The level of drug use among adolescents in schools has fluctuated over the past 25 years, but illicit drug use reported by seniors rose from 30 percent in 1992 to 41 percent in 2002. This paper examines drug testing in schools as a means of preventing student drug use, which is one aspect of the "war on drugs," a public and policy response to the growing problem of drugs in schools. To explore this issue, researchers used school-level survey data about drug testing from 894 schools participating in two related studies: the Youth, Education, and Society study and the Monitoring the Future study. Using cross-sectional data, they used descriptive analyses and logistic regression to examine how the presence of drug testing relates to 12-month use of marijuana and any other illicit drugs by students. The study also sought to determine how many schools use drug testing, the characteristics of these schools and which students are tested for drugs. The study does not use a pre-post design nor does it have random assignment to treatment conditions, both of which would be desirable for an ideal evaluation of an intervention program. The research showed that relatively few schools report testing students for drug use. Data suggest that drug testing, as practiced in recent years in American secondary schools, does not prevent or inhibit student drug use. The two forms of drug testing that are generally assumed to be most promising for reducing student drug use - random testing applied to all students, and testing of athletes - did not produce encouraging results.