A test of the false positive bias hypothesis.

An evaluation of the hypothesis that adolescent gamblers are biased toward the production of false positive responses proposed by Derevensky et al. was tested. The results of secondary analysis of available data supported the hypothesis for items on the South Oaks Gambling Screen and items from the DSM-IV screening instrument. The conclusions were presented as tentative in view of the modest sizes of the observed effects. A number of limitations to the analysis are also described.