Measurement invariance of the burnout assessment tool (BAT) across seven cross-national representative samples

Abstract

The aim of this study was to investigate the measurement invariance of the Burnout Assessment Tool (BAT) across seven cross-national representative samples. In this study, burnout was modeled as a second-order factor in line with the conceptual definition as a syndrome. The combined sample consisted of 10,138 participants from countries in Europe and Japan. The data were treated as ordered categorical in nature and a series of models were tested to find evidence for invariance. Specifically, theta parameterization was used in conjunction with the weighted least squares (mean- and variance adjusted) estimation method. The results showed supportive evidence that BAT-assessed burnout was invariant across the samples, so that cross-country comparison would be justifiable. Comparison of effect sizes of the latent means between countries showed that Japan had a significantly higher score on overall burnout and all the first-order factors compared to the European countries. The European countries all scored similarly on overall burnout with no significant difference but for some minor differences in first-order factors between some of the European countries. All in all, the analyses of the data provided evidence that the BAT is invariant across the countries for meaningful comparisons of burnout scores.