The effects of the use of computer-enhanced instruction with remedial students were assessed, using 4,293 ninth through twelfth graders--3,308 Black, 957 White, and 28 Other--involved in the Governor's Remediation Initiative (GRI) in Georgia. Data sources included the Comprehensive Tests of Basic Skills (CTBS), a data collection form developed for the study, and scores from tests accompanying each module of the computer-enhanced curriculum. Data were analyzed using the Statistical Package for the Social Sciences (1984). Analyses of variance were conducted on selected variables. The computer-based instruction used was found to be effective and showed superiority to traditional classroom instruction for the remedial students in the program. All comparisons showed significant gains on the CTBS, with the exception of the tenth-grade gain for subtest concepts and applications. The average gain for all students on total math was 4.772 normal curve equivalents. The magnitude of progress is still in question as the CTBS gain scores may not be representative of the math curriculum and the module unit tests are not norm referenced. GRI administrators need to reexamine in-depth the use of classroom aides, evaluate the module test items for sex bias, check the program screening process for racial bias, assess student gains for various racial categories, and assess any standardized assessment instrument used for evaluation of the project for criterion-related validity before use. Ten tables are provided. Lists of software and primary instructional materials, and the GRI Math Lab Data Collection Score Report and Demographics Forms for 1986 are appended. (TJH)