A detailed analysis of the significant gender differences observed in performance on the mathematics portion of the Scholastic Aptitude Test (SAT) in 1986 was performed for students enrolled in the Montgomery County Public Schools (Maryland). A disaggregation of data from the 1986 SAT administration was performed to examine homogeneous subgroups and shed light on two of the factors believed to affect differential performance: (1) courses taken; and (2) socioeconomic status. SAT data are presented for the county's 2,924 female and 2,715 male SAT-takers (in grade 12) who graduated in 1987. Data assessed included information on participation and performance of females in grades 1 through 12; SAT scores of men and women in different ethnic groups; and SAT scores of white women and white men of comparable educational background. A thorough analysis of the data available on course participation, school performance, socioeconomic status, student and parental attitudes toward mathematics, and possible school effects suggests that the reasons for the persistence of the gender gap in mathematics test performance lie outside these factors. The study did not contribute to the current effort to examine test bias as a source of the persistent differential performance on the SAT math test by male and female students. Seven data tables are provided. (TJH)