Based on a review of the evidence of previously published studies and on new evidence presented here, the author argues that the broken home is more strongly related to the commission of some offenses than to others; and that sex, race, and/or socioeconomic status may influence the strength of the relationship. Broken homes seem to be more related to ungovernable types of behavior and to marijuana use, rather than to alcoholism, theft, or vandalism. The association seems to be stronger for some types of children, such as white girls, but the evidence is slight. Data were collected for this investigation through a 1975 survey of Arizona high school students, in which anonymous questionnaires were returned with answers to questions on self-reported delinquent behavior, the latter identified in twenty delinquent and status offenses. Eight of these offenses were selected so that researchers could examine their relation to the broken home: vandalism, assault, shoplifting, auto theft, marijuana use, drinking alcohol, truancy, and running away. The definition of broken home for the survey was limited to father absence. Sex, ethnicity, and religious affiliation were determined through direct questions. The findings of the survey were not conclusive, thus explaining the somewhat contradictory findings of other recent broken home research. The broken home-delinquent behavior relationship can be established for some offenses and some children, but it appears, nevertheless, weak. Due to the complexity of the etiology of delinquent behavior, the broken home can be regarded as one criminogenic factor among many. Explanatory endnotes and 31 references are appended.