To help assess the effects of school-based random drug testing programs, the U.S. Department of Education's Institute of Education Sciences (IES) contracted with RMC Research Corporation and Mathematica Policy Research to conduct an experimental evaluation of the Mandatory-Random Student Drug Testing (MRSDT) programs in 36 high schools within seven districts that received Office of Safe and Drug-Free Schools (OSDFS) grants in 2006. This report describes the implementation of the MRSDT programs and their impacts on students-focusing primarily on student-reported substance use but also examining other outcomes. The study's key findings indicate that Consistent with the goals of the program, students subject to MRSDT reported less substance use than comparable students in high schools without MRSDT. Specifically, student-reported past-30-day use of substances tested under their districts' MRSDT policies was lower in schools implementing MRSDT than in schools without such policies. A similar, though not statistically significant, pattern was observed on other student-reported substance use measures. However, the MRSDT program had no "spillover effects" on the substance use reported by students who were not subject to testing and had no effect on any group of students' reported intentions to use substances in the future. Contrary to concerns raised about the possible unintentional negative consequences of random drug testing, the MRSDT program had no effect on the proportion of students participating in activities subject to drug testing or on students' attitudes toward school and perceived consequences of substance use. Finally, there was some evidence that impacts of the MRSDT program were related to the ways in which the programs were implemented. Both testing for a larger number of substances and testing for alcohol and tobacco were significantly correlated with lower substance use in the treatment schools relative to the control schools. However, it was not possible to distinguish between these two factors due to the fact that districts tested for a larger number of substances were also those districts that tested for alcohol or tobacco. Impacts were not significantly related to other implementation characteristics examined. Appended are: (1) Random Assignment; (2) Obtaining Parental Consent; (3) Sample Sizes and Response Analysis; (4) Diagnostic Analyses; (5) Outcome Measures; (6) Estimating Impacts; (7) Assessing the Robustness of the Impacts; (8) Impacts on Individual Substances; (9) Impacts on Student Subgroups; and (10) Study Instruments. (Contains 81 tables, 9 figures, and 26 footnotes.)