= 27d567bc2d48d9f40e9ccc20ea370b15)
TimS comments on Rationality Quotes Thread January 2016 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (244)
Pardon my ignorance, but all the "intellect realism" theories seem like they can be charitably paraphrased as group X:
- has a different mean IQ than the general population and/or
- has a different standard deviation for IQ and/or
- has a significantly skewed distribution from the normal curve
I've seen claimed IQ means in the 80s for black Americans. Observationally, American public life includes many black people for whom I find it implausible that they aren't pretty smart - eg Clarence Thomas, Colin Powell, Condeleeza Rice.
If I assume no difference in std dev or skew in intelligence distribution, it seems to me that I observe too many intelligent black folks for the mean to be in the 80s. Moreover, adding an assumption that std dev is lower doesn't help - now the successful black folk are explained, but I don't observe enough extreme low IQ folk.
That's why I conclude some error exists in the assumption of an 80s mean IQ.
Why would we have to assume the IQs for groups, when we could just go out and give people tests?
More technically, the assumption that IQ is a good measure of intelligence across different sub-cultures.
Why would we have to assume that IQ is a good measure of intelligence across different sub-cultures? Aren't there experiments we could perform to measure the validity?
It seems like a core assumption of the "intellectual" realists. I'm conceding it to strong-man the opposing argument. If we don't assume IQ is culturally independent, the correlation between IQ and life outcome looks like a hidden-variable measure of social acceptability - i.e. an expected status quo bias if people prefer those they perceive as in-group. That just weakens the realist argument.
This would mean the IQ scores are meaningless for cross-country comparison. And that just aint' so.
That's not what that link says - best I can tell, the book summarized states living in a high-IQ country is more predictive of good life outcomes than your own IQ.
Getting down to brass tacks, we are assuming a lot when we compare IQ numbers from different tests. WIAS-IV is not necessarily comparable to other tests in English. Assuming that the French language test measures the same thing as the WIAS-IV assumes the very conclusion that I'm not agreeing with. (Although I'm not arguing this point in our other discussion).
You keep on saying "assuming" and Vaniver keeps on telling you that there is no need for assumptions: we have data. It's not hard to give the same people different tests and then look at how do the scores correlate.
In fact, that's how the whole concept of IQ came into being -- IQ is an estimate of the general intelligence component (g) that is common to performance on a variety of intelligence-measuring tests.