[Development of the Japanese version of the Rathus Assertiveness Schedule]

The present study investigated the reliability and validity of the Japanese version of the Rathus Assertiveness Schedule (RAS) preparatory to study the relationship between self-expression skills and mental health status of employees at a Japanese worksite. Assertiveness is one of self-expression skills and defined as standing up for one's own rights as well as for others. RAS that the most widely used was developed by Rathus in the U.S.A. and consists of 30 items. We developed the Japanese version of RAS (J-RAS) along with a back-translation. The subjects were 173 participants of a 2-day assertiveness training course from July to December, 2001 and 364 workers in a manufacturing company in the Kyushu area. We sent the J-RAS to the 173 participants before they commenced the course, and of them 170 participants returned sufficient answers for analysis. During the course their trainers completed the objective evaluations on assertive self-expression. We compared their answers with the trainers' evaluations by the Pearson's correlation to certify the J-RAS's validity. We also mailed the J-RAS to the manufacturing workers in June, 2001. Of them, 266 workers returned sufficient answers by the end of the month. We calculated the Cronbach's alpha coefficient by them to certify the internal consistency reliability of the J-RAS. In August, 2001, we mailed the J-RAS again to 98 workers who agreed to retake the test. We studied the Pearson's correlation between the initial answers and the following ones to certify the test-retest reliability of the J-RAS, using the 98 answers of the workers who agreed to be retested. Our results found that 19 item sets, excluding Q3--7, 9, 13, 19, 20, 25, 28, had significantly positive correlation with the objective evaluations. The sum of the 19 items of the J-RAS (19-item J-RAS) had significantly higher correlations with the objective evaluation than that of 30 items of the J-RAS (30-item J-RAS). The internal consistency reliability of the 19-item or 30-item J-RAS was high, because their Cronbach's alpha coefficient was more than 0.80. Their test-retest reliability was also high, because the Pearson's correlation coefficient between the initial test and the subsequent test was more than 0.84 (ps < 0.001). Our results implied that the validity and reliability of the J-RAS was could be considered as acceptable and that the 19-item J-RAS measured the assertiveness of Japanese people more exactly than the 30-item J-RAS.