90% of Statistics are Bogus, But….
One of my first acts as a opposition researcher with the NZ Labour Party, aged about 7, was to put together a graph that purported to show a sharp increase in violent crime in a provincial town; which one I can’t recall. It was a stunning revelation to me at the time that, by fiddling with the parameters of a graph, you could dramatically increase its impact. If, for example, there were six armed robberies in Wanganui in 1992 and nine in 1993, by starting the Y axis on a bar graph at five you can create an instantaneous crime spree. Because the average Wanganui (Taupo? Bay of Plenty?) journalist is probably as scientifically rigorous as I am, it dominated the headlines for at least one news cycle and earned me some welcome kudos along the way. In retrospect, this very unsophisticated statistical trickery taught me three lifelong lessons, namely:
Never trust a statistic when it supports the agenda of the person who came up with it;
There is no such thing as an objective fact;
It’s much easier to win an argument when you set the rules of engagement.
My little Wanganui (or Taranaki? Hawkes Bay?) crime wave produced a treasure trove of Machiavellian insights, foremost among them being that the presentation of statistics as a proxy for ‘objective truth’ is one of the most pernicious and pervasive forms of bullshit in the world today.
Having said all that, there are some statistical findings too sumptuous and gratifying to resist. Okcupid, an online dating site, has recently crunched its member profiles to gain a better understanding of its target audience. They sliced and diced this data in many revealing ways, but most interesting to me was how they used the Coleman-Liau Index to approximate the U.S. grade level necessary to comprehend the self-composed profiles of each of its 500,000 plus members. They then cross-referenced the findings with religiosity — as well as degrees of religiosity (i.e. strong or weak belief). Here is the delectable outcome:
Now, I am not saying that reading age is a perfect indicator of intelligence. Sorry, let me correct myself: I am saying that.
I can’t decide which of the following findings I like the most:
The highest reading age of all is among half-hearted Buddhists
Whereas strong religious belief has a negative correlation with reading level, the opposite is true of atheists who appear to become smarter the further they move from god.
Protestants are dumber than…well, everybody, and Muslims are smarter than Catholics (though less so than Jews, which must sting a little).
To highlight what this means in practice, especially for people who don’t immediately grasp the US grade system, here is a representative sample of books and plays recommended for 7th graders — or very committed Protestants:
Anne of Green Gables
The Miracle Worker
Harry Potter books
Chronicles of Narnia
In contrast, atheists and non-committal Buddhist are reading at a 9th grade level:
All Quiet on the Western Front
One Flew Over the Cuckoo’s Nest
Grapes of Wrath
Most pointedly, though, consider this: the most popular translation of the bible — the New International Version — requires a reading grade level of 8.4. In other words, the only Christians able to read and comprehend it are lapsed Catholics — or, to frame yet another way, the less you comprehend the words of the Bible, the more you are likely to believe them.