Evidence is also mixed as to the effects of gender on college attendance, but most research has shown that females are more likely to attend college after controlling for background factors. Marsh (1991c), in an analysis of a nationally representative data set, found that females are more likely than males to attend college, after controlling for background factors. He found this difference to be largely mediated by differences in psychosocial and behavioral factors, including their higher average academic self-concepts, school effort, and grades. Similarly, Zaff, et al. (2001) found that males were more likely to attend college after controlling for a set of background, parental control, and school-level factors. Haveman and Wolfe (1995) found that being a Black female increased the likelihood of attending college, after controlling for background factors. Similarly, Borus and Carpenter (1984) found that males were less likely to attend college. However, Leventhal, Graber, and Brooks-Gunn (2001), in an analysis of a sample of Black urban children, did not find gender to be significantly related to likelihood to attend college.