This study compared the effect of two sources of instruction and verbal encouragement on the participation of individuals with mental retardation (MR) in a 10-week physical activity program. Participants were 17 adult employees of a sheltered workshop (5 females, 12 males) ranging in age from 30 to 65 years. The program was offered at work 3 days per week and involved aerobic dance activities. Group engagement in moderate to vigorous intensity physical activity (MVPA) was systematically observed and was compared using a reversal design. Condition A involved an exercise leader plus an exercise video to deliver instructional cues and verbal promotion of participation. Condition B used an exercise video as the only source of instruction and verbal promotion. The exercise videos were designed specifically for the participant group to address the low fitness levels and limited ability to make activity transitions. Data indicated that, on average, a higher percentage of the group was engaged in MVPA when the leader-plus-video condition (A) was applied. However, the difference was not practically meaningful when the administrative ease and cost-effectiveness of videos are considered. Further, a considerable overlap of data points in the graphical analyses indicated that withdrawing the leader did not control exercise behavior. Program attendance was variable but remained high over the course of the study. Group engagement levels were higher during the sessions with fewer participants, which suggested that a small group of highly compliant participants were more consistently on-task. Work performance was not negatively impacted when employees took time out of their workday to participate in physical activity. It is of importance that several participants continued to participate in the exercise program over the 4-week maintenance phase. This study provided a convenient, inexpensive method for adults with MR to independently participate in physical activity. / Graduation date: 1999