The psychometric properties of demand-control and effort–reward imbalance scales among Brazilian nurses

Abstract

Purpose To compare the psychometric adequacy of the demand-control-support (DCS) and the effort–reward imbalance (ERI) questionnaires in relation to their respective theoretical models among workers within the same psychosocial work environment. Methods A self-report questionnaire was administered to 1,509 nursing personnel at two Brazilian hospitals. Analyses were based on internal consistency, corrected item-total correlation, and confirmatory factor analysis models for ordinal data (conducted with Mplus software) to test the fit of each model. Results Internal consistency was satisfactory (>0.70) for most scales of the DCS and ERI questionnaires, except for the job decision latitude scale (α = 0.50). With reference to corrected item-total correlations, adequate performance was observed for most items of both questionnaires. The item ‘Do you have to do same thing over and over again’ had a very low corrected item-total correlation (−0.09) and the removal of this item increased the internal consistency of the skill discretion subscale. Overall, fit indices (of factorial validity) for the DCS and the ERI questionnaires were adequate. However, a low correlation between subscales of the decision latitude scale (−0.164) was observed. For the psychological job demands and overcommitment scales, the best adjustment of the model was obtained when correlated errors between items were considered. Conclusions Strengths and limitations were observed in both the DCS and the ERI scales. The questionnaires could be improved by the revision of some items.