Validation of the Japanese Version of the Burnout Assessment Tool

Abstract

The current study aimed to validate the Japanese version of the Burnout Assessment Tool (BAT-J), a new burnout measure. We conducted an Internet survey to confirm the validity and reliability of the BAT-J, using registered monitors from a Japanese survey company. The first-wave survey was conducted in May 2018, with 1,032 monitors. Of these, 498 participated in the second-wave survey in June 2018 to confirm 1-month test-retest reliability. We examined the factorial validity of the BAT-J core symptoms (BAT-JC) and BAT-J secondary symptoms (BAT-JS), as well as their reliability (internal consistency and test-retest reliability) and construct validity. Factorial validity was examined using confirmatory factor analyses and exploratory structural equation modeling bifactor analyses. Convergent and discriminant validity were examined using multitrait-multimethod frameworks well as the average variance explained. Exploratory structural equation modeling bifactor solutions for the BAT-JC, BAT-JS, and BAT-J demonstrated the best fit to the data. They also indicated that the general factor accounted for over two-thirds of the common variance explained. Internal consistency and test-retest reliability were confirmed. Convergent and internal discriminant validity of the BAT-JC were confirmed vis-ȧ-vis burnout, as assessed with the Maslach Burnout Inventory-General Survey. Moreover, external discriminant validity of the BAT-J was demonstrated for work engagement and workaholism. Finally, both BAT scales showed significant positive relationships with job demands and turnover intention. All validity results were in line with the job demands-resources model. The results of the current study provide the first evidence for the BAT-J's reliability and factorial and construct validity.

Burnout Assessment Tool (BAT)—Development, Validity, and Reliability

“…Additionally, the MTMM-framework also indicates that there is some divergence in the way in which each questionnaire conceptualizes burnout, which is not surprising given their different conceptualizations. Additionally, the recent Japanese study of Sakakibara et al [ 68 ] who also used the MTMM framework corroborated the current findings by demonstrating convergent and discriminant validity of the BAT vis-à-vis the MBI-GS. Taken together, these results concur with the conclusions drawn by Halbesleben and Demerouti [ 72 ] who used a similar framework as well to test the convergent and discriminant validity of the OLBI and MBI-GS.…”

“…Namely, (a) burnout is considered to be a syndrome —which is compatible with the idea of a second-order model; and (b) a distinction between core and secondary burnout symptoms is in line with our definition of burnout that is grounded in our conceptualization of burnout as well as in our qualitative analyses (Part 1). This is confirmed by a recent study of the BAT in a representative sample of Japanese workers [ 68 ]. This study corroborated that a second-order factor model including only the core symptoms as well as a second-order factor model including the core symptoms plus the secondary symptoms fitted well to the data.…”

“…Given the novelty of our study, this was a first step in validating the BAT. Although the results are promising and seem to be corroborated by three very recent studies [ 68 , 69 , 70 ], they are still preliminary. Additional, more elaborate testing should take place in the future in order to evaluate the validity of the BAT.…”

“…Secondly, the results further suggest that the construct of burnout, as measured by the BAT, can be discriminated from other well-being constructs, such as engagement (i.e., UWES), job boredom (i.e., DUBS), or workaholism (i.e., DUWAS). Using the same questionaries and methodology, a recent Japanese study corroborated the discriminant validity of the BAT vis-à-vis work engagement and workaholism [ 68 ]. These results concur with Schaufeli, Taris, and Van Rhenen’s [ 73 ], who conclude that engagement, workaholism, and burnout can be distinguished from each other empirically, as well as with Schaufeli and Salanova [ 45 ] who theorize about the differences between burnout, job boredom, and engagement.…”

A Rasch analysis of the Burnout Assessment Tool (BAT)

“…Our results with a strong general factor are also confirmed in a recent article that investigated the measurement invariance of the BAT across seven cross-national representative samples, and in which the BAT was modelled as a second-order factor and showed a good fit to the data [ 22 ]. In a similar vein, a recent study investigating the Japanese version of the BAT suggested the presence of a strong common factor as well [ 51 ].…”