The South Oaks Gambling Screen-revised Adolescent (SOGS-RA) revisited: a cut-point analysis

Abstract

The lack of a criterion validity gold standard defining adolescent pathological gambling represents a major limitation of the adolescent gambling literature. The present study employed Receiver Operating Characteristic curve analysis (ROC) to examine the performance of the South Oaks Gambling Screen-Revised Adolescent (SOGS-RA) recommended cut-point of 4 against two-proxy gold standards: self-identified need for, or receipt of, help for gambling. Logistic regression analysis examined the correlates of self-identified need for help with gambling. The sample was comprised of 12990 adolescents from Atlantic Canada whose average age was 15 years and 50% of whom were male. The SOGS-RA performed as well or better with these proxy gold standards than with proxies used in previous studies. We concluded that the proxy gold standards based on self-identified need for or receipt of help represent a pragmatic solution to the lack of an adolescent-specific definition of problem or pathological gambling. Students in grade seven and SOGS-RA scores of 4 or greater were independent predictors of self-identifying a need for help.