Abstract
While cohesion has been identified as an important predictor of individual exercise adherence in group settings (Spink & Carron, 1993), to date, this relationship has only been established in structured exercise programs. It is worth recognizing that many individuals are often active in groups that are less structured (e.g., workouts with friends). As distinctions have been reported in the psychosocial correlates most associated with exercise in structured and unstructured exercise settings (Spink et al., in press), this study aimed to examine the relationship between cohesion and adherence in an unstructured group exercise setting. Participants (N = 104) who reported exercising with others in an unstructured setting in the previous 4 weeks indicated the frequency of their exercise in this setting and completed the modified GEQ (Carron et al., 1985) to assess cohesion within that group. Regression results revealed two cohesion dimensions predicted exercise frequency: GI-Task, ? = .38, t (99) = 2.31, p < .05, and GI-Social, ? = -.30, t (99) = -1.99, p < .05. In contrast to previous results in structured settings, the GI dimensions of cohesion emerged as predictors of attendance behaviour, suggesting possible differences between structured and unstructured exercise settings. Also, the fact that social cohesion was found to be negatively related to adherence contrasts with previous findings that have suggested that cohesion is typically positively related to adherence.
