This study sought to examine American College Test (ACT) and basic skills placement test score patterns; the use of these test scores in academic advising; student performance in freshman level reading, English, and mathematics courses; relationships among these factors and their power to predict or account for placement and performance; and differences in the patterns of placement and performance by academic/enrollment unit, race, and sex. The study's population was all first-time students (N=1,854) admitted for Summer or Fall 1988 at the University of Louisville (Kentucky). The study found that advisors do use the ACT and other scores to place students in certain types and levels of courses, which serves to institutionalize the patterns of difference by race and sex that appear in test scores. However, the study also found that when students were allowed the opportunity to perform, their academic performance patterns did not conform to the test score and placement patterns noted. For example, female students earned consistently higher grades than male students, although female students had lower ACT scores and mathematics placement test scores. Race had no statistically significant bearing on performance except in mathematics. (JB)