Objective. (1) To investigate the contribution of job characteristics and personal characteristics to the explanation of health effects among university personnel; (2) to investigate the differences between scientific personnel (SP) and non-scientific personnel (NSP); (3) to investigate whether health effects occurred one after another. Methods. The well being at work of employees at a Dutch university (n=2,522) was investigated by means of a questionnaire. A model was constructed in which several job and personal characteristics were set out against health effects. The latter were assumed to occur in phases: decreased 'job satisfaction' as an early effect, followed by increased 'tension' and 'emotional exhaustion', and possibly also by increased 'perceived health complaints'. The contribution of job and personal characteristics to the explanation of health effects was investigated by means of linear regression analysis, with separate analyses for SP and NSP. Results. Positive job characteristics, especially professional expertise and work variety, contributed to the explanation of 'job satisfaction'. The major contributors to 'tension' and 'emotional exhaustion' were negative characteristics, such as work pressure. Besides the negative aspects, the major contributors to the explanation of 'perceived health complaints' were sex, age and other health effects. In NSP, social support contributed to the explanation of 'tension' and 'emotional exhaustion', but not in SP. The explained variance of 'job satisfaction' by the positive job characteristics in NSP was much higher than that in SP. To investigate whether health effects occurred one after another, we considered explained variance. Explained variance in 'job satisfaction' was much higher than in 'perceived health complaints'. 'Emotional exhaustion' and 'tension' were in between. Conclusions. Contrary to expectations, decision latitude and social support played only minor roles. Also, the differences between SP and NSP were smaller than expected. As individual condition and burden of outside work also influence health effects, we recommend the inclusion of relevant personal factors in future studies.
