A MULTIVARIATE MODEL FOR ABSENCE AMONG WORKERS IN MENTAL HEALTH (BURNOUT, JOB DESIGN, REGRESSION ANALYSIS)

Abstract

The major focus of the present investigation was the examination of a multivariate model of absence from work. This project entailed examining the relationships between and among variables which represented three major aspects of worker absence: the behavior itself, attitudes regarding absence, and the demographic characteristics of the subjects studied. The subjects for the study were 77 staff members (50 men, 27 women) of an urban children's psychiatric hospital.;The subjects were administered a battery of attitudinal questionnaires, which assessed two work-related attitudes, locus of control, and intrinsic motivation. Their work attendance records were reviewed for a six month period, and the inactivity rate (percent of absence) was computed for each subject. The seven independent variables: age, sex, locus of control, intrinsic motivation, salary, motivating potential score, and burnout were arranged in a causal path model with percent of absence as the dependent variable. Causal path analysis, multiple regression, correlations, and examination of the distributions of each variable were the data analysis methods utilized.;The specific ordering of variables in the path model was not entirely supported by the data. However, part of the model was supported at a significant level (p < .05) and other relationships within the model were supported by theory trimming. Paths, where B < .15, were deleted. Nonetheless, the amount of residual unexplained by this model was substantial. The study supports the notion that absence can be better understood in a multivariate model and that absence is in part a function of behavior, attitudes, salary, and other personal characteristics. Finally, the discussion notes that absence may have beneficial effects and clearly needs more research.