Child maltreatment and insecure attachment: a meta‐analysis

Abstract

The validity of the insecure/disorganized attachment pattern has been clearly established in over 80 studies with more than 100 samples. However, given that child maltreatment is considered one of the most important causes of insecure/disorganized attachment, it is notable that few studies have been published specifically examining the maltreatment/insecure attachment hypothesis. The purpose of this paper was to review the research findings and conduct a meta‐analysis to examine the effect size of maltreatment and insecure attachment across studies. A second goal was to conduct a subpopulation analysis to investigate effect size by type of maltreatment. Following a literature search which yielded 25 articles, we identified eight studies (n = 791) that: (a) consisted of children who experienced different types of maltreatment including cases of malnutrition and failure‐to‐thrive, (b) consisted of children under 48 months of age, (c) used the Strange Situation or a modification of it to measure attachment, (d) had comparison groups, and (e) contained sufficient detail to conduct a meta‐analysis. Results showed that the maltreated infants were significantly more likely to have an insecure attachment than the controls. A subpopulation analysis by type of abuse was conducted. Overall the findings support the maltreatment/insecure attachment hypothesis. The subpopulation analysis, although extremely small, indicates that different types of maltreatment affect the magnitude of the effect. These findings suggest further investigation into the differential impact by type or multiple types of maltreatment as well as the frequency of events on outcomes.