That's fairly weird how more people pick 50-75% than 25-50%. One would have to assume the vast majority of quantitative traits lie in the 25-50% range (though of course the general populace don't study behavior genetics, and this wasn't exactly a straightforward example).

Liberals seem to be more genetically deterministic than their conservative counterparts, which is also weird since you'd expect they would be more educated (or was this controlling for education?)

Thanks Razib, a good study.

Liberals seem to be more genetically deterministic than their conservative counterparts, which is also weird since you'd expect they would be more educated (or was this controlling for education?)

i didn't control. but the difference in education is rather modest. 14% of liberals have grad educations vs. 8% of conservatives, and 19% have college vs. 20% of conservatives. if you limit the graduate educated sample to only liberals and conservatives (remove moderates), 55% are liberal, and 45% conservative. (i limited to years 2000-2008)

but the ideological results are weird enough that it might be something wrong with the coding. but pretty much the same thing occurs with parties, republicans are a touch more environmentalist than democrats. even when i limit the sample to whites there isn't much of a change. the sample sizes get smaller, but if you limit to those with college degrees (at least), still the same trend. perhaps those on the political right are trying harder *not* to seem racist?

That's fairly weird how more people pick 50-75% than 25-50%. One would have to assume the vast majority of quantitative traits lie in the 25-50% range

Can someone fill me in on what such a percentage figure means to a geneticist (or closely related fields) in practice? Without defined units of comparison, it strikes me as ridiculously ill-defined, like asking "is this New York cheesecake sweeter than it is expensive?"

Depending on the units, one could get arbitrarily near 100% or 0%. Are we supposed to think about the ratio of the genetic difference between George and Verne Troyer, to the environmental difference between George and his separated identical twin who gets raised to watch a little more TV? Or is it more like the ratio of genetic difference between George and his defect-free first cousin, to the environmental difference between George's upbringing in San Jose and his twin's in Darfur?

i really don't think what they're asking is scientifically coherent. but i think it's a rough sense of how plausible people might think a high or low level of heritability might be, even if they don't know the terms. additionally, i think the absolute level is less informative than the rank orders or relative differences.

I think that educated, intelligent people are also less likely to have much personal experience with seriously competitive sports that require a lot of strength or size so they just don't realize how important genetic luck is. I mean, sure, all those middle-class guys "go to the gym to stay in shape", but they never tried to gain 20 kilos of muscle or had their dreams of being a hockey player smashed by bigger, faster guys, nor do they have many friends or relatives who did stuff like that.

Anyone who spends a lot of time working hard at certain sports is going to see that some people get a lot more results with a lot less work, and when asked probably attribute a big chunk of that to genes.

I do realize a lot of Americans would answer differently if George wasn't a black man - on American weightlifting forums you get your standard-issue racist flame wars whenever anyone wonders if black guys have a genetic advantage at anything or even mentions genetics without bringing up race - but I think more educated and intelligent people would still lean more towards the nurturist side even in non-PC-sensitive societies.

@Drogomir:

Your point is well taken, but such evidence doesn't directly point to genetic dominance. It shows that physical, as opposed to some kind of psycho-social factors like "motivation", are a much bigger immediate cause of performance differences. But couldn't those hormonal, metabolic, etc. differences still owe a lot to environmental differences in childhood or adolescence?

Razib, those education stats you give make me think that what the survey is capturing in liberalism is more economic liberalism than social liberalism, the former which would be more endorsed by the lower class. The liberal party can be seen a heterogenous group of ivory tower progressives plus lower class people voting self-interestedly on economic issues.

Anyway, the GSS seems to be capturing mainly the lower class group, otherwise you would expect the education level of liberals to be significantly higher than conservatives, and their response on the heritability question closer to the "educated" or "college degree" levels. As it stands the Liberal group looks more similar to the "no college degree" or even "dumb" group.

I didn't know how to aggregate in the GSS, thanks for posting that code.

Very few studies have been done of general athletic ability. I'd like to see more. General intelligence (g) has been studied a lot. We can give people a wide range of tests like remembering numbers, folding shapes in your head, vocabulary analogies, and it turns out they all correlate to g. Some correlate more strongly, some more weakly.

The same thing could be done for athletic ability (I'll call it a). Give people tests like the bench press, 50 meter dash, 100 meter swim, and see how they correlate. We can then ask:

To what extent is it heritable? (the usual twin/adoption studies could help)

Which individual skills best correlate with a?

Is there a "Flynn effect" for a? Is it going up or down?

And, of course, the controversial question...does the average value of a differ by race?

JG: I wasn't really thinking in terms of evidence of genetic dominance, only about the reason why the views of intelligent, educated people differ from the rest of us. Actually probably the guys most likely to say success in athletics is close to 100% genetic are skinny dudes who tried to lift weights to get bigger, didn't eat enough to make progress and gave up after a few weeks. They like to use genes as a convenient excuse for their failure when it's really a lack of hard work in the kitchen.

Chemdude: for tests of general athletic ability the standing broad jump is often used to evaluate kids' potential for many sports. Some data on kids collected in China or the Soviet Union might already exist.