Work engagement in Japan: Validation of the Japanese version of the Utrecht Work Engagement Scale

The purpose of this study was to validate the Japanese version of the Utrecht Work Engagement Scale (UWES-J). Employees from three independent samples completed the questionnaire (total N = 2,334). Confirmatory factor analyses using the multiple-group method revealed that, instead of the original three-factor model, a one-factor model that assumes that all engagement items load on one single factor fitted the data. Moreover, the one-factor structure was invariant across all three samples. Internal consistency of the scale was sufficiently high (alpha = .92) and the test-retest reliability with an interval of two months was .66. A series of second-order confirmatory factor analyses to examine construct validity confirmed the hypothesised model, indicating that work engagement was positively related to job satisfaction whereas it was negatively related to strain and burnout. However, as in previous studies, professional efficacy, a component of burnout, loaded on the engagement factor. These results, in general, suggest the reliability (internal consistency and stability), factorial invariance, and construct validity of the UWES-J in three independent samples.