The Correlates of Self-Reported Delinquency: An Analysis of the National Longitudinal Survey of Children and Youth

4. Discussion

The annual self-reported delinquency rate in our sample (39%) was substantially higher than the official rate of 5% found in the Uniform Crime Reports (UCR). [7] Our rates of property offending (39%) and violent offending (15%) were also substantially higher than official rates. UCR data indicated that the prevalence rate for property crime was approximately 3%, while the violent offence rate was less than 1% of all youth. Discrepancies between official rates and self-reported rates are to be expected. In fact, according to Moffitt (1993), although only a portion of youth come to the attention of police, deviant behavior is so prevalent during adolescence that youth who never engage in offending are considered statistical anomalies.

In terms of gender, our results indicate that the official data sources also vastly underestimate the extent of female delinquency. We found that females constituted approximately 44% of all youth who reported engaging in delinquent behavior while the official data from the UCR indicate that females constituted 22% of all youth charged. Female youth were less likely, however, to report serious and frequent offending, which may partially explain this discrepancy. Police may not be as likely to charge youth for the minor offending that is common among female youth.

While the Aboriginal self-reported delinquency rate (41%) was similar to non-Aboriginal youth (39%), Aboriginal youth were more likely to report engaging in more serious offences. Unfortunately, we were unable to compare our sample to official data sources, as Aboriginal status is not adequately reported to the UCR.

In general, the correlates identified in this study provide further support to the findings within the literature. The primary correlates of general delinquency are negative school attachment, anti-social peers, victimization, aggression, and negative parenting. The findings are encouraging in that these are primarily dynamic factors amenable to change through targeted interventions with youth and their families.

While unique correlates were identified for female and male delinquency, the broad categories of negative school attachment, anti-social peers, aggression, and negative parenting were still present. The central difference was that victimization was strongly correlated with female delinquency and poor motivation was uniquely correlated with male delinquency. Appropriate gender-based interventions designed to reduce recidivism, therefore, should provide an increased focus on reducing the consequences of victimization for females and increasing self-motivation for males.

Sexual offending was not strongly related to any of the factors we tested. Clearly, the pathways into sexual offending are rather unique and may be related to a different set of independent variables.

For violent offending, the same five core concepts (school, peers, aggression, victimization, and parenting) emerged with the addition of age, self-esteem, hyperactivity/inattention, and gender. That is, violent youth displayed attributes similar to other delinquents, but tended to be older, hyperactive males who presented a positive self-image.

For property offending, four of the five core concepts entered the model with negative parenting being the notable exclusion. Gender, hyperactivity/inattention, and socio-economic status were also correlated with property-related offending. In other words, property offenders, as a unique group of delinquents, tended to be hyperactive males with higher socio-economic status.

Drug trafficking was also correlated with four out of the five core concepts. Victimization, however, did not enter the model.


[7] The UCR data in this report represents all youth aged 12 to 17 charged with a common offence ( Criminal Code and other federal statute offences) across all 13 provinces/territories from the same year as the NLSCY data ( 1998).

Date modified: