Construct validity of the demand-control model: A double cross-validation approach

Abstract

According to Karasek's demand-control model, negative work- and health-related outcomes are espccially likely to occur when job demands are high and a worker's control over the job is low. Many studies have examined this hypothesis, but no consistent supportive evidence has been found. The current paper reviews earlier studies and critiques, and focuses on the internal and external validity of the demand-control model as it is usually understood. In a first study the factorial structure of the model was examined. Confirmatory factor analyses of data from two samples (n = 179 and 302, respectively) demonstrated that a three-factor model (with factors Skill Discretion, Decision Authority and Psychological Job Demands) fitted the data best. In a second study using the same samples the authors examined how these three factors were related to a range of health- and work-related outcome variables. The interaction hypothesis suggested by Karasek was not supported, but the authors did find main effects of the three factors. Implications and limitations of the study are discussed.