The Washington Post reports:
Among white teens, Fryer and Torelli found that better grades equaled greater popularity, with straight-A students having far more same-race friends than those who were B students, who in turn had more friends than C or D students. But among blacks and especially Hispanics who attend public schools with a mix of racial and ethnic groups, that pattern was reversed: The best and brightest academically were significantly less popular than classmates of their race or ethnic group with lower grade point averages.
"For blacks, higher achievement is associated with modestly higher popularity until a grade point average of 3.5 [a B+ average], then the slope turns negative," Fryer and Torelli wrote in a new working paper published by the National Bureau of Economic Research. A black student who's gotten all A's has, on average, 1.5 fewer same-race friends than a straight-A white student. Among Hispanics, there is little change in popularity until a student's average rises above a C+, at which point it plummets. A Hispanic student with all A's is the least popular of all Hispanic students, and has three fewer friends than a typical white student with a 4.0 grade point average.
Fryer and Torelli based their conclusions on a federally funded survey of 90,118 junior high and high school students in 175 schools in 80 communities nationwide during the 1994-95 school year. The resulting data set contained a wealth of information on each student, including the number of friends they had and who those friends were. To prevent an inflated tally, the researchers counted students as friends only if each listed the other as a friend.
This supports what I've been saying for some time: that Hispanics have a worse attitudinal problem toward education than do blacks. Contrary to the claims of John McWhorter, African-American culture isn't particularly anti-intellectual or anti-education ... at least relative to the average black IQ of 85. Considering that only about 1 out of 6 African-Americans has a three digit IQ, blacks spend quite a few years in school and a surprising fraction at least attempt college.
In general, blacks may suffer from inflated expectations about education: the Yale or jail syndrome. How many times have you seen interviews with poor ghetto children who announce they are going to be a doctor or a lawyer? When it eventually dawns on them that no way no how are they ever going to be doctors or lawyers, too many decide that then they might as well deal drugs.
The average Hispanic IQ is somewhere around 91, but Hispanics don't average more schooling than blacks. In some ways, this is healthy: Hispanics with two digit IQs are more likely to go get a job than waste time at a community college. Still, it reflect an anti-educational bias in Hispanic culture that keeps down many Hispanics who do have the brains to make use of education.
Looking at the actual report by Fryer and Torelli, the peer pressure effect doesn't seem terribly huge:
Put differently, evaluated at the sample mean, a one standard deviation increase in grades is associated with roughly a .103 standard deviation decrease in social status for Blacks and a .171 standard deviation decrease for Hispanics. For students with a 3.5 grade point average or better, the effect triples.
So, for blacks, if their grades go up by a standard deviation, their social status falls by one tenth as much. Is that the cause of their low grades? Perhaps to some extent. It might well be an explanation for why blacks get even worse grades on average than their standardized test scores would predict. But how big is the impact of peer pressure against "acting white" relative to the brute factor of lower average IQ? And would blacks consider getting good grades to be "acting white" if blacks had the same average IQ as whites? Occam's Razor keeps bringing us back to recognizing IQ as the 800 pound gorilla of the racial education gap.
As you might expect, Fryer and Torelli don't mention IQ. Nor do they mention Asians.