This research uses the National Education Longitudinal Study of 1988 (NELS) to examine student misbehavior in the context of individual, familial, and institutional influences. The NELS is the third longitudinal study of students and schools conducted by the National Center for Education Statistics, and the NELS surveys encompass not only students but also their parents, teachers, and school administrators. The purpose of the paper is to add to the experiential base of students using data collected by NELS. Latent factor structures of student misbehavior as well as variables that aid in the explanation of student misbehavior were derived through common factor analysis of selected NELS items. Factor structures were tested for internal consistency and generalizability by gender, race, and socioeconomic status. A series of exploratory univariate and multivariate logistic regression models were used to determine risk factors for and protective factors against behavior problems. The NELS project provides a wealth of information with which to study student misbehavior, but it does pose certain limitations, including those of missing data, survey revisions, and survey breadth over depth. It was hypothesized that misbehavior would be distinct from substance abuse in high school, though strongly correlated, but scaling procedures yielded a misbehavior variable heavily defined by substance abuse. Misbehavior was inseparable from substance abuse, but substance abuse was not dependent on misbehavior for viability. One explanation for this finding may lie in the survey itself since misbehavior items were sparse and could have been explored more thoroughly. It was hypothesized that six variables would explain adolescent behavior, but only four reliable constructs were realized: student self-reported self-concept, peer influence, parent involvement, and administrator-reported school climate. It is hoped that this research can be used to design appropriate treatment and prevention programs for adolescents and to draw attention to the penetrating influence of a child's friends on his or her behavior. It can also show the wealth of data collected and the usefulness of NELS data. (Contains 1 table and 10 references.) (SLD)