Is discrimination a driving force behind the over-representation of children with an immigrant background in Swedish out-of-home care? A quantitative study from Stockholm City

Abstract

ABSTRACT To research potential discrimination in concluding decisions of child-welfare investigations, i.e. to place children into out-of-home care or let them remain at home while providing services, this study applies logistic regression to a random sample of 132 (n = 132) investigations conducted by Stockholm City, Sweden, in 2014. First, the risk of entry into care associated with a set of practice-related factors was estimated. Second, the interactions between the significant factors and having first or second-generation immigrant background from outside of Europe was estimated using interaction terms. All interactions were insignificant: Lone parenting: OR = 0.301, 95% CI: 0.052–1.737; Number of reporters: OR:0.405, 95% CI: 0.082–1.964; Social problems in the child: OR = 0.788, 95 CI: 0.140–4.706; Neglect: OR = 2.210, 95% CI: 0.239–27.327; Number of social problems in the parent(s): OR = 2.209, 95% CI: 0.797–8.184; Previous placement: OR = 0.236, 95% CI: 0.017–2.950. Discrimination is therefore not indicated by the results. The large over-representation of immigrant children in Swedish out-of-home care is therefore more likely driven by disproportionate exposure to common risk factors for child welfare involvement (cf. Drake et al., 2011. Racial bias in child protection? A comparison of competing explanations using national data. Pediatrics, 127(3), 471–478. https://doi.org/10.1542/peds.2010-1710). Poor housing conditions, poverty and neighbourhood segregation are of relevance for Stockholm City.