Self-Reported Violent Offending

Serious violent offenders are identified on the basis of self-reported involvement in violent FBI Index Offenses from a large representative sample of adolescents in the United States. The study involves a longitudinal prospective design with five annual waves of data on this national youth panel. Estimates of four basic parameters of violent offending are presented: prevalence rates, individual offending rates, career lengths, and crime mix patterns. These estimates are then compared to estimates based upon official arrest data. Self-reported estimates are substantially greater than those based upon arrest data, suggesting that those arrested for serious violent offenses constitute only a fraction of youth actually involved in serious violent behavior. The analysis also indicated that compared to self-reported data, arrest data overstate the magnitude of sex, race, and age differentials in the prevalence of serious violent offending. Further, self-reported individual offending rates were also substantially greater than those based upon arrest history data and there were no significant differences in self-reported individual offending rates by sex, race, class, or age. These findings suggest that prisoners incarcerated for serious violent offenses are not representative of those in the general population who are involved in serious violent offenses. The analysis of career length and crime mix patterns suggested that the period of involvement in violent offending was short (one year), individual offending rates increased with time, and the crime mix pattern became more diversified with time in the career.