Exploring longitudinal measurement invariance and the continuum hypothesis in the Swedish version of the Behavioral Regulation in Sport Questionnaire (BRSQ): An exploratory structural equation modeling approach

Abstract

Objectives: The aims of the present study were to: (a) examine longitudinal measurement invariance in the Swedish version of the Behavioral Regulations in Sport Questionnaire (BRSQ) and (b) examine the continuum hypothesis of motivation as postulated within self-determination theory. Design: Two-wave survey. Method: Young competitive athletes (N = 354) responded to the BRSQ early in the season (November) and at the end of the athletic season (April). Data were analyzed using exploratory structural equation modeling (ESEM) and bifactor ESEM. Results: We found support for strict longitudinal measurement invariance in the BRSQ. Latent mean comparisons showed an increase in external regulation and amotivation across the season. The latent factor correlations indicated some deviations from a simplex pattern related to amotivation, external regulation, and introjected regulation. In the bifactor model, intrinsic motivation items had negative factor loadings on the global factor, identified regulation items had factor loadings approaching zero, and introjected and external regulation and amotivation items all had moderate to strong positive factor loadings. Conclusion: The present study adds longitudinal measurement invariance to the psychometric evidence of the BRSQ. Research on why the latent means of the behavioral regulations changed over the athletic season is warranted. The continuum hypothesis was partially supported. Latent factor correlations and factor loadings on the global factor in the bifactor ESEM highlighted that the discriminant validity of the controlled regulations and amotivation needs further investigation.