The students included 12 males and 12 females each from the 3rd, 6th, 8th, and 12th grades. All were enrolled in an urban school district serving a total of 35,000 students. Participants were interviewed individually. They used a five-point Likert-type scale to indicate their agreement or disagreement with 12 statements about drug education and its impacts. Results revealed that the students generally believed that each type of delivery mode would positively affect knowledge but not behavior related to alcohol and other drugs. In addition, students believed that education delivered by experts would be the most effective and parent-delivered knowledge the least effective means for increasing students' knowledge about alcohol and other drugs. Findings indicated the need to rethink drug education programs and perhaps acknowledge their limited purpose, because current education efforts are not achieving their intended outcomes. Tables, discussion of study limitations, and 50 references (Author abstract modified)