'Education - something' seems to fill a lot of lower rungs.

I know that these scores were a semi-common meme around grad school when I was there. People wondered why philosophers/physicists/mathematicians gravitated towards each other, but it was silently suspected among a few of us that it was due to similar analytical ability. The results were even more pronounced when the logic test still existed.

I think that one shouldn't overemphasize the mean though. There's a lot to be said for the distribution -- in physics, the entire lower part of the curve gets weeded out. That doesn't seem to happen in CS. And philosophers learn to write arguments and navigate arcane vocabularies as a matter of course. Perhaps the math result is indicative of general ability, but they have to do logic, too.