The Interparental Relationship and Children's Behavior: Conceptual Issues and a Meta-analytic Integration

• In Event: 3-116 - The interparental Relationship, Parenting, and Child Adjustment: New Insights from State-of-the-Art Methods

Integrative Statement

In the last decades, new theories have developed and empirical research has increasingly focused on multiple specific dimensions of the interparental relationship in relation to children’s problem behavior (e.g. Davies & Cummings, 1994; Grych & Fincham, 1990). However, a systematic integration of this growing corpus of findings is lacking and complicated due to two challenges. First, there is no consensus about the multidimensionality of the interparental relationship. Second, a jingle-jangle fallacy is evident in the wide differences in the use of conceptual labels for similar or identical measures. Therefore, building on the meta-analysis of Buehler and colleagues (1997), the overall aim of this meta-analysis was to provide a renewed and expanded empirical integration of existing research on linkages between the interparental relationship and children’s externalizing and internalizing problem behavior. In order to reach this aim, we first needed to tackle a foundational question: what is the interparental relationship? A conceptualization recognizing nine distinctive dimensions was formulated, based on measures in the current empirical field (i.e. omnibus adjustment, satisfaction, negative quality, conflict frequency, hostility, disengaged behavior, constructive behavior, child-related conflict, and composite conflict). Finally, this conceptualization guided our empirical integration and we used state-of-the-art multi-level meta-analytic techniques to provide in-depth and comprehensive knowledge about how different dimensions of the interparental relationship are related to externalizing and internalizing problem behavior for children from 0 to 18 years old. Studies with children up to 18 years old (k = 230 samples from 226 studies) that reported on associations between the interparental relationship and children’s problem behavior were included. Information from 1722 effect sizes was analyzed in a three-level meta-analysis, which is an extension of the traditional two-level random-effects model in which the dependency among multiple ESs from the same study is modeled by adding an intermediate level. Results showed an overall small association between (negativity in) the interparental relationship and externalizing and internalizing problem behavior. Moreover, associations appeared to be overall stronger for externalizing behavior than internalizing behavior. Also, all dimensions, except satisfaction, showed significant, small correlations with both externalizing and internalizing behavior. For externalizing behavior, child-related conflict appeared to be most important, for all children independent of age and gender, whereas for internalizing behavior, multiple dimensions were equally important. Finally, few moderating effects were shown. Associations between conflict frequency and problem behavior became stronger when the age of the children increased. Family composition did matter for associations between hostility and externalizing behavior and composite conflict and internalizing behavior. Moreover, almost all associations depended on common method variance. Overall, this meta-analysis provides a few conclusions that are interesting for further theory-building (i.e. modeling hypothesis, cognitive-contextual model and emotional security hypothesis). And importantly, this study shows that there is indeed compelling evidence that children of all ages, gender, and in different family compositions, who witness frequent, poor resolved (i.e. hostile, disengaged, low constructive) or child-related conflict between their parents are at risk for developing behavioral problems, underlining that the interparental relationship can be a relevant starting point for support aimed at improving youth outcomes.

Authors

• Willemijn Maria van Eldik, Erasmus University Rotterdam
  Presenting Author

• Amaranta D de Haan, Erasmus University Rotterdam
  Non-Presenting Author

• Patrick Davies, University of Rochester
  Non-Presenting Author

• Lidia R Arends, Erasmus University Rotterdam
  Non-Presenting Author

• Maartje Luijk, Erasmus University Rotterdam
  Non-Presenting Author

• Peter Prinzie, Erasmus University Rotterdam
  Non-Presenting Author