1. Objectifs

Intentions :
« The first goal of the study was to identify the independent contribution of each level of analysis (individual, family, and neighborhood) on variation in the frequency of physical aggression and prosocial behavior. The second goal was to identify the independent contribution of each selected variable within each level of analysis. » (p. 567)

2. Méthode

Échantillon/Matériau :
« Participants were 2,745 2–11-year olds nested in 1,982 families, which were themselves nested in 96 Canadian neighborhoods. » (p. 567)

Instruments :
Objective census tract data on area size and poverty de Statistique Canada;
Data of The National Longitudinal Survey of Children and Youth (NLSCY) are drawn from :
- Revised Simcha–Fagan Neighborhood Questionnaire;
- Barnes McGuire’s (1997) revised Simcha–Fagan Neighborhood Questionnaire;
- Center for Epidemiological Studies Depression Scale;
- McMaster Family Assessment Device;
- Social Provisions Scale;
- Parent Practices Scale.

Type de traitement des données :
Analyse statistique

3. Résumé

« The study identified independent individual, family, and neighborhood correlates of children’s physical aggression and prosocial behavior. [...] Hierarchical linear modeling showed that the total variation explained by the three-level model was 28.03% for physical aggression and 17.57% for prosocial behavior. For both childhood behaviors, approximately 66% of this explained variance was between individuals and up to 30% was between families. The smallest amount of observed variation was between neighborhoods. Significant individual-level predictors common to both childhood behaviors were child’s sex and maternal hostility toward the target child. Specifically, boys had more mother-reported physical aggression and less prosocial behavior. Children who experienced greater-than-average maternal hostility (compared to siblings) were more physically aggressive and less prosocial. At the family level, significant common predictors were mother depressed mood and punitive parenting. Children had higher levels of physical aggression and lower levels of prosocial behavior in families where mothers had greater depressed mood and used more punitive parenting practices. At the neighborhood level, greater perceived problems and lower poverty level were associated with higher levels of physical aggression. Results are discussed with reference to past and future studies of multilevel effects on children’s socialization. » (p. 565)