The ‘Obama effect’ provides a great chance to prove or disproove the stereotype threat theory. The last four SAT test dates were 10/4/08, 11/1/08, 12/6/08, 1/24/09. The last one is 4 days after the inauguration, peak of Obama mania. In future months we are going to get the statistics of the SAT data. This is real world data, not some artificial small data set.

Does anyone really thinks that the achievement gap would disappear or at least reduced in the Jan result?

The issue of academic comparability…In past studies, they were able to control for student SAT. We did not have that data, but did have education level, and controlled for that (the means we report are adjusted means, controling for education). That said, this population is probably more balanced on education than the world in general (all participants were saavy enough to be part of an on-line subject pool). That is similar to prior studies, where students were from the same school (e.g., Stanford). These studies (to my knowledge) have not been done with kids who were really not involved in school. Indeed, the effect assumes that the subjects care enough about their performance on the test to be anxious about the results. So, there is comparability, but when we (and prior studies) show equal scores, that does not imply that we would expect equal scores in the general population. We believe the “Obama effect” helps blacks who are engaged and relatively well educated to overcome stereotype threat, but do not expect that the “Obama effect” will eliminate the national black-white difference in test scores. (Some people believe that is what we were claiming, which is understandable if they see a 2-minute CNN report.)

This summary is correct, except that our sample was adults, not students:
“In two of the tests, the black students did, as predicted, have lower scores than their academic abilities would predict. But just after the Democratic Convention that nominated Obama, and again just after the election, the black students reminded of their race didn’t get lower scores than academically comparable white students.”

In all four testing situations, the students were reminded of their race before being given the test. In previous research, this has been shown to induce “stereotype threat,” where black students underperform compared to academically comparable white students. In two of the tests, the black students did, as predicted, have lower scores than their academic abilities would predict. But just after the Democratic Convention that nominated Obama, and again just after the election, the black students reminded of their race didn’t get lower scores than academically comparable white students.