Montreal Longitudinal Study, Canada

This research by the University of Montreal was to check the validity of previous findings that physical aggressiveness and academic problems are predictors of delinquency that are identifiable early life. The research candidates were therefore identified in kindergarten, although the programme did not commence until they were aged 7. All the families were of low socio-economic status.

Project design for Montreal Longitudinal-Experimental Study

Kindergarten teachers rated the behaviour of each of the boys in their classes at the end of the school year. 1,037 boys from 53 schools who were identified as being at-risk (i.e. disruptive, hyperactive, aggressive) were selected for the longitudinal study and randomly assigned to 3 groups: the treatment group, created for the experimental study of prevention; the observation group, created for the longitudinal observational study of the social interactions of disruptive boys; and, the no-treatment, no-contact control group, for evaluating effects of the experimental study and of the longitudinal follow-up.

"An effective intervention at age 7 was able to remove two thirds of the disadvantage reflected in the behaviour of a group of disruptive kindergarten children."

Post-treatment, behaviour was assessed annually (from age 9 to age 12) by teachers, peers, mothers and the boys themselves. For all of the boys, ratings were obtained on a number of indicators: educational achievement, fighting behaviour, overall school behaviour and performance, delinquent behaviour, mothers' perception of antisocial behaviour and the nature of parent-child relationships, making it possible to assess the impact of intensive treatment on antisocial, disruptive behaviour.

Findings

By the end of primary school, the behaviour of the disruptive boys in the untreated group confirmed previous research findings that physical aggressiveness and academic problems, are predictors of delinquency that are identifiable early on in a child's development. The research also confirmed that social intervention can positively affect the social development of disruptive boys. Compared with the untreated boys, the boys who received the treatment exhibited less aggression in school, performed well academically more often, experienced fewer difficulties in adjusting to school, and reported committing fewer delinquent acts up to three years after the end of treatment. With respect to teacher-rated fighting, the boys in the control and observational groups had a significantly higher fighting score than those in the treated group. Untreated boys were twice as likely to be rated as having serious school behaviour and school performance problems as the boys in the treatment group (44% vs. 22%). A smaller proportion of the treated, than of the untreated, reported having committed delinquent acts involving trespassing (40% vs. 62%), stealing items worth less than $10 (19% vs. 45%), stealing items worth more than $10 (7% vs. 20%), and stealing bicycles (5% vs. 19%). No significant differences were found between the treated and untreated boys in terms of hyperactivity, prosociality and vandalism. There was also a brief positive update at age 15.

A follow up study when the boys were aged 24 found that, compared with the control group, the boys in the treatment group had significantly higher levels of high school graduation and with lower rates of acquiring criminal records. Boisjoli et al (2007) also compared the two groups with a normal or ‘normative’ group of young men of the same age, parental occupation mix etc. These comparisons are shown in the table below.

Using the ratio of High School Graduations to numbers with criminal records at age 24, one could crudely deduce that an effective intervention at age 7 was able to remove two thirds of the disadvantage reflected in the behaviour of a group of disruptive kindergarten children. Richard Tremblay, one of the authors of this study, argues that the most effective time to intervene is when differences in aggression show (as he demonstrates that they do, with predictive validity) at ages 2 and 3. An effective intervention delivered at that age might be expected to achieve an even stronger result.