This study investigated the possible role of self‐efficacy in the stress process by examining relations between stressors (role ambiguity, situational constraints, and hours), strains (job dissatisfaction, anxiety, frustration, and turnover intent), and efficacy beliefs (both individual and collective). Individual efficacy was related to only two of the four strains and had no mediating or moderating effects. Collective efficacy, however, was strongly related to both stressors and strains. Collective efficacy also moderated the effect of work hours and mediated the relation between situational constraints and two of the strain measures. It was concluded that the theory of individual self‐efficacy (Bandura, 1977) may not adequately explain collective efficacy. Future research on the self‐efficacy construct as well as its role in the stress process was suggested.