This study investigated possible explanations for increases in marijuana use among eighth grade students in Texas from 1992 to 1996. The variables examined were availability of marijuana, perceived dangerousness of marijuana use, use of alcohol and cigarettes, parental attitudes toward marijuana use, and a number of other variables related to family, peers, and school. Although many of these variables were strongly related to marijuana use during the past year, only measures relating to availability of marijuana showed increases in prevalence similar to those of marijuana use across three time periods investigated. Another finding of interest was the dramatic increase (from 4% in 1992 to 24% in 1996) in marijuana use among African American students.