does only show of those that are into online dating not a broad spectrum of the religion etc. Is that even relevant? who knows unless a comprihensive study is done. All im saying is that this is like doing a survey and then only going to the middle and upper class (those that can addord internet) with a survey about online dating (not everyone is interested in online dating, married various other reasons). It is not truely representative unless all class groups races and other relevant things atre taken into account with the correct ratios.
Other than that it is interesting and not that suprising, except for a few things. I mean why on earth would you list soul food on anything?!
Whatever goes upon two legs is an enemy.
Whatever goes upon four legs, or has wings, is a friend.
No animal shall wear clothes.
No animal shall sleep in a bed.
No animal shall drink alcohol.
No animal shall kill any other animal.
All animals are equal.
Tapey,
It's not just a dating site- there are married people on there. Statistics could be normalized for that, but it's hard to imagine the relevance.
Socioeconomic normalization might be more practical, but there's a strong known correlation with poverty, lack of education, and religiosity already.
I have no doubt there are married people there dating site or not. That wasn't the point, certain types of people join these type of sites others don't. I don't mean to be rude or anything but the internet isn't generally peoples first choice but even that is irrelevant. It is just not a true representation of the population of, in this case the world, its not just americans on there. You can tweek the results all you want but you will not find anyone puting academic creditablity on this. Yes its a fun little project and no doubt representative for that site but the fact is it is done under far from ideal conditions when looking for the truth.
What this says to me is of the users of okcupid or whatever these are the results that apply to this population of users. I wouldn't push the results past that. From these results I couldn't in good faith say on average atheists have better writing skills, cleverer or whatever. Really because also you could say it means people of these religious groups put more effort into there profiles than other religious groups due to there beliefs. Your beliefs affect everything you do it wouldn't suprise me if it affected how you went about writing about yourself because naturally your beliefs affect how you view yourself. There are just to many loose ends to take this seriously. Why someones profile looks the way it does and the language they use is more likely how they wish to be percieved after all you are selling yourself with your profile. Someone wishes to be seen as smart they are not likely to go about writing in crap english, someone wanting to promote there hobbies won't put as much effort in writing as well as the person wanting to look smart. To many variables in this. Think about it this way writing a formal email to company to complain and writing an email to your friend, you use comlpetely differant langauge and your writing style changes, its the same with writing an internet profile. If you want a study with these results to mean something you need somehing better than an online profile, get 2000 people and get them to write a a bunch of things on a whole range of topics for a whole range of differant functions, in a whole range of styles with all the nessicary representative factors then I will take it seriously.
Fun little project they have going and I enjoyed it, but its clear they did it as a fun little thing, you just cannot push it past that in my opinion.
Whatever goes upon two legs is an enemy.
Whatever goes upon four legs, or has wings, is a friend.
No animal shall wear clothes.
No animal shall sleep in a bed.
No animal shall drink alcohol.
No animal shall kill any other animal.
All animals are equal.
evidently there are some exceptions*cough*
(I can understand using "they" as a singular pronoun- it is very wrong, though it is a common affront to English- but using "there" (which is neither a pronoun nor possessive, but relates to location) instead of the otherwise incorrect "their" (which is at least a possessive pronoun) when you should be using "one's" or in the very least "his or her" or even (in a sexist manner) simply "his" is quite frankly vulgar. You might want to look into correcting that habit; you make that mistake consistently, which indicates that it probably isn't a typo.)
Aside from that, causation is irrelevant to what this analysis actually says. The computational processes of the analysis are likely perfectly valid, and it says exactly what it says to the group it refers to- wherein you are prepared to complain about the sample, you'd do better not to, because this is no more skewed than a survey based on phone calls, or one based on collecting random people from the street (I would expect it to be less skewed because participation is *involuntary* and completely blind).
It says the atheists are better writers- not *why* they are better writers. If they are better writers because they care more, that's irrelevant- they are still better writers.
The sample is more than sufficient to compensate for any small bias towards the "single internet users" to make a very probable general statement, but regardless of that you might note that the very demographic you criticize as being too narrow is becoming the majority of the population in North America (it's already potentially over 40% of adults, and teen usage blows that out of the water).
One can not just say that the internet is for the middle and upper class anymore- the ubiquity of social networking has extended into every relevant demographic short of Amish (a group I would argue aren't very relevant to national demographics) and people with either at least one foot in the grave or still in diapers (the latter probably unable to participate due to being unable to write).
Within this sample, if one wanted to, one could normalize for any relevant factor (such as for those who report being married, although this might be a little difficult given the overwhelming number of married people on the site who are also lying about it)- what you have not demonstrated is that any of the major biases that we might normalize for are relevant.
The site contains a higher percentage of atheists than the population as a whole, and yet because the analysis is comparative between religious groups, that isn't relevant.
The site may contain higher proportions of single users as compared to married users, but unless you have a fairly good reason to believe that the writing proficiencies of atheists tank when they get married, and that the writing proficiencies of theists enjoy a drastic improvement upon the event of marriage, there's really no apparent reason to consider that.
All this analysis is really doing is restating the obvious- in yet another statistically significant way that adds to the existent mountain of evidence.
Did we need more evidence to prove this? No. Would this be enough in itself? Given the sheer size and spread across the population it represents- yes, it should be. Until something better comes along that says otherwise, this is perfectly relevant.
I'd love to see something like this done with Facebook (encompassing something like half of all people who use the internet in the states); but the scope here is really sufficient to say something already.
You know what would be funny? If all the atheists were from America or Canada, and all the religious were from Portugal or Lebanon
Everyone likes to have a reason to toot their horn in some way or another, but the data cited here does not give us a good reason. To even call it a reason, good or bad, is being too charitable. It is obviously true that one is better able to express themselves and to find and refine ideas when one has a better appreciation of language and its many shades of meaning, but the degree to which one writes well cannot be determined by the Coleman-Liau Index.
The author of this Political Math Blog entry has already shown this to be the case. He created two sample profile texts that used language so simple that it seems like an attempt to copy the prose style of Isaac Asimov. The first one professed atheism in the form of metaphysical naturalism and the other professed Christianity. The Coleman-Liau Index reported them as being five reading levels apart, even though they would be put at about the same reading level by any reasonable person.
There was some discussion in this topic earlier about the sample size used. The sample size is large enough that a proper interpretation would tell us something very interesting, but the reports of the Coleman-Liau Index cannot be trusted as a proper interpretation, and because those reports are all we were given, we have no way to derive a sound conclusion from any of this.
Stultior stulto fuisti, qui tabellis crederes!
Having gotten some sleep, I read through the comments on the Political Math Blog entry. It turns out that I should have been more careful, because the author used the Gunning-Fog Index, not the Coleman-Liau Index. As such, my previous reply was premature.
Aside from having been tired, another reason I trusted the results so readily is because they were completely reasonable to me, given my background knowledge. I have been using StyleWriter for years because it gives me the results of many readability tests and points to areas where a text can be improved. Ever since I started using such software, I have learned that readability tests are useful but easy to trick and they sometimes give results that are downright crazy.
Stultior stulto fuisti, qui tabellis crederes!
I knew that from tech writing class. We were to bring in a magazine article that we had analyzed and then share in class. I choose one that was short, with short words and sentences from a business weekly. Scored like 3rd grade. Only a couple people in class understood what it was talking about (x.25 networks - very trendy at the time). Another article was about electricity generation. Since it had electricity (5 syllables!) in about every other sentence, it scored at graduate level, though everyone in class understood it. Haven't trusted those type of analysis since.
-- I feel so much better since I stopped trying to believe.
"We are entitled to our own opinions. We're not entitled to our own facts"- Al Franken
"If death isn't sweet oblivion, I will be severely disappointed" - Ruth M.
I have never been in a writing class that did such a comparison, but I know exactly what you are talking about. For some strange reason, I like to experiment with my writing and see how the readability tests react to the result.
One time, I took a chapter of "The System of Nature" and rewrote it according to the rules of Ogden's Basic English (which limits you to 850 words and the simplest sentence structures) and dropped the reading level about ten points, and the changes I made were almost completely artificial.
Later, I rewrote the same text according to the rules of English Prime (which forbids the use of the verb 'to be' in all its forms) and the grade level raised by quite a bit, even though it was about as easy to read as it was before. Sometimes a simple change in punctuation can produce a huge change in reading level, such as switching out a semicolon for a period and moving the 'however' to a later part of the sentence, which is a completely artificial change.
Because I have done many experiments like these to hone my skills, I have come to understand that readability tests cannot be trusted to tell you the true readability of a text, let alone the writing proficiency of the author. I mean, take Isaac Asimov for an example. It takes an extraordinary proficiency with language to express things as simply and clearly as he did, but the people who take the tests seriously as a measure of proficiency would conclude that he can only barely escape the charge of illiteracy. If you know how the tests react to the various writing styles (e.g., they vomit with Hawthorne-like writing, they smile condescendingly to Asimov-like writing, etc.), then you can gauge their reaction to see if you are coming close to the writing style you like the best, but the grade level numbers themselves are garbage.
Stultior stulto fuisti, qui tabellis crederes!
Good thing I have a university degree, or I wouldn't be able to follow this sentence.
What pissed me off the most was when my sons were in school and they were using these assessments to determine what the children would be able to read at their grade level. When I finally wised up, I started getting older children's literature from the library - my oldest went through a stage of reading every Hardy Boys mystery. My middle son learned to read by reading The Lord of the Rings trilogy. He was an excellent reader when he finished it. My youngest liked romances. yech. My husband says he learned to read from Kipling's stories. I liked fairy tales. Find a genre the child likes, get them books pre-1950 in that genre, and s/he will learn to read.
-- I feel so much better since I stopped trying to believe.
"We are entitled to our own opinions. We're not entitled to our own facts"- Al Franken
"If death isn't sweet oblivion, I will be severely disappointed" - Ruth M.
In Northern Ireland, members of terrorist organisations on both sides of the Protestant/Catholic devide can be counted among the "not serious". Yet, when it comes to the atrocities they commit, they are more than happy to attribute them to religion.