Commentary: Are siblings birds of a feather? – reflections on Jenkins et al. (2012)

Abstract

Sibling relationships have garnered increasing research attention, though they still lag far behind parent–child and marital relations. As Jenkins and colleagues mention in their accompanying article (Jenkins, Rasbash, Leckie, Gass, & Dunn, 2012), the developmental sequelae of sibling relationship quality, including internalizing and externalizing behaviour problems, are well documented. Sibling conflict itself is also a persistent problem for many parents, frequently cited as a reason for attending family therapy, and sibling violence is the most common form of domestic abuse (Naylor, Petch, & Ali, 2011). For all of these reasons, identifying predictors of sibling relationship quality is important. Antecedents to sibling relationship quality are also valuable windows into the process of socialization within families. Traditional theories of socialization are top-down, giving parents both the power and responsibility to influence their children’s behaviour, including their relationships with one another. In support of this, research has consistently linked parent–child and sibling relationships (e.g. Pike, Coldwell, & Dunn, 2005). Taking this logic to its extreme, one would predict that where families include three or more children, the sibling relationships within families would be similar in character. That is, according to this top-down model, a mother and father in a high conflict marriage will likely engage in coercive, harsh parenting, and have children with poor quality sibling relationships (thankfully, the reverse would also be true). Until recently, the analytic tools have been lacking to incorporate more than two children per family in systematic, quantitative research. It was thus not possible to test the similarity of sibling relationships within families. Multi-level modelling applied to family data has made it feasible to tease apart betweenand within-family variance. In the present context, this enabled Jenkins and colleagues to quantify individual differences in sibling relationship quality into aspects shared by children in the same family (between-family effects) and those aspects unique to each sibling dyad within a family (within-family effects). This differentiation was the first goal addressed by the authors, with their emphasis being the identification of significant between-family variance. While significant betweenfamily variance was identified for both positive and negative aspects of sibling relationship quality, the within-family variance was also significant and greater in magnitude. This finding is not the main focus of the article, but I believe it to be the more important and groundbreaking result. This is only the second time that such an analysis has been reported, and the first time without mono-reporter bias. What does it mean? The moderate between-family effects are supportive of the traditional socialization perspective outlined above. But the larger within-family variance indicates that children from 3+ child families can and do experience different sorts of relationships with their different brothers and sisters. The implications of this finding are not limited to (relatively) large families. This result indicates that individual characteristics of the children themselves (such as personality, social competence, mental health, etc.), and the unique interaction of these characteristics among dyads accounts for the lion’s share of individual differences in sibling relationship quality. My point is not to let parents ‘‘off the hook,’’ especially as children’s characters develop within the context of parent–child relationships. However, family systems theory along with traditional socialisation theory has perhaps swung too far in the direction of holding parents responsible for their children’s behaviour, including their relationships with brothers and sisters. A positive outcome of recognizing that the antecedents to sibling relationship quality do not reside entirely with parents is that it may be possible to empower parents to engage in effective conflict resolution strategies without the concomitant blame. What of the second goal addressed by Jenkins and colleagues? Here the findings are not new: maternal malaise (depressive symptoms) and parenting were predictive of subsequent sibling relationship quality. However, these findings were contextualized as explaining between-family differences. Thus, the mechanism of these maternal factors is identified as operating at the level of the family – influencing the overall family climate, including the warmth and/or hostility of sibling relations. As the authors point out, such an approach could also be used to identify factors operating at the within-family level, and factors that interact across these levels. Specifying determinants of within-family variance in sibling relationships is foreshadowed by research on children’s experiences that are not shared with their siblings (see Turkheimer & Waldron, 2000). Journal of Child Psychology and Psychiatry 53:6 (2012), pp 630–631 doi:10.1111/j.1469-7610.2012.02536.x