Christian Round Winners

Yesterday I revealed true authorship for the Turing Test entries and congratulated the most convincing Christians in the Atheist round.  Now it’s time to look at the results of the Christian round.  Here’s what the Christian respondents (sample size = 93) made of the responses.

Click image for larger view

When participants were ranked according to their average score, only two fell into the ‘definitively Christian’ range (i.e. an average score between two and three, where a vote for Atheist was assigned zero points, a vote for ‘Lean Atheist was assigned one point, etc).  And the winners were:

  1. Entry #3 (Kat – Christian) with 76% of respondents convinced she was Christian
  2. Entry #5 (Leah – Atheist) with 78% of respondents convinced I was Christian
*Kat places first, although I have a higher percent of people fooled because more people rated her as very likely to be Christian, pulling up her average score

The Christians managed to catch out two fakers (people with average scores between zero and one, placing them squarely in the Atheist to Lean Atheist category).  The culprits were:

  1. Entry #1 (Peter – Atheist) with only 23% convinced
  2. Entry #4 (Michael – Atheist) with 27%

Unlike in the last round, where atheists and Christians mostly agreed in their assessments, the Christian responses provoked different responses.  Here’s what the atheists (sample size: 85) concluded.

Click image for larger version

Not a single entry cleared the average > 2 cutoff I had set.  So, that noted, here are the top four scorers among the atheists according to a different scoring scheme I came up with (about which, more later).

  1. Entry #5 (Leah – Atheist) with 74%
  2. Entry #2 (Alex – Atheist Christian) with 66%
  3. Entry #11 (Lorilyn – Christian) with 60%
  4. Entry #6 (Adam – Atheist) with 62%

*To sum up the alternate scoring system briefly, I’m still computing an average, but the ‘Christian’ responses are weighted more heavily than the ‘Lean Christian’ ones and the ‘Atheist’ responses count for more than the ‘Lean Atheist’ ones.


Even using the weighted average score, not a single entry was identified by the atheists as definitively atheist.  Also interesting, all the people the atheists chose as Christian were either atheists or somewhat liberal sounding Christians (cf. Lorilyn claiming that other denominations might offer the different paths to truth).

More charts/demographic breakdowns/etc to come.  For now, the full rankings are under the jump.

Entries are listed from most christian average score to least christian, so the percentages may not be strictly decreasing.  The entries are ranked according to the Christian answers, but the percent of atheists convinced is given in parenthesis and italics at the end of each row.

  1. Entry #3 (Kat – Christian) with 76% (47)
  2. Entry #5 (Leah – Atheist) with 78% (74)
  3. Entry #12 (Tristyn – Christian) with 71% (61)
  4. Entry #6 (Adam – Atheist) with 66% (62)
  5. Entry #8 (Nathan – Christian) with 60% (54)
  6. Entry #10 (Brian – Christian) with 62% (41)
  7. Entry #13 (Chris – Christian) with 61% (61)
  8. Entry #9 (Anonymous – Christian) with 56% (46)
  9. Entry #11 (Lorilyn – Christian) with 52% (60)
  10. Entry #14 (Charles- Atheist) with 53% (48)
  11. Entry #15 (Bo – Atheist) with 49% (59)
  12. Entry #2 (Alex – Atheist Christian) with 43% (66)
  13. Entry #7 (Jeannine – Christian) with 30% (53)
  14. Entry #1 (Peter – Atheist) with 23% (48)
  15. Entry #4 (Michael – Atheist) with 27% (40)

About Leah Libresco

Leah Anthony Libresco graduated from Yale in 2011. She works as a statistician for a school in Washington D.C. by day, and by night writes for Patheos about theology, philosophy, and math at www.patheos.com/blogs/unequallyyoked. She was received into the Catholic Church in November 2012."

  • http://www.blogger.com/profile/07592851515207841008 Christina

    "Typo" notice…The charts are identical.

  • Anonymous

    Not counting Alex (again), Christians 0.59, Atheists 0.48. Big win for Christians, I think. Since many Christians (including myself) think Alex is a nonChristian, including him with the atheists gives them 0.47. I'm surprised it was this close since, as I said before, I found it surprisingly easy, being 13/15. The atheists were hard and I was only 6/15.

  • http://www.blogger.com/profile/08572976822786862149 Darwin

    Huh. Again, really interesting to see how this shook out.One thing that seems interesting is that there seems to have been a lot of uncertainty among Christian respondents. (I certainly felt a lot, I only classified five pieces as definitely Christian or definitely atheist.) All Christians except Jeannine scored above 50% with Christians, but no one got overwhelming numbers of votes. (And three out of seven atheists scored above 50% with Christians as well.)It would be interesting if there was some way to filter by type of Christian portrayed and type of Christian judging. Some of the identifications I got wrong or had trouble with were cases whether the type of Christian portrayed by the answer is one I strongly disagree with or have little agreement with.

  • http://whatloveteaches.blogspot.com Slow Learner

    One possible explanation for a skew that occurs to me is that I got the impression most of the Christians writing are Catholics.Given that my experience of Christianity is primarily with Anglicanism and Evangelicals, I think that both my ability to accurately spot Catholics, and my ability to write as one, could well be much worse than my ability to imitate and spot (Poe's Law notwithstanding) Evangelicals.On the flipside, we atheists are a quite disparate bunch; more so than Christianity as a whole, arguably, but certainly more so than Catholics. As such I don't think there are as many quirks of language or argumentation that would enable Atheists to accurately peg one another.I'm sure there's a lot more to it, of course – I was fairly sure that I'd spotted Adam purely because I've read Daylight Atheism extensively, and I know his writing 'voice', if you will; but that's the first explanation which occurs to me for the discrepancy in scores.

  • http://www.blogger.com/profile/16496144988509668275 Leah

    Yikes, good catch Christina. The Christian graph is now accurate.

  • http://www.blogger.com/profile/01210899232931814200 JC

    I'm enjoying that according to your statistics, your boyfriend was the more convincing atheist, but you were the more convincing Christian. Tie-breaker? Which of you won the Catholic vote for more convincing Christian? :)

  • http://www.blogger.com/profile/08572976822786862149 Darwin

    On the flipside, we atheists are a quite disparate bunch; more so than Christianity as a whole, arguably, but certainly more so than Catholics. As such I don't think there are as many quirks of language or argumentation that would enable Atheists to accurately peg one another.Heh. In a classic example of how people with different viewpoints look at each others groups, I had been thinking pretty much the exact opposite. My basic theory was that the Christians generally scored very high when pretending to be atheists because atheism is a fairly simple idea (God doesn't exist) and thus easy for someone used to working with religious and philosophical concepts to fake. When one gets into judging the Christian side, on the other hand, there are a whole lot of different ways one can believe in God (even after we restrict things down to only Christianity) so I figured this explained the overall low positive scores (with Christian judges most successfully identifying those Christians most like them) but the overall complexity would also make it difficult to fake from the outside (thus the overall underperformance of atheists in passing for Christian.)

  • http://www.blogger.com/profile/07592851515207841008 Christina

    Another thing to consider is the shift that happens over the course of the test. When I took the Christian one I noticed that the more I read the less I was able to distinguish. (and I didn't help with accounting for this by then reading them out of order).If you look at the charts, there appears to be a gradual shift towards "not Christian". In the Atheist questions there appears to be a gradual shift towards "is Christian".

  • http://www.blogger.com/profile/07592851515207841008 Christina

    And have you considered a site like this (so we can all play)? http://www-958.ibm.com/software/data/cognos/manyeyes/

  • http://www.blogger.com/profile/00681934865643964687 JSA

    This just shows that Leah isn't really an atheist. She just pretends to be one on this blog, but in her heart of hearts, she's secretly a Christian.

  • http://www.blogger.com/profile/05377685250633624137 Tristyn Bloom

    I find it amazing that Slow Learner would say atheists are more diverse than Christians, or even Catholics. I would've said precisely the opposite and thought this experiment bore that out! Granted, as one of the Christian participants myself, I felt some obligation not only to speak honestly, but to attempt to be convincing– not that I was actually a Christian, but that Christianity was good and true. When you're tailoring something for a particular audience in that way, a lot of the quirks get ironed out. Given the diversity we saw despite that factor weighing on all of us, I'd say we're all rather mad.

  • http://www.blogger.com/profile/10019240793982424774 Christian H

    "As such I don't think there are as many quirks of language or argumentation that would enable Atheists to accurately peg one another."I'd find atheist prose hard to fake, but I've found that atheist sites and commenters (with the exception of Leah) sound very similar. Christians do this, too (count how often "discern" is used in conversation), so it's not especially a judgement. I just don't think it's true that there are no quirks. It's not always about vocabulary, of course, but also about tone and readerly/writerly assumptions.

  • Anonymous

    I was a Christian respondent. On my first read thru I pencilled down Leah's answer at VLC, but on a second reading I had reservations and NEARLY put it VA but settled for VC. When I was deciding, the phrase about "keeping our existence from being flat" raised some suspicions with me. My second suspicion was the phrase "clinched my conversion". I don't know why, but that didn't seem like something a Christian would have written.In spite of these reservations I still put Leah down as a Christian so well done Leah!

  • http://www.blogger.com/profile/07345409714343072179 Ray

    It seems like the same people get high scores on both the atheist and christian rounds for the most part. Maybe some people are just better writers and therefore come off as more plausible on both rounds.Given the small number of writers, though, I doubt this says all that much about atheists and Christians in general.

  • Antrobus

    (Disclosure: atheist). Despite finding, from personal experience, that expressing doubt tends to give the impression of a more sincerely-held, deeply-examined opinion, I somehow supposed that this was my own private insight. Surprise, surprise, it seems to have been the single most favoured ruse, on both sides.I thought that Catholics were liable to be over-represented among the Christian subjects, because the blog-owner has a Catholic boyfriend, but it didn't seem to help my detection rate. As happened to one above commenter, Leah-as-Christian managed to completely divide my judgment (but I teetered between VLC and VLA…and I made the wrong call).Blunt statements have become fashionable in atheist literature over the past decade, and atheists who write this way are easily found on blogs and BBs. I have seen plenty of sincerely-held claims that religions pander to the worst parts of human behaviour, that all religious people are scared into morality by Hell/bribed into it by Heaven, that organised religion is a financial exploit run by clergy who worship only the profit motive, etc. So I was more inclined to flag fundamentalist Christian statements as the work of an imposter, than I was to do the same for blunt or simplistic statements about religion made by someone claiming to be an atheist.

  • http://www.blogger.com/profile/11111405959451703182 PhysicistDave

    Slow Learner wrote:>Given that my experience of Christianity is primarily with Anglicanism and Evangelicals, I think that both my ability to accurately spot Catholics, and my ability to write as one, could well be much worse than my ability to imitate and spot (Poe's Law notwithstanding) Evangelicals.Yeah, I’ve known *lots* of Evangelicals who are sincerely convinced that all atheists are dogmatic materialists, intransigent moral relativists (if not anti-moralists), etc. Since attempts to impersonate atheists by people with such beliefs would probably betray those false assumptions, and since atheists do not suffer from those misconceptions, it would be pretty easy to spot one of them trying to impersonate an atheist.I think part of the difficulty of detecting the fakes in Leah’s exercise was that she chose fairly well-educated Christians, none of whom seem to be naively dogmatic Evangelicals, and that her atheists, when impersonating Christians, seemed to assume that it was this sort of Christian that they were trying to impersonate. With a different choice of participants (e.g., poorly educated and dogmatic Evangelicals, who do make up a lot of the Christians in this country), the exercise probably would have turned out differently.Incidentally, I myself once unintentionally passed the Christian Turing test on another Web forum when I posted a satirical post making fun of Christian beliefs, a post which I thought was an obvious reductio ad absurdum. One of the atheist women on the forum did not realize my post was satirical, was quite convinced that I was a Christian, started attacking me, and has hated me ever since (even more so, I think, when she finally realized it was a satire that I had assumed that everyone would grasp). Dave

  • Gilbert

    Some random points:1. Both sides have people who can imitate the other side plausibly and people who can't. The test wasn't (and wasn't meant to be) rigorous enough to establish actual proportions, but it pretty much does establish, that neither side is inherently unable to take the other sides viewpoint.2. Kat and Leah are the clear personal winners of both rounds. It might be a statistical fluke, but maybe woman are better at intellectual mimicry. 3. One could get basically the same ranking by taking all answers as sincere and then ranking by the quality of their arguments. That is actually not a bad algorithm, because the premise of the test is one side not understanding the other side and therefore being unable to give good arguments for it. Which is why it so easyly cought Peter and Michael, who acted the part of their tribes stereotype of the dumb Christian, rather then of an actual Christian.The problem is, that what we're actually after (the difference in understanding the other side) is probably dwarfed by the generic quality of argument (some combination of the time spent writing the arguments, intelligence, rhetoric skills, &c;). This basically makes the test into a debating competition. If anyone repeats this, it might be a good idea to have a "explain two interesting ideas unrelated to religion"-question and then to randomly assign the two explanations to the two anwer sets. Also the "honest" side shouln't be bound to honesty. For example, Jeannine could clearly have given much better answers then admitting to the irrationality of her faith.

  • http://www.blogger.com/profile/13286545868199457106 Oran

    This is strange. I'm an atheist, and I was not raised Christian, and yet looking at my own guesses, I was better at detecting Christians than atheists. I think this reflects two things:(1) Christians and religious people in general do have a greater diversity of ideas and feelings while atheism all basically boils down to the same short list of points no matter which religion or religions they are primarily rejecting. I think they are good points, but they are not that difficult to grasp and emulate [not to devalue some of the excellent writing that was done emulating it].(2) More importantly, a couple of atheists perhaps did reflect a simplistic stereotype. I believe that stereotype is an accurate depiction of a large number of people, but based the quality of responses in the atheist round and the make-up of the community I did not expect those people to be represented.(3) On the other hand, there was one real Christian I thought was a parody, but it's good to see that Poe's law didn't creep into this activity too heavily. I think the way this challenge was framed kept Poe's law out of the mix in a way that doesn't happen in a less structured conversation.

  • http://www.blogger.com/profile/11111405959451703182 PhysicistDave

    Gilbert wrote:>Which is why it so easyly cought Peter and Michael, who acted the part of their tribes stereotype of the dumb Christian, rather then of an actual Christian.But, of course, some actual Christians are indeed “dumb Christians”: anyone who doubts this has little experience attending evangelical churches. (No, I’m not saying they are all dumb, but *some* certainly are! Anyone who denies this did not attend the Southern Baptist church my parents forced me to attend as a kid.)So, if Leah had included some of those “dumb Christians,” many of those who were judging might have assumed they were atheists making a poor attempt to impersonate Christians.The results of the test were really an interaction between what the judges were expecting and what the contestants assumed the judges were expecting. As it happened, the contestants were well matched to the judges expectations (and the impersonators accurately guessed the perspective of the non-impersonators) which made it impossible to accurately guess who were real and who were fakes.A different group of contestants more representative of real-world Christians, different expectations on the part of judges and contestants, etc., and the contest obviously would have turned out in a different way.In fact, by the standards of most evangelical Christians I know, I’d say that *none* of the contestants, atheists or “Christians” were “real Christians” – i.e., what many evangelicals consider to be true-blue, born-again, Bible-believing Christians.Of course, since I’m not a Christian and attach no positive value to the word (quite the contrary!), I don’t really care who counts as a “true” Christian.Dave

  • Gilbert

    @Dave>But, of course, some actual Christians are indeed “dumb Christians”Im a little amused, because in this case the scare quote hedge actually diminishes the truth of the statement. Of course some Christians are dumb. Some adherants of pretty much everything are dumb. If you doubt it for your tribe, take a look at reddit, then weep. But that's not what I was talking about. The new atheist stereotype of the dumb Christian is – like the standard issue atheist of a Chick-trakt – an inversion of a collective self-image. It is dumb, because the self-image includes intellectual superiority. And dumbness is what cought it out here, just like it would have cought out the Chick-atheist. Dumbness is, however, not the only stereotypical property of the "dumb Christian" or of the Chick-atheist.For example, I have no reason to doubt the abundance of fideists at your childhood parish. Nor to doubt the existence of dumb fideists in that parish, because fideists come at all levels of intelligence. But I do doubt that parish having any significant number of "dumb Christians". Conservative evangelical fundamentalists would not have written Peters answer to the third question or Michaels answer to the last one. Those answers mix in liberal Christians anti-intellectualisms very rarely combined with the conservative ones. So the stereotyping goes beond dumbness, but in this test the dumbness is clearly what did it in.Now my point was, that in this test the heuristic of sorting by quality of argument (which includes, but isn't identical to, intelligence) was (a) plausible a priori, (b) probably the actual heuristic and (c) more susceptible to the confounder of generic quality of argument then to the specific quality of imitation.I can only guess for (b), but (a) and (c) would also be true in a perfectly representative sample. In that case, too, the strategy would have penalised generic bad arguers and people incapable of understanding the other side, thus being biased against fakers. Imitating a dumb opponent would, therefore, still be a bad strategy. So, assuming no atheists fell for the bad strategy again, it would be basically a test of generic quality of argument. It's anyones guess how that might turn out (though anyone thinking their answer obvious is obviously deluded), but the main point remains: the answer would tell us more about generic quality of argument than about understanding of each others position. A real religious turing test would, therefore, have to control for generic quality of argument.Btw, I'm trying to interpret, not to critisize this test, which was never meant to settle the question.

  • Marcel Kincaid

    "Peter and Michael, who acted the part of their tribes stereotype of the dumb Christian, rather then of an actual Christian.""their tribes stereotype"? "an actual Christian"? Now that is dumb, dishonest, and hypocritical.

  • http://www.blogger.com/profile/08395703772492059721 Benjamin Baxter

    "You attacked reason. It's bad theology." — Father Brown

  • http://www.blogger.com/profile/11111405959451703182 PhysicistDave

    Gilbert wrote to me:> But that's not what I was talking about. The new atheist stereotype of the dumb Christian is – like the standard issue atheist of a Chick-trakt – an inversion of a collective self-image. It is dumb, because the self-image includes intellectual superiority.Actually, the usual atheists’ stereotype of Christians *greatly* under-estimates the stupidity of many of the dumb Christians I have known personally. For example, I had a Sunday-school teacher who would declare that the Bible is true because it is the Word of God and we know that it is the Word of God because it says so and we know the Bible is true. But that is not what made him really dumb. To demonstrate his circular reasoning to him, we would keep asking the questions again and again so that he kept going around in circles: How do we know the Bible is true? How do we know it is the Word of God? But how do we know the Bible is true? Etc.And he’d just keep going around in logical circles, grinning at us with this enormous confidence that he was really proving his point, and not noticing that we were all just shaking our heads at each other. I suspect that everyone in that Sunday-school class became an atheist, thanks to poor, incredibly stupid Glen Manion!Now *that* is a dumb Christian, and most atheists have no idea how many dumb Christians there are just like that, since, unlike me, most atheists have not encountered their share of Glen Manions!Most human beings, atheists or not, are indeed intellectually superior to dumb Glen. Not a stereotype, just reality.Gil also wrote:> For example, I have no reason to doubt the abundance of fideists at your childhood parish.Now, you’re betraying your own ignorance of Christianity, Gil. Most denominations do not have parishes. You must be one of those non-Christian Catholics, eh? I did have another Sunday-school teacher who not only insisted that all Catholics were going to Hell (not that uncommon of a view in our church) but also that they should all be imprisoned if not executed (I think this was to protect the children from evil Catholic teachings).See, another dumb one. And, yes, how we enjoyed encouraging him to expound on those views just to make sure everyone knew how dumb a Christian could be. (Yes, we were mean, but do you have any idea how boring a Southern Baptist Sunday School is? No, I thought not!)Gil also wrote:> Nor to doubt the existence of dumb fideists in that parish, because fideists come at all levels of intelligence.Well, not too many “fideists” among top scientists if you look at the Larson-Witham study.Anyway, I think you agree with me that if the atheists in the test had expected the Christians to be like my former Sunday-school teachers, or if the Christians had really been like my former Sunday-school teachers, the test would have turned out differently.Leah’s circle of acquaintances seems to over-represent liberal, intellectual Catholics, which is hardly representative of the country at large.Dave

  • http://prodigalnomore.wordpress.com/ Benjamin Baxter

    Maybe this is because Christianity caters to everyone, Dave, not just those clever enough to hide in abstraction.It's too bad, too, that she includes people who hark to Aquinas rather than Paley. It almost seems like, by her acquaintances, she's giving Christianity a fair shot.http://prodigalnomore.wordpress.com/

  • Gilbert

    @DaveYes, upon looking it up I see I made mistake, tough it was about language rather then Christianity. I'm in Germany and the dictionary translation of "church" ("Kirche") applies almost exclusively to Christian denominations or buildings, not to the local religious community. That is almost exclusively called "Gemeinde" (apparently not quite as equivalent to "parish" as I tought) not only if it is protestant but even if it is Jewish, Muslim or even Buddhist. It's sometimes hard to remember languages differ not only in the words but also in the borders between words. And just to preemt your next tought: Yep, over here fundie Evangelicals are in the low single digits population-percentage wise. Still they are around and noticible if you care for such things.Next, yeah your childhood expsure to Christianity seems to have been rather traumatic. I see how this is peronal for you. But that is a special case. Those people are fairly embarassing even for most Evangelicals. Yes, they do exist, but no, they are no more representative of Christianity then the sample we have seen here. Now for the flip-side: Aggressive New Atheism style Atheists weren't all that common in my childhood. But liberal to green anti-Catholic Agnostics were. (Actually on the definitions Atheists now typically insist on you would have to count them as Atheist but I think we can skip the fruitless fight on semantics.) And I can tell you they match your childhood experience of stupidity. One of them (a school teacher) spent a whole week insisting to my class, that modern technology allowed melting a ton of iron out of half a ton of ore. In my more self-critical moments i wonder how much of my reactionary outlook on life is actually based on defience, becaus the rigidly dogmatic idiots I suffered almost universally identified as liberal to Greenish. Also there are still actual Stalinists around, so Atheists can easyly match the evil of your would-be iprisioner of Catholics. So yes, in a fully representative sample you would have the kind of people you suffered. But that would be compensated by the people I suffered.As for the high profile scientists: I can think of loads of confounders and reasons for a reversed causality. But actually I don't care to have this paricular pissing contest. So I'll just note, that for an actual incompatibility you would have to find zero rather then "not many".

  • http://www.blogger.com/profile/12714581082612893449 Colin

    Hey, Gilbert, shoot me an email @mesmeridicus@gmaildotcomif you have the time.Thanks

  • http://www.blogger.com/profile/08016230732925516069 Gilbert

    @ColinDone.In case anyone else wants to contact me, there is now an email address at the profile you reach by clicking on my nick.


CLOSE | X

HIDE | X