Relations Between Parental Affect and Parenting Behaviors: A Meta-Analytic Review

SYNOPSIS Objective. The goal of the present investigation was to provide a meta-analytic review of the research on affect and parenting in nonclinical samples. Design. The authors conducted analyses on the overall mean effect size for 63 studies (k = 18,211). Affect was coded as either positive or negative, and parenting behavior was coded as either supportive-positive or harsh-negative. Moderators included definition of affect, time frame of measurement, reporter, child age, and parent gender. Results. The authors' analyses support the association between parental affect and parenting behavior, and this relation was consistent across types of affect and parenting, as well as gender of parent. Significant methodological moderators of these relations include time frame match and reporter match. Child age moderated the relation between negative affect and supportive-positive parenting, but not harsh-negative parenting. Last, both negative and positive affect showed specificity in their association with parenting behavior. Conclusion. Parental affect appears to be a reliable correlate of parenting behaviors in the general population. In addition, consistent with theory (T. CitationDix, 1991), negative affect was more strongly related to hostile parenting, and positive affect was more strongly related to supportive parenting. Methodological factors, such as time frame of measurement and reporter, as well as demographic variables, should be carefully considered in the design and interpretation of future studies of parental affect and parenting behaviors. Notes 1There were other articles in the present study that addressed positive affect that was not pure trait affect, but these could not be included in these analyses for the following reasons: CitationLorber and O'Leary (2005) was coded as mixed positive affect, but the effect size was reverse coded and included with studies focused on mixed negative affect because it represented the happy–sad construct. CitationWeis and Lovejoy (2002) included state and trait affect, and when an overall effect size was calculated, this study could not be classified as either state or trait and was excluded from the analyses. Studies included in the meta-analysis are marked with an asterisk.