"Stereotype Threat"

I commented fairly acerbically in April 2004 on the latest gambit among psychologists for explaining away low average black IQ and educational attainment. It is the "stereotype threat" theory. The theory seems to have a number of incarnations but at its wackiest, it says that blacks do poorly because they are afraid of letting down their race! Nonsense as gross as that hardly needs comment but let me simply ask: If such fears exist at all, why do they not make the student try harder and thus do better?

A somewhat more reasonable theory is that blacks "drop their bundle": They know that they are on average unlikely to do well so don't really try to do well. All the studies that I have heard of over the years tend to show, however, that blacks have unusually HIGH motivation in the testing situation -- so that theory need not detain us.

In that situation, psychologists have had to devise ever more dubious experiments to support their ideas. The latest appears to be a study reported here of which we read: "In their spring study of 81 students at Boca Ciega High School in Gulfport, Brett Jones and Tom Kellow investigated "stereotype threat," a phenomenon in which students worry their failure might confirm a negative belief about their race'.

They found that giving a test under "threatening" conditions -- where students were told that the result would predict their later educational success -- produced much lower scores among blacks than when the test was given with a more reassuring introduction. Under the more reassuring condition, scores of blacks and whites were about the same. The authors concluded that the findings supported their theory.

It is difficult to know where to start in commenting on such nonsense. Quite aside from the total disregard for sampling that is characteristic of most psychological research, Occam's razor has been completely ignored. If the findings show anything, they simply show that blacks handle stress less well. There is no need to bring "stereotypes" into the explanation. But the gaping oddity in the findings is that blacks and whites did in one condition do about equally well on the test. That is totally contrary to all prior findings where some attempt at sampling was used and suggests that the whole setup was severely contrived in some way and that the results therefore tell us nothing at all. To put it as politely as I can, there was at least a very strong "Rosenthal" (experimenter expectation) effect at work.

I am not the first to find the "theory" risible. Steve Sailer also dissects it and there is an even more savage takedown at Gene Expression. Charles Murray is probably the most succinct of us all, however. Of the original study in the field, he says: "Claude Steele and Joshua Aronson, demonstrated experimentally that test performance by academically talented blacks was worse when a test was called an IQ test than when it was innocuously described as a research tool. Press reports erroneously interpreted this as meaning that stereotype threat explained away the black-white difference. In reality, Messrs. Steele and Aronson showed only that it increases the usual black-white difference; if one eliminates stereotype threat, the usual difference remains". See also here and here.

Comments? Email John Ray