Référence bibliographique 
British Journal of Educational Psychology, vol. 87, no 3, p. 456-477.
«This study examines the differential associations between a wide range of key risk factors and high school dropout as a function of immigration status. [T]he study pursues three main objectives. First, we examine whether dropout rates vary between first-, second-, and third-generation-plus students. Second, we assess the differences between these groups in terms of individual, social, and family risk factors at the beginning of secondary school. Finally, we aim to estimate the extent to which the direct associations between these early adolescence characteristics and school dropout vary for first-, second-, and third-generation students.» (p. 459)
«[W]e hypothesized that poverty would have a lower negative impact on school dropout for first- and second-generation students compared to their third-generation-plus counterparts.» (p. 459)
«The sample was drawn from a larger population-based sample of 30,000 students from 77 public secondary schools in Quebec Province (Canada) […]. These students were all attending public schools located in low socio-economic neighbourhoods. […] Overall, 3109 students participated in the project.» (p. 460)
Type de traitement des données :
«[T]his study proposes that student’ immigration status matters and should be considered when implementing prevention strategies against school dropout. Three main findings stemming from this large-scale longitudinal sample of low socio-economic youths support that conclusion and merit further discussion. […] First, our results highlight that immigrant’s students, albeit exposed to less favourable economic environment, showed better school and psychosocial adjustment than their non-immigrant peers. […] Second, although the dropout rate of first- and second-generation students was similar, these students differed on some academic, family, and social dimensions from the beginning of high school. As compared with their second-generation peers, first-generation immigrants presented more positive attitudes and beliefs about school and shared warmer and less conflictual relationships with teachers. […] Third, a key set of findings concerns the identification of many differential associations between individual, psychosocial, and family characteristics and first-, second-, and third-generation- plus students’ dropout. Differential associations emerged for over 40% of the factors considered in the present study, and many of these differential associations remained significant after considering all risk and protective factors in a single model.» (p. 471-472)