1. Objectifs

Intentions :
«[T]he current meta-analysis aims at providing a robust estimate of the association between family and peer factors susceptible to either increase or decrease the risk for victimization in adolescents’ and emerging adults’ dating relationships.» (p. 2)

2. Méthode

Échantillon/Matériau :
«A computer search of the PsycInfo and Pubmed databases, Google Scholar, as well as other French databases available at was conducted between May and December 2015. […] This process led to a final sample of 87 studies (scientific articles, dissertations, and research reports […]).» (p. 3)

Type de traitement des données :
Analyse statistique

3. Résumé

«First, we examined if the strength of the associations between the various factors and DV [dating violence] differs according to the specific form of violence sustained. Second, we quantitatively summarized two classes of determinants of DV, namely family and peer factors, in one study, allowing the comparison of effect sizes between the two groups of factors. Finally, we are aware of no systematic effort to summarize knowledge about the family and peer-related protective factors for DV. […] With respect to family risk factors, results show significant associations between all forms of maltreatment (sexual abuse, psychological abuse, neglect, witnessing IPV, and physical abuse) and DV victimization, with effects sizes of small magnitude. The results also suggest no difference in the strength of the associations across the various types of maltreatment and DV, suggesting that no particular form of childhood abuse or neglect seems to have a greater influence than the others on DV and also that each form of DV was impacted in a similar way by the risk and protective factors included in this meta-analysis. […] Peer victimization, sexual harassment from peers, and affiliation with deviant peers were examined as potential risk factors for victimization in dating relationships. All three factors evidenced an association a small magnitude with DV, with peer sexual harassment almost reaching the threshold for a moderate effect.» (p. 8-9)