Speaking of this, let’s talk about Tex’s business as if it were our business. After all, he’s the original prophet of the whole neanderwhatsit.
The Patrick Kinney I was friends with growing up was scored at 156 Stanford-Binet. At this time I had only scored 148 at the age of 9. It was only later in the military that was revised to 183. Both Patrick and I just naturally understood he was the brighter of the two of us.
So we have a starting point.
Now, I mentioned this claim before (183 IQ) and Heaviside said he didn’t think the military had IQ tests with that kind of range. I replied that because the Army has shown serious interest in IQ since at least the days of the Army Alpha, they probably retain at least some interest in putting the right people in the right jobs. Well, it seems that’s roughly true, because they didn’t do anything remarkably stupid with this score. And that’s saying something, when you’re talking about the people who inspired the phrase “Army-proof”.
In the olde days of psychometrics, they would have taken this Stanford-Binet score and applied the simple ratio IQ formula: Test score x (Mental Age / Actual Age) = IQ. In Tex’s case, this would give the hilarious overestimate of 148 x (9 / 18) = 296. They could also have taken the score at face value, saying “148 is high enough and you probably aren’t any smarter than when you were nine”. Instead of doing either, like Adragon de Mello’s retarded charlatan of a father (who estimated his son’s IQ as 400), it seems the Army either exercised remarkable restraint in their estimation process or applied a secret IQ test for stratifying spooks. The latter seems more likely to me, because it’s only in the last couple of decades that they started selecting for conscientiousness and selecting against psychoticism.
Let’s compare their estimate with the intuition of a career psychometrician, Paul Cooijmans, whose website is canonical around here.
This is a first attempt to show how intelligence rises and falls throughout life. The rough graph below is an educated guess of the normal development; that is, the development in persons without severe brain damage or severe degenerative brain diseases:
I.Q. development with age modelled
In the conservative case we would only expect the adult Tex to add 20 points to the test given at the age of 9, which still yields a hefty 168.
(Though I’ve said this approximately ONE BILLION TIMES, I want you to keep in mind that IQ scores above 130 get weird. I first realized this while reading The Bell Curve, where the authors give hundreds of examples, one after another, where more IQ is always better. But there was one blip in the data that I haven’t been able to find since then, where performance in college was actually lower for the highest IQ decile than the 9th decile. The authors put this down as a statistical anomaly, but I remember thinking at the time “wait, that’s me and that’s exactly right!”. See, I score 99th percentile on IQ tests but 90th percentile on most academic achievement tests. This was probably 2010, when I first got really interested in psychometrics and strange ideas like “conscientiousness”.)