The findings suggest that an active consent procedure introduces a small degree of bias into surveys of students' substance use, which raises concerns about the validity of population estimates of substance use obtained from these studies. This bias may be exacerbated in prevalence studies that examine trends over time. Active consent procedures reduced the number of younger students who participated in the survey. Among younger students, there was a tendency for fewer boys and fewer students from non-English-speaking backgrounds to participate in the study. Active parental consent also produced greater homogeneity in the students surveyed in a school. This suggests that a greater sample size will be needed under active parental consent, requiring that more students from a given school be surveyed. Since schools are becoming increasingly reluctant to participate in surveys due to the disruption to students' study, the request to survey an increasing number of students may make schools even more reluctant to participate in such surveys. The increased sample size would also increase the cost of conducting the study. A total of 4,754 students in grades 7 through 12 (mainly ages between 12 and 17) were surveyed. Of these students, 1,968 (41 percent) were from schools that required active parental consent. There were 41 schools in the active-consent condition and 44 schools in the passive-consent condition. 3 tables and 14 references