The sample consisted of 1,203 youth who were attending 1 middle school (n=619) and 1 high school (n=584) in the third largest city of Coahuila, a northern Mexican border State. The study was conducted in May 1998. Indexes of classification accuracy of the POSIT Substance Use/Abuse scale were evaluated against a "drug abuse" problem severity criterion that combined youth who met DSM-IV criteria for alcohol abuse/dependence disorders with youth who had used other illicit drugs five or more times in their lifetime. The study found that using a cut score of one or two on the POSIT Substance Use/Abuse scale generally yielded optimal classification accuracy indexes that varied somewhat by gender and school subgroups. The classification accuracy of the POSIT Substance Use/Abuse scale was slightly better when used with high school males. This was due, in part, to the higher base rate of serious involvement in substance use among this group compared to other study participants. As with any screening tool, these study findings suggest that users of the POSIT in Mexican schools must weigh the relative importance of identifying a maximum number of at-risk cases (sensitivity) at the expense of specificity and overall classification compared with maximizing classification accuracy at the expense of sensitivity. Although the current study findings support those of Marino et al. (1998), they should be interpreted with caution while awaiting additional research with urban and rural youth across multiple school systems and States within northern Mexico that use multiple assessment strategies. 2 tables and 32 references