This study was designed to determine if observing a similar or dissimilar model who makes varying self-efficacy statements influences an observer's efficacy expectations and, in turn, muscular endurance performance. Females (N = 150) were randomly assigned to groups in a 2 × 4 × 3 (model similarity by model talk by trials) factorial design or to a no-model control group. Model similarity was manipulated by having subjects view a female described as a nonathlete (similar) or a male described as a varsity track athlete (dissimilar). The four levels of model talk included: a positive self-talk model who performed and made positive self-efficacy statements, a negative self-talk model who made negative self-efficacy statements, an irrelevant-talk model who made statements unrelated to self-efficacy, and a no-talk model who remained silent throughout the performance. Self-efficacy measures were assessed in addition to performance on three trials of a muscular endurance task. Results revealed that similar model subjects extended their legs significantly longer than dissimilar model and control subjects. Moreover, the similar-positive-talk and similar-no-talk groups performed significantly better than the dissimilar-positive-talk, dissimilar-negative talk, dissimilar-no-talk, and the no-model control groups. Subject self-efficacy, however, was not found to be the major mediating variable affecting these performance changes.
This experiment was partially funded by an All University Research Grant at Michigan State University. The authors would also like to acknowledge Mary 30 Faust for her assistance in the data collection and Nancy Steel for her participation as a model. Reprint requests should be sent to Daniel Gould, Department of Health, Physical Education, and Recreation, Room 203 - IM Sports Circle, Michigan State University, East Lansing, MI 48824.