The Alignment of Measures and Constructs in Organizational Research: The Case of Testing Measurement Models of Creativity

Organizational creativity research has a curious misalignment between construct definitions and measurement model specifications-definitions embrace multiple facets, but empirical measures do not. The purpose of this study was to examine potential measurement model misspecification in organizational creativity research. We compare whether creativity is best assessed as a unidimensional common latent construct model with reflective indicators or as a multi-dimensional composite latent construct model with formative indicators. To assess potential measurement model misspecification, two studies were conducted in organizational and professional settings. For Study One, MBA students (n = 152) evaluated stimuli from entrepreneurship and advertising. For study, two professional artists (n = 167) evaluated art domain stimuli. CFA results suggest composite latent construct models with two factors (novelty and usefulness) represent creativity assessments in entrepreneurship, advertising, and art better than one- and three-factor models. Results suggest that failure to acknowledge inconsistencies between construct definitions and measurement models may put researchers at risk of reporting findings with limited statistical conclusion validity. Further, improved theories and empirical models should include facets of creativity. Broader implications of measurement model misspecification for organizational science research are also discussed. This is one of the first studies to examine the potential measurement model misspecification in organizational creativity research. We examined this potential using data from three domains and across two domain-specific samples. The results were robust across all samples and settings and suggest concern with respect to current methods used for measuring organizational creativity