INTRODUCTION: Despite the well-known health benefits of physical activity, a large proportion of college students are not physically active on a regular basis. Technological advances and the accessibility of computers have made Internet instruction a viable and feasible option. The purpose of this study was to examine the efficacy of a web-based instructional program for changing Social Cognitive Constructs that are correlated with physical activity behavior. METHODS: College students enrolled in health and wellness courses were recruited for this study. Four classes received ten weeks of instruction on the health and fitness aspects of exercise plus the web-based program (treatment, n=80), three classes received ten weeks of instruction on the health and fitness aspects of exercise and no web-based program (comparison, n=91), and two classes were in unrelated areas (cancer and aids education) (control, n=137). The web-based program consisted of 9 web assignments (1 per week) that targeted: analysis of past and present physical activity behaviors, behavioral goal setting, social support, self-regulation, self-efficacy, self-monitoring, and awareness as related to physical activity. Each of the SCT variables was measured using an instrument with previously established validity and reliability. RESULTS: Sixty-eight percent of the students found the assignments helpful in planning for long-term physical activity. The number of web assignments completed was significantly correlated with days absent (r=-.28) and overall grade (r=-.36). The percentage of students completing the web assignments was higher for students receiving a grade of C or better (61%) in the class compared with students receiving a C or worse (35%) (p=.001). Social support from friends and family, self-efficacy, and self-regulation were similar between the treatment, comparison, and control groups before the start of the course. Post-test change scores for self-regulation skills were significantly higher in the treatment group (16.9) vs. the comparison (8.8) and control group (-0.7). Knowledge concerning the application of the skills taught by the web assignments (time management, self-regulation, exercise tailoring, behavioral strategies for physical activity) was significantly greater in the treatment vs. the other groups after the course. CONCLUSION/DISCUSSION: This study demonstrates that a web-based instructional program has a positive impact on knowledge and skills related with physical activity. The program was easy to administer and it was well received by a majority of students. Further application and evaluation of this approach to changing physical activity behavior is warranted.