In risk assessment practice, correctional staffs have been recommended to collect necessary information through face-to-face interviewing for reducing intentional distortion of response or social desirability bias. However, currently in correctional practice, it is discussed that self-report questionnaire is one of options as well as clinician-administered method in the accuracy of predicting recidivism. The purpose of this study was to examine how procedures for the assessment affect inmates’ responses in a sample of Japanese prisons. Participants were 1,574 inmates who were newly admitted into prisons, and they completed self-report questionnaire (including 47 items) either in front of correctional staffs or in their room. Responses to two different settings were compared. The results revealed that in approximately three-fourths of the items, there were no statistically significant differences between two groups. However, ceiling and floor effects were found in 28 items, and these results seemed partly because of the social desirability. Implications for offender assessment and future directions are discussed.
