This study investigated whether some coping strategies are better or worse than others, by virtue of their links with health behaviours, or their stress buffering effects. It also investigated whether some ways of attributing for success and failure in health behaviour change are better or worse than others, through their effects upon feelings, expectations, and intentions. Employees of a hospital that was closing down completed questionnaires at two points in time, one year apart (baseline N= 109; of which 102 successfully followed up). Analysis was by a combination of multivariate analysis of variance, latent class analysis, exploratory and confirmatory factor analysis, and regression analysis. The evidence supported a four-dimensional view of coping (cf., Cox & Ferguson, 1991), and (less strongly) a two-type model of routine health behaviours. There was little evidence that coping strategies were linked to either routine health behaviours or to health behaviours used as ways of coping. However, the use of health damaging behaviours as ways of coping was predicted somewhat by avoidance coping. Having controlled for negative affectivity, there was little evidence that coping or resources buffered the effect of stressors on well-being, or that coping mediated between resources and well-being (cf., Cohen & Edwards, 1989). There were, however, main effects of coping on well-being (e. g., avoidance coping acted to increase mental symptoms); and main and interactive effects of stressors and resources on coping (e. g., stressors acted to increase avoidance coping but resources buffered this effect). The evidence supported a four-dimensional view of attributions (cf., McAuley, Duncan, & Russell, 1992), but with differences between attributions for success and failure. There was only limited support for Weiner's (1986) model applied to health behaviour change. Of particular interest were the interactive effects of attributions on affective reactions and on intention; these effects invariably involved the stability dimension.