Social support, occupational stress, and health.

The present paper is concerned with the buffering hypothesis that social support ameliorate. the impact of occupational stress on job-related strain and health. Previous studies of this hypothesis have yielded conflicting results. Our purpose, therefore, is twofold. First, we summarize the literature in this area and review several studies in detail, all of which found main effects of social support on perceived occupational stress and on some health outcome measures. Three of the studies were specifically designed to examine the buffering effects of support. Of the three, two found little or no evidence for buffering (LaRocco and Jones, 1978a; Pinneau, 1975), whereas the third reported buffering effects (House and Wells, 1978). Second, we attempt to reconcile these different conclusions by reanalyzing one data set-first analyzed by Caplan et al. (1975) and then by Pinneau (1975)-using a moderated regression technique identical to that used in the LaRocco and Jones (1978a) and House and Wells (1978) studies. The data usedfor this analysis consist of a randomly stratified sample of men from 23 occupations (N 636). Our review andfindings support the buffering hypothesis for mental and physical health variables (anxiety, depression, irritation, and somatic symptoms), but, as in the previous three studies, fail to support the buffering hypothesis in regard to job-related strains (job dissatisfaction, boredom, dissatisfaction with work load).