Abstract
General Health Questionnaire (GHQ) results are given for a large (N = 1013) sample of South Australian young people (average age 19.6 years), to compare the usefulness of the 12-, 28-, and 30-item forms of the GHQ. Internal reliabilities are generally adequate and the Likert scoring method produces significant correlations with psychological measures such as self-esteem. the case-prevalence rate using the binary scoring method was comparable with other studies, but misclassification rates were unacceptably high when DSM-Ill Axis I diagnosis was used as the criterion for the presence of any psychiatric disorder.
