Vulnerable Children and Youth Studies, 2007, 2, 2, 127-141 Reliability of Self-Report of Health in Juvenile Offenders Dianna T Kenny a and Jennifer Grant b Abstract The aim of the present study was to investigate the accuracy of self- reports of juvenile offenders on physical factors (e.g., sleep difficulties, weight related behaviors and weight perceptions), health risk behaviors (e.g., alcohol use), trauma history (e.g., physical and sexual abuse) and psychological factors (e.g., anxiety, suicidal and self-harm behaviors). Self-reports obtained via a Health Questionnaire from 242 incarcerated juvenile offenders were compared with standardized measures (Body Mass Index, Adolescent Psychopathology Scale and Child Trauma Questionnaire) to investigate the reliability (via construct validity) and veracity of their self-report. Using kappa estimates and receiver operating characteristic curves, results generally showed high agreement across measures, suggesting that self-report questions from the health survey could all be used reliably. The degree of accuracy indicated that young offenders are as reliable as clinical and community samples of adolescents in their self-report. These findings have implications for routine assessments and practice evaluations that rely on self-report as the method of data collection and as the basis for clinical formulation and treatment planning. Keywords: juvenile offenders, self-report, reliability, health Correspondence concerning this article should be addressed to Professor Dianna T. Kenny, Faculty of health Sciences C42, University of Sydney PO Box 170, Lidcombe NSW 1825 Ph: +61-2-9351-9644; Fax: +61-2-9351-9540 Email: d.kenny@usyd.edu.au a University of Sydney, Australia b University of New South Wales, Australia Introduction Page 1 of 15