Measuring Emotional Labor: Survey Construction, Measurement Invariance, and Structural Equation Model

The capacity to engage in international comparative studies of emotional labor depends on having a robust survey instrument that works as well in one culture as in another. Comparative research projects are complicated undertakings because they must accommodate different languages as well as linguistic nuances. Challenges are compounded when empirical methods are employed. Translation and back-translation of the survey must be conducted to ensure that the meanings of items are as intended. Measurement equivalence must be assured in order for findings to be credible and for comparisons to be made across countries and cultures. This chapter describes development of the survey items, explains how the challenge of making comparisons across nations was overcome, and shows the structural equation model used in each nation’s analysis. Items for each of three emotional labor constructs—emotive capacity, pretending, and deep acting—were developed by first analyzing pilot data from five nations on three continents. After variables were refined, public service workers from a larger sample of nations were surveyed: seven nations on four continents. Multi-group confirmatory factor analysis (MGCFA) then tested for cross-national invariance. Results confirm both configural and partial metric invariance of the items, enabling comparison of emotional labor demands and its consequences across nations.