I don’t think it’s an argument for disregarding social science
It’s not an argument for disregarding social science, but it is an argument to be more sceptical of its claims.
I also don’t think you necessarily need a high IQ to be a successful social scientist.
I disagree but let me qualify that. If we define “successful” as “socially successful”, that is, e.g., you have your tenure and your papers are accepted in reasonable peer-reviewed journals, then yes, you do not need high IQ to be be successful social scientist.
However if we define “successful” as “actually advancing the state of human knowledge” then I feel fairly confident in thinking that a high IQ is even more of a necessity for a social scientists than it is for someone who does hard sciences.
Ah, I’m sorry—I actually agree with everything you just wrote. I fear I may have miscommunicated slightly in the comment you’re replying to.
You’re right, I did point that out. And I do think that it can be harder in social science to weed out the good stuff from the bad stuff, and as such, you can get reasonably far in social science terms by being well-spoken and having contacts with a similar ideology even if your science isn’t great. This is an undesirable state of affairs, of course, but I think it’s just because doing good social science is really difficult (and in order to even know what good social science looks like, you’ve gotta be smart enough to do good social science). It’s part of the reason I think I can be useful and make a difference by doing social science, if I can do good rational social science and encourage others to do more rational social science.
My point isn’t that you don’t need to be as smart to do social science; doing it well is actually harder, so you’d expect social scientists to be at least as smart as hard scientists. I think that social science and hard science require slightly different kinds of intelligence, and IQ tests better for the hard science kind rather than the social science kind.
It’s really difficult to make a formula that calculates how to get a rocket off the ground. You have to crunch a lot of numbers. However, once you’ve come up with that formula, it is easy to test it; when you fire your rocket, does it go to the moon or does it blow up in your face?
It’s really easy to come up with a social science intervention/hypothesis. You just say “people from lower classes have worse life outcomes because of their poor opportunities (so we should improve opportunities for poor people)” or “people from lower classes are in the lower class because they’re not smart, and their parents were not smart and gave them bad genes, so they have worse life outcomes because they’re not smart (so we should do nothing)” or “people from lower classes have a culture of underachievement that doesn’t teach them to work hard (so we should improve life/study skills education in poor areas)”. I mean, coming up with one of those three is way easier than designing a rocket. However, once you’ve come up with them… how do you test it? How do you design a program to get people to achieve higher? Run an intervention program involving education and improved opportunities for years, carefully guarding against all the ideological biases you might have and the mess that might be made by various confounding factors, and still not necessarily have a clear outcome? There’s not as much difficulty in hypothesis-generation or coming-up-with-solutions, but there’s a lot more difficulty in hypothesis-testing and successful-solution-implementing.
Hard science requires more raw processing power to come up with theories; social science requires more un-biased-ness and carefulness in testing your theories. They’re subtly different requirements and I think IQ is a better indicator of the former than the latter.
It’s not an argument for disregarding social science, but it is an argument to be more sceptical of its claims.
I disagree but let me qualify that. If we define “successful” as “socially successful”, that is, e.g., you have your tenure and your papers are accepted in reasonable peer-reviewed journals, then yes, you do not need high IQ to be be successful social scientist.
However if we define “successful” as “actually advancing the state of human knowledge” then I feel fairly confident in thinking that a high IQ is even more of a necessity for a social scientists than it is for someone who does hard sciences.
As you pointed out yourself , hard sciences are easier :-)
Ah, I’m sorry—I actually agree with everything you just wrote. I fear I may have miscommunicated slightly in the comment you’re replying to.
You’re right, I did point that out. And I do think that it can be harder in social science to weed out the good stuff from the bad stuff, and as such, you can get reasonably far in social science terms by being well-spoken and having contacts with a similar ideology even if your science isn’t great. This is an undesirable state of affairs, of course, but I think it’s just because doing good social science is really difficult (and in order to even know what good social science looks like, you’ve gotta be smart enough to do good social science). It’s part of the reason I think I can be useful and make a difference by doing social science, if I can do good rational social science and encourage others to do more rational social science.
My point isn’t that you don’t need to be as smart to do social science; doing it well is actually harder, so you’d expect social scientists to be at least as smart as hard scientists. I think that social science and hard science require slightly different kinds of intelligence, and IQ tests better for the hard science kind rather than the social science kind.
It’s really difficult to make a formula that calculates how to get a rocket off the ground. You have to crunch a lot of numbers. However, once you’ve come up with that formula, it is easy to test it; when you fire your rocket, does it go to the moon or does it blow up in your face?
It’s really easy to come up with a social science intervention/hypothesis. You just say “people from lower classes have worse life outcomes because of their poor opportunities (so we should improve opportunities for poor people)” or “people from lower classes are in the lower class because they’re not smart, and their parents were not smart and gave them bad genes, so they have worse life outcomes because they’re not smart (so we should do nothing)” or “people from lower classes have a culture of underachievement that doesn’t teach them to work hard (so we should improve life/study skills education in poor areas)”. I mean, coming up with one of those three is way easier than designing a rocket. However, once you’ve come up with them… how do you test it? How do you design a program to get people to achieve higher? Run an intervention program involving education and improved opportunities for years, carefully guarding against all the ideological biases you might have and the mess that might be made by various confounding factors, and still not necessarily have a clear outcome? There’s not as much difficulty in hypothesis-generation or coming-up-with-solutions, but there’s a lot more difficulty in hypothesis-testing and successful-solution-implementing.
Hard science requires more raw processing power to come up with theories; social science requires more un-biased-ness and carefulness in testing your theories. They’re subtly different requirements and I think IQ is a better indicator of the former than the latter.