For instance, Jordan and Nettles (1999), in an analysis of data from the National Educational Longitudinal Study of 1988 (NELS), found that girls had lower scores on math and science tests in the 12th grade than boys. Additionally, Gamoran (1992) found girls scored lower on a test of math and verbal achievement. Likewise, Entwistle, Alexander, and Olson (1994) found that adolescent males outperformed adolescent females on the Math Concepts and Applications test, a subtest of the California Achievement Test focused on math reasoning. The sample included 279 Black and White students from Baltimore. Further, the difference in boys' and girls' performance was most severe among the highest achieving youth. Among the relatively small samples of high achieving (31 boys and 38 girls), the average score for boys was 22.5 points higher than that for girls. Further, this difference was even stronger among African Americans, in which there was a greater gap between high-performing males' and females' test scores (Entswisle et al., 1994). This pattern held despite the fact that African American females were far more likely to be enrolled in algebra class (62 percent of females versus 33 percent of males).

Not all studies support this pattern. For instance, Entwisle et al, (1994) also found an opposite pattern among the low-performing youth, with girls outperforming boys within this subgroup. Also, Guo (1998), in an analysis of NLSY, found that males and females scores did not differ significantly on the math achievement tests, after controlling for a number of background factors.