Researchers have devoted a great deal of attention to gender differences in math perceptions and performance. Although the gap has closed at some levels, important differences persist and have been addressed in detail in previous chapters. However, when one throws ethnicity into the mix, the picture becomes far more complicated.

While the ethnicity gap has been decreasing over the past few decades, gender differences are a primary focus within the mathematics domain, ethnicity differences are a concern across a wide range of educational areas. This ethnicity gap has been decreasing over the past few decades, however, African American and Latino/Latina students consistently receive lower scores than do European American students across a wide range of high-stakes standardized tests (Camara & Schmidt, 1999; Jencks & Phillips, 1998). Moreover, African American and Latino/Latina high school students and college freshman (Ramist, Lewis, & McCamely-Jenkins, 1993) have lower average grades and class rank than their European American and Asian American counterparts (National Task Force on Minority High Achievement, 1999). In contrast, on most tests, the performance of Asian American test takers is similar to that of European American students. The one exception to this pattern is the Quantitative Test of the Graduate Record Examination where Asian American students score higher than European American students (Camara & Schmidt, 1999).

Given these gender and ethnic differences, then, some researchers suggested that women of color should be most disadvantaged group in a math context because they suffer from a double bind: they are ostensibly victims of additive effects related to both gender and ethnicity.