Jack comments on Undiscriminating Skepticism - Less Wrong

97 Post author: Eliezer_Yudkowsky 14 March 2010 11:23PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (1329)

Comments (1329)

Sort By: Best

Show:

You are viewing a single comment's thread. Show more comments above.

Comment author: Jack 17 March 2010 08:19:53PM * 0 points [-]

But Nisbett is quoting from a study "which found improvement on almost all tests for African American 12th graders". That study may not even have contained the data on 9-year-olds. You can ask "Why didn't that study include that data?", well because they were comparing data for 12th graders.

Comment author: brazil84 17 March 2010 09:08:32PM 0 points [-]

Actually, it's not clear to me what study he is talking about. Here's what he says:

The largest study, conducted by the NAEP, indicated that, if trends were to continue, the gap in reading scores would be eliminated in approximately 25 years and the gap in science scores in approximately 75 years.

So I went to the NAEP web site and looked at the very first graph I saw. What study do you think he is referring to?

Comment author: Jack 17 March 2010 09:56:19PM * 0 points [-]

Hedges and Nowell (1998) found improvement on almost all tests for African American 12th graders compared with other 12th graders over the period 1965– 1994. The best estimates in terms of the stability the scores provide, and in terms of their correlations with IQ, are in the form of composites, for example, reading + vocabulary + mathematics for the EEO survey. The Black–White gap on these composites over the period decreased on average by 0.13 standard deviation per decade, yielding an estimate of a reduction of the gap by around 0.39 standard deviation over the period. The largest study, conducted by the NAEP, indicated that, if trends were to continue, the gap in reading scores would be eliminated in approximately 25 years and the gap in science scores in approximately 75 years.

I take this to say that Hedges and Nowell examined lots of test results for African Amercicans 12th graders from 1965-1994. The test with the largest sample was the NAEP test. Since Hedges and Nowell were looking at 12th graders Nisbett is probably talking about the 17-year-olds.

I could be wrong. In any case, the trends have changed since 1994 so obviously the predictions don't hold.

This all seems pretty beside the point to me since the evidence that really matters is the adoption and skin tone studies. The other thing that becomes obvious is that there just isn't nearly enough data-- all the studies are decades old presumably because 1975 was the last time you could get grant money to study the issue. There certainly isn't enough to conclude, as you did, that there is obviously a genetic component.

Comment author: brazil84 comment score below threshold [+] (147 children)

Comment author: cupholder 18 March 2010 02:25:41AM 6 points [-]

That's only if you feel you need to rely on scientific studies to reach conclusions. Some things don't require such a study.

Yes, but you have to be super careful when deciding which things need scientific studies.

A few years ago I would've said women were so much more chatty than men - and that the difference in chattiness was so obvious - that it would be a waste of time to check it out scientifically. But sometimes, when you check things out systematically, you're surprised. I think the argument about blacks, whites and IQ is a bit like that, although that argument is more about the cause of the differences and not their mere existence.

Comment author: wedrifid 18 March 2010 03:22:05AM * 4 points [-]

Participants were assigned conversational partners at random, and asked to talk for up to ten minutes on one of forty topics

I would never have predicted that women would be more chatty in such a test. I would have predicted that men would talk more on a supplied topic. I believed, and still believe that women are more chatty under the commonly intended meaning of 'chatty'. A more relevant test:

Asign random pairs of people and send them on a 5 hour hike together. Count words.

I would expect female pairs to say more words than the male pairs. Mixed pairings I would find somewhat more difficult to predict due to possible interference from courtship protocols.

Comment author: mattnewport 18 March 2010 04:34:09AM 2 points [-]

I have my doubts about how strongly that particular test would correlate with what I understand by 'chatty'. It's a pretty artificial setup. When I think about it I have a pretty fuzzy idea of what 'chatty' means though. I would still say women are more chatty than men but that is partly because some part of the fuzzy definition involves 'the type of small talk that women tend to engage in more than men' rather than some idea of total word volume.

Comment author: brazil84 18 March 2010 09:44:21AM * 1 point [-]

Yes, but you have to be super careful when deciding which things need scientific studies.

I would say "reasonably careful" not "super careful." One thing I've noticed about the race and intelligence debate is that many people apply an extremely heightened standard of skepticism to the question.

By analogy, suppose we were debating the existence of God. There may very well be a few scientific studies out there which lend some degree of support to the theistic point of view. Further, there probably has not been a lot of scientific research into the subject. So one could take the "super careful" approach and say that the jury is still out on the subject. But that's silly. That's just exaggerated skepticism on the part of folks who don't like a particular conclusion.

For those of us who do believe in God but aspire to rationality, the best we can do is to concede there's a contradiction there.

Comment author: cupholder 18 March 2010 11:18:35PM -1 points [-]

When I suggested being 'super careful' I meant being super careful about deciding which things are so obvious as to not need systematic debate and study in the first place, not about deciding how skeptical to be of certain 'sides' or conclusions in a debate.

Comment author: brazil84 19 March 2010 12:26:48AM 1 point [-]

When I suggested being 'super careful' I meant being super careful about deciding which things are so obvious as to not need systematic debate and study in the first place, not about deciding how skeptical to be of certain 'sides' or conclusions in a debate.

I'm not sure what you mean by "systematic debate and study," but assuming it means the same thing as "scientific studies," it seems to me it amounts to basically the same thing. At least in this case.

Comment author: cupholder 19 March 2010 01:05:19AM 4 points [-]

I'll try and clarify with the non-race and IQ related example that first put the idea into my head: gravity. The idea of things falling to the floor is so obvious to me, and agrees so well with my common sense, that I would not even bother to debate somebody who wanted to argue that things don't fall to the floor. That's the behaviour I'm saying it's a good idea to be super careful about: rejecting challenges to your existing view out of hand.

Stepping back to the race and IQ argument, I'm saying that I would exercise a lot of care before I put the argument into the 'no need to even bother debating it' box. Having entered into the debate, though, I would be content to apply my ordinary standards of evidence to the different 'sides' in the debate. I mean the 'super careful' warning to apply pre-debate, not during the debate.

Comment author: brazil84 comment score below threshold [+] (137 children)

Comment author: NancyLebovitz 18 March 2010 03:12:30AM 1 point [-]

I'm interested that you believed women were much more chatty than men as recently as a few years ago.

I can remember it being a default belief in the culture that women were more chatty, but I thought it had faded out in the 80s or thereabouts.

Comment author: cupholder 18 March 2010 04:18:45AM 0 points [-]

I imagine it's less widespread a belief than before the 80s, but it's just one of those things you get by osmosis from the broader culture when you're young. It's part of the stereotypes there are about the sexes: women can't drive, men won't ask for directions when they're lost, blah blah blah.

Jack comments on Undiscriminating Skepticism - Less Wrong

Article Navigation

Comments (1329)

Nearest Meetups

Virtual Study Room

Recent Comments

Recent Posts

Recent Rationality Quotes

Recent Wiki Edits

Top Contributors, 30 Days

Recent Karma Awards

Comments (1329)

Top Contributors, 30 Days

You'll need to login or register to do that

(Don't worry, it only takes a few seconds)

Create

Login