Abstract Background Evaluation of psychosocial factors requires instruments that measure dynamic complexities. This study explains the design of a set of questionnaires to evaluate work and non‐work psychosocial risk factors for stress‐related illnesses. Methods The measurement model was based on a review of literature. Content validity was performed by experts and cognitive interviews. Pilot testing was carried out with a convenience sample of 132 workers. Cronbach's alpha evaluated internal consistency and concurrent validity was estimated by Spearman correlation coefficients. Results Three questionnaires were constructed to evaluate exposure to work and non‐work risk factors. Content validity improved the questionnaires coherence with the measurement model. Internal consistency was adequate (α = 0.85–0.95). Concurrent validity resulted in moderate correlations of psychosocial factors with stress symptoms. Conclusions Questionnaires' content reflected a wide spectrum of psychosocial factors sources. Cognitive interviews improved understanding of questions and dimensions. The structure of the measurement model was confirmed. Am. J. Ind. Med. 56:100–110, 2013. © 2012 Wiley Periodicals, Inc.