Table 3
Variation in violent delinquency across schools

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Variation in violent delinquency across schools
  Empty model School neighbourhood characteristics model School characteristics model Student characteristics model Full model
  odds ratio
Fixed effects  
Intercept 0.15 ** 0.13 ** 0.06 ** 0.03 ** 0.01 **
School-neighbourhood characteristics  
Socioeconomic disadvantage 1  
Low 2 1.00 1.00
Middle 1.14 0.94
High 1.08 1.00
Police-reported crime rate 1  
Low 2 1.00 1.00
Middle 1.36 * 0.93
High 1.00 0.90
School characteristics  
School capital 1  
Low 1.53 ** 1.49 **
Middle 1.41 * 1.50 **
High 2 1.00 1.00
School disorder 1  
Low 2 1.00 1.00
Middle 1.44 ** 1.25
High 1.45 ** 1.41
School status  
Public 1.70 1.56
Private 2 1.00 1.00
Student characteristics  
Gender  
Male 1.96 ** 1.99 **
Female 2 1.00 1.00
Grade level  
Grade 7 1.59 ** 1.85 **
Grade 8 1.52 ** 1.68 **
Grade 9 2 1.00 1.00
Low parental monitoring  
Yes 3.60 ** 3.53 **
No 2 1.00 1.00
Friends accept criminal behaviour  
Yes 4.28 ** 4.24 **
No 2 1.00 1.00
Usually does not like school  
Yes 1.59 ** 1.57 **
No 2 1.00 1.00
  variance
Random effects  
Variance component (µ 0j) 0.215 ** 0.207 ** 0.106 0.153 * 1.076
  percent
Prevalence and variation across schools 3  
Average likelihood in a 'typical' school 13.2 11.4 5.8 2.6 1.2
Low prevalence 5.8 5.0 3.1 1.2 0.7
High prevalence 27.4 23.8 10.4 5.4 2.0
*
statistically different from the reference category p ≤ 0.1; or for variance components, regression coefficients in the model vary statistically among schools p ≤ 0.1
**
statistically different from the reference category p ≤ 0.05; or for variance components, regression coefficients in the model vary statistically among schools p ≤ 0.05
1.
For this analysis, neighbourhood and school variables were divided into tertiles, or low, middle and high scores, each containing one-third of schools.
2.
Reference category in models.
3.
See details on calculation of prevalence and variation across schools in 'Calculating between-school variation in prevalence of violent delinquency' in the Methodology section.
Note(s):
Unweighted sample size = 3,137 students attending 149 Toronto schools.
Source(s):
Statistics Canada, International Youth Survey, 2006, Census, 2006 and Incident-based Uniform Crime Reporting Survey, 2006.