RESUMO
This paper describes the factor analysis testing and construct validation of the Japanese version of the Caffrey Cultural Competence Health Services (J-CCCHS). The inventory, composed of 28 items, was translated using language and subject matter experts. Psychometric testing (exploratory factor, alpha reliability, and confirmatory factor analyses) was undertaken with nurses (N = 7494, 92% female, mean age 32.6 years) from 19 hospitals across Japan. Principal components extraction with varimax rotation yielded a 5-factor solution (62.31% variance explained) that was labeled: knowledge, comfort-proximal, comfort-distal, awareness, and awareness of national policy. Cronbach α for the subscales ranged from 0.756 to 0.892. In confirmatory factor analysis using the robust maximum likelihood estimator, the chi-square test was as follows: χ2 (340) = 14604.44, P < .001. After correlated errors were introduced, there was evidence of improved model fit (χ2 (335) = 8681.61, P < .05) but the other indices showed improvement (RMSEA = .058 [90% CI, 0.057-0.059], TLI = .891, CFI = .903, and SRMR = .059). The discriminating power of the J-CCCHS was indicated by statistically mean differences in J-CCCHS subscale scores between predefined groups. Taking into consideration that this is the first foray into construct validation for this instrument, and that fit was improved when a subsequent data driven model was tested, and it has the ability to distinguish between known groups that are expected to differ in cultural competence, the instrument can be of value to clinicians and educators alike.