Psychometric assessment of workaholism measures

Purpose - Although the concept of workaholism has existed in the academic literature for decades, exploration of its measurements seems to lag behind. The purpose of this study is to present an investigation of the three most commonly used workaholism measures; the Workaholism Battery (WorkBAT), the Work Addiction Risk Test (WART) and the Dutch Work Addiction Scale (DUWAS) in terms of their cross-validation, their temporal stability and their factor structure. Design/methodology/approach - The three measures were administered to 661 cross-occupational Norwegian workers. A total of 368 of these completed the same measures 24-30 months later. Findings - The cross-validation showed that the correlations between the scores of the different instruments were too low to conclude that they measure the same construct. The 24-30 month test-retest reliability coefficients for the measures revealed that the scores were quite stable over time. None of the previously suggested factor solutions for the three measures had a good fit with the data. Explorative factor analyses supported a four-factor solution for the WorkBAT and for the WART. A two-factor solution for the DUWAS was found. Research limitations/implications - All the data are based on self-report, which might bias the results. Practical implications - Different workaholism measures cannot be used interchangeably. Originality/value - This is the first study that cross-validates the three most used workaholism measures and which investigates the reliability of these instruments over a long-term period (24-30 months).