Christian Round Winners [Turing 2012]

As was the case yesterday, you may want to keep this year’s answer key and the list of Christian answers at hand while you look through the results from round two.

The top three scorers in round two were all sincere Christians.  Eve Tushnet took home ‘Most Christian Christian’ with 88% of Christian respondents rating her entry number nine as ‘Likely Christian’ or ‘Very Likely Christian.’  The next two slots were won by Elliot (entry five, 86%) and Gilbert (entry six, 85%).

So before I tell you how the shamming atheists did, let me offer a big congratulations to Elliot and Gilbert, whose entries were in the top three for honesty in both rounds.  Since I know Elliot in real life, I suppose I’ll stand him a drink next time I see him, and the same goes for Gilbert, if he visits DC.

The next entry down was the first successful atheist: Guy, who wrote entry number four and got 68% of the Christians to think he was Christian.  Following him was Matt DeStephano of Soul Sprawl (entry two, 60%).  I’m inclined to rate these two as the only winners among the atheists, since the next atheist in the listings only scraped a 38%.  Let’s check the chart.

The atheist answers are highlighted in blue. The first bar is the percent of Christian respondents rating the entry as “Very Likely Christian” and the second is the percent rating it “Likely Christian.” The entries are sorted descending by the sum of the ratings.

The two Christians to get trapped down in the bottom with the unsuccessful atheists were Christian H of The Thinking Grounds (entry twelve, 35%) and Sweet Tea (entry eleven, 21%).  I wonder to what extent they scored low because the Christian part of my readership tends to be fairly conservative.

Meanwhile, Jacob of The Thoughtful Atheist (entry one, 32%) was identified as a fake by both groups.  Last year’s quasi-agnostic also scored low in both his rounds.

The full stats appear in the I’d-love-suggestions-on-cleaning-it-up graph below. Now. the answers are sorted according to the percent of Christian readers that rated the entry “Very Likely Christian” so the rankings have shifted. The atheist participants are highlighed by setting their grayscale gradient to run backwards, so you can think of it as the more accurate the category for that essay, the darker the datapoint.

About Leah Libresco

Leah Anthony Libresco graduated from Yale in 2011. She works as an Editorial Assistant at The American Conservative 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://not-atamelion.blogspot.com Michael H.

    On visual representation: if you kept the color distinction between atheists and Christians in the all-together graph (i.e. create a gradient that starts with one color and goes to a sufficiently distinct color), it’d be more readily legible, I think…

  • Maiki

    “The Christian answers are highlighted in blue.”

    I think this is backwards.

    • leahlibresco

      Merci, fixed that and a couple other errors in that paragraph.

  • deiseach

    I have to apologise to Sweet Tea, because I was convinced he was an atheist impostor due to the word choice in this part of his answer:

    “However, my religion demands the use of genitalia only in sexual acts with the potential for reproduction”

    “Ha, ha!” I said to myself. “A good try – but not good enough, Mr. Bond! No real Catholic would use the term ‘genitalia’ in this context, because it’s not about only using your bits in the approved fashion.”

    I suppose I’ll just get on with wiping the egg off my face, now :-)

    • Sweet Tea

      Eh, your point that it’s an badly precise wording is quite true, my apologies for being misleading.
      ….I still can’t really believe that, as bad an atheist as I am when I’m emulating one, that I’m the worst Christian at seeming Christian.

  • blotonthelandscape

    Hello, really interesting experiment! Re. data viz, in my professional opinion, they grayscale is offputting, and the colours are all jumbled up. I’d recommend a more diverse colour scheme, and making sure the answer groups occur on the same parts of the graph for each individual (i.e. Very Christian/Likely Christian/Likely Atheist/Very Atheist).

    Do you use Tableau? Free and easy, with a lot of options for making neat visualisations.
    (http://www.tableausoftware.com/products/public). If you’d like, send me the data and I’ll do a quick re-jig for ya, free of charge ofc. ;)

    • http://delphipsmith.livejournal.com Delphi Psmith

      Tableau rocks. I had to use it in a class last year and loved it (anything that get me voluntarily playing with statistical data HAS to be good!). If you plan to do lots of playing with data, I highly recommend it.

    • blotonthelandscape

      Oh yeah, and FORTHESAKEOFALLTHATISHOLYANDKITTENS label your axes. /statsrant

    • leahlibresco

      I’m sending you the spreadsheets. I’m quite interested to see what you do, thanks for the offer!

  • http://prodigalnomore.wordpress.com/ The Ubiquitous

    This was a more interesting, difficult, and rewarding Turing Test. Thanks a bunch, Leah! (Oh, and did you get around to reading my response to your response to that one thread everyone gets obsessed with?)

  • http://thinkinggrounds.blogspot.com Christian H

    You might remember, Leah, that in my e-mail to you I predicted I’d score low in both categories because I wasn’t intending to try to look like the sort of Christian or the sort of atheist that either side would like. I was kind of surprised that I did as well in the atheist round as I did, for that matter.

  • http://www.twitter.com/fodigg Matt

    I thought I would catch more people by imitating an actual Catholic (C8), but I think people went for consistency and so talking about internal conflicts probably wasn’t the way to go. Also, because most of the conversations I have with this person were topical to current politics, that was reflected in the answers.

  • Pingback: Returning to Turing Test Methodology

  • Pingback: Trackback

  • Pingback: Trackback


CLOSE | X

HIDE | X