RESULTS OF THE NATIONAL STUDY, WHICH WAS SPONSORED BY THE NATIONAL INSTITUTE ON ALCOHOL ABUSE AND ALCOHOLISM, WERE BASED ON ANSWERS TO A 35-PAGE QUESTIONNAIRE ADMINISTERED TO 13,122 ADOLESCENTS SELECTED FROM GRADES 7 THROUGH 12 THROUGH A 2-STAGE STRATIFIED RANDOM SAMPLING DESIGN. MANY OF THE QUESTIONS WERE BASED ON THE BEHAVIOR SCALES DEVELOPED BY DONOVAN AND PSYCHOLOGICAL AND SOCIAL VARIABLES TESTED BY A NUMBER OF RESEARCHERS. IN THIS ANALYSIS, THE METHODOLOGY OF THE NATIONAL SAMPLE, THE QUESTIONNAIRE, AND THE PROCEDURES USED TO ANALYZE THE DATA TO DEVELOP 14 SOCIAL-BEHAVIOR TYPOLOGIES ARE DESCRIBED. THE DATA WERE ALSO USED TO DEVELOP HYPOTHESES REGARDING SOCIAL GROUP DIFFERENCES IN GENERAL DEVIANCE AND PREDICTIONS OF DEVIANT BEHAVIOR BASED ON BEHAVIOR TYPOLOGY. NUMEROUS TABLES PRESENT THE RESULTS OF THE MULTIVARIANT ANALYSES. IT WAS FOUND THAT THE BEHAVIOR TYPOLOGIES ACCOUNTED FOR ABOUT 32 PERCENT OF THE VARIANCE IN GENERAL DEVIANT BEHAVIOR. THE TYPOLOGIES ARE ACCOUNTED FOR 40 PERCENT OF ADOLESCENT INVOLVEMENT WITH MARIHUANA AND 20 PERCENT OF THE VARIANCE IN CHURCH ATTENDANCE. MORE DEVIANT PRONE TYPES WERE LESS CONCERNED WITH ACADEMIC ACHIEVEMENT, HAD LOWER EXPECTATIONS FOR ACADEMIC RECOGNITION, GREATER ATTITUDINAL TOLERANCE OF DEVIANCE, LOWER RELIGIOSITY, STRONGER ORIENTATION TO FRIENDS THAN TO PARENTS, AND GREATER SOCIAL SUPPORT FOR DEVIANCE (GREATER FAMILY APPROVAL OF TEENAGE DRINKING AND GREATER EXPOSURE TO PEER DRINKING AND DRUG USE). APPENDIXES CONTAIN A BIBLIOGRAPHY AND STATISTICAL DATA. (GLR)