Although many attempts have been made to assess problem or pathological gambling in adolescents, concerns have been raised about whether existing measures are ideally suited for this purpose. Such measures are heavily influenced by traditional addiction models common to the study of substance use. In contrast, more recent public health approaches to gambling place a greater emphasis on the role of behavior and its harmful consequences and this is implicit in many currently accepted definitions of problem gambling. This paper reports on the use of one such measure (Victorian Gambling Screen -VGS), with 926 grade 7-12 adolescents surveyed in the Australian Capital Territory. The VGS was shown to correlate well with the gold standard Diagnostic & Statistical Manual-IV-Juvenile Screen (DSM-IV-J) for problem gamblers producing similar prevalence estimates. The measure also has sound internal reliability and concurrent validity. The findings suggest that harm-based measures such as the VGS are credible with adolescent populations in Australia and that various forms of harm observed in adult populations can also be observed in adolescent problem gamblers.