This study was intended to determine whether sex-linked mathematics achievement and contextual differences persist into the university years. A random sample of 150 first-year mathematics students enrolled at a midwestern university was identified. A common 30-item calculus readiness test and a 37-item questionnaire, dealing with students' perceptions of the context of mathematics learning, were administered to all students present on the testing day. One hundred examinees returned the questionnaire, and 131 completed the calculus readiness test. Of the 100 questionnaire respondents, 91 had taken the test. Data on these 91 subjects were used in subsequent analyses where data for both instruments were involved. When high school algebra grade was controlled, males scored significantly higher than did females on the calculus test. However, a multivariate analysis of variance of the seven components of the test produced a significant sex effect and univariate differences in favor of males were found for geometry, inequalities, and conceptual problems. The sex variable, coded as a dummy effect, did not enter the regression equation. Algebra study and hours of mathematics homework accounted for 53.5% of the variance in the test grades. Five data tables are included. (TJH)