Reliability and validity of the Japanese version of the organizational justice questionnaire

Objectives: Previous European studies reporting low procedural justice and low interactional justice were associated with increased health problems have used a modified version of the Moorman's Organizational Justice Questionnaire (OJQ, Elovainio et al., 2002) to assess organizational justice. We translated the modified OJQ into the Japanese language and examined the internal consistency reliability, and factor-based and construct validity of this measure. A back-translation procedure confirmed that the translation was appropriate, pending a minor revision. Methods: A total of 185 men and 58 women at a manufacturing factory in Japan were surveyed using a mailed questionnaire including the OJQ and other job stressors. Results: Cronbach alpha coefficients of the two OJQ subscales were high (0.85- 0.94) for both sexes. The hypothesized two factors (i.e., procedural justice and interactional justice) were extracted by the factor analysis for men; for women, procedural justice was further split into two separate dimensions supporting a three- rather than two-factor structure. Convergent validity was supported by expected correlations of the OJQ with job control, supervisor support, effort-reward imbalance, and job future ambiguity in particular among the men. Conclusion: The present study shows that the Japanese version of the OJQ has acceptable levels of reliability and validity at least for male employees.