This study investigated: the impact of addition of learning styles identification and interpretation on an existing academic skills improvement program, the effect of student's learning new material at preferred or non-preferred times of day, and learning style characteristics for different sexes, ethnic groups, and college majors. Student GPAs and probationary status were compared for 144 freshman students admitted on Individual Approval status, i.e., 71 students who completed the Academic Skills Workshops during the Fall of 1983 and 73 students who completed a revised program in 1984. Reading gain scores and learning style characteristics were studied for the 1984 students. Learning style characteristics were measured by the Productivity Environmental Preference Survey and reading gains were measured by the Nelson-Denny Reading Test. Analysis of variance, simple effects analysis, and chi square analysis were used to determine whether GPAs and probationary status significantly improved after the addition of learning style information for the total sample, sex and ethnic subgroups. Reading gain scores were compared by means of a t test. Analysis of variance and simple effects analysis were used to determine whether different learning style preferences existed for different sex, ethnic, and college major groups. Findings indicated that GPAs and probationary status did not significantly improve for students who received learning styles assessment and interpretation as compared to those who did not, either by total sample or subgroup analysis. Reading gain scores were not significantly better for time-congruent than time-incongruent students. However, scores on some learning style elements were significantly different for male, female, Anglo-American, Mexican-American, and college major groups. Differences in program instructional format and setting, sample, and outcome measures between this study and previous research were discussed as possible reasons for the lack of significant improvement in GPA, probationary status or gain scores. Some of these factors may also have been responsible for the low number of elements found to be significantly different for the subgroups studied.