Job control, task-related stressors, and social stressors: Identifying groups with different developmental patterns in these variables, and predicting psychological and physical well-being by group membership

Cumulative effects of working conditions on health and well-being have been reported repeatedly; however, changes over time have also found to be important. We 1) identified groups with different trajectories of job control, task-related and social stressors (simultaneously) and 2) predicted well-being and health by group membership. Method: We used data of the N = 483 participants (spanning 10 years; Nt1 = 1394) of the “Work Experience and Quality of Life in Switzerland” study (Semmer et al., 2005). Task-related stressors, job control (ISTA; Semmer et al., 1999) and the social stressors (Frese & Zapf, 1987) were the predictor variables. Outcome measures were irritation (Mohr et al., 2006), job satisfaction (Baillod & Semmer, 1994), and BMI. Trajectories of working conditions were determined by Growth Mixture Modeling (GMM). We then tested if outcome variables at t5 differ between those trajectories, adjusted for the baseline value of the respective outcome (ANCOVA). To determine to what extent results were due to trajectories vs. current working conditions (short-terms effects), we compared these results to an analysis in which predictors in t5 were controlled as well. Results: We decided for a five-class model. Two of the smallest classes were characterized by the least favorable constellation of the level and development of work characteristics over time: The “high strain class” was characterized by increasing social and task-related stressors and by decreasing job control over time. The “active job & high social stressors” class was characterized by high job control and high job demands as well as by high social stressors. When adjusted for the base level of the respective outcome variable the effect of class membership on all outcomes was significant. Pairwise comparisons revealed that “active job & high social stressors” class were significantly associated with a higher level of irritation and BMI, and lower level of job satisfaction compared to the more “favorable” groups. The “high strain class” showed the same effects for job satisfaction. Most effects also held when predictors at t5 were controlled as well. Theoretical implications and conclusions: Although classes with unfavorable trajectories were small, their well-being at t5 was lower than that of more favorable classes. Adjusting the level of predictors at t5 represents a rather strict test; it shows that not only short-term exposure (level of predictors at t5) but also preceding developments were important. Our results also suggest that social stressors at work are especially important and can offset effects of job demands and control