A study was conducted at Miami-Dade Community Colleges (MDCC) to investigate whether a combination of freshman and sophomore grades and student scores on the College Level Academic Skills Test (CLAST) would be a better predictor of student academic success at the four-year level than CLAST scores alone. Since there were insufficient data to determine the predictive validity of CLAST, an analogous situation was used to test the premise; i.e., the college performance of students with high grade point averages (GPA's) from high school and low scores on the Comparative Guidance and Placement (CGP) test was analyzed. The study sample was a group of 455 credit students who had been offered the Scholar's Grant award based on their standing in the top 10% of their high school class. Of these students, 326 scored high enough on the CGP to register for college-level course work, and 129 scored low enough on at least one of the sub-tests to be eligible for remedial course work. The performance of these two groups was compared on the basis of whether they earned an associate degree, number of credits earned, and student GPA. There was no significant difference between the groups on the basis of credits earned or degree attainment; however, a larger proportion of students not eligible for remedial course work earned GPA's of 3.0 or higher. Based on study findings, it was concluded that achievement test results are not better predictors of overall success than previous academic performance, and that use of CLAST results alone will probably lead to erroneous admission decisions. (HB)