Poster #38 - Gender as a predictor of parental favoritism

• In Event: Poster Session 08

Abstract

Parents often have valid reasons for treating their children differently. For example, they may have different developmental needs (Jensen et al., 2020; Kowal & Kramer, 1997) that are based in their dyadic differences (Jensen et al., 2022). Despite the reasons why parents may treat their children differently, the extant literature suggests that children who receive less favored treatment tend to have poorer developmental outcomes, including more externalizing behaviors (Boyle et al., 2004; Peng et al., 2018), more internalizing behaviors (Meunier et al., 2012; Shanahan et al., 2008), and negative family relationships (Boll et al., 2005; Kowal et al., 2004).
Because of the negative outcomes associated with being less favored, it is critical for scholars to examine which children may be favored or less favored by their parents. Although few studies have explicitly examined this question, enough data exist in the literature on parental favoritism to assess the link between gender and favoritism.
Data for the current analysis came from 29 unique samples, 57 sources nested within those samples, and 378 effect sizes nested within those sources. Overall, the analysis reflected data collected on 15674 unique children (M Age = 19.39, SD = 13.55). Sources for the data came from published studies, doctoral dissertations, masters’ theses, and unpublished raw data shared from several scholars.
Meta-analysis often collapses multiple effect sizes within each study. In the current analysis, however, we took advantage of the variation that exists within studies and within samples by using a four level multi-level meta-analysis. To further leverage the multi-level model, we examined moderators at each level to see whether the overall correlation between favoritism and relative birth order varied by characteristics of the sample, source, and/or effect size. We ultimately excluded moderators from the source level (level three) because analysis showed that there was not significant variation at that level.
Our findings suggest that overall, parents did not tend to favor either sons or daughters (see Model 1 in Table 1). Although though were no main effects, multiple moderators showed significant interactions (see Model 2 of Table 1). Specifically, in the United States (b = -.18, se = .04, p < .0001) and in countries other than the United States (b = -.12, se = .04, p < .001), parents reported that they favored daughters more than sons, but the effect was stronger in the United States. Additionally, different effects emerged for the reporter of favoritism. Specifically, children reported no association between gender and favoritism (b = .02, se = .03, p > .05) but parents reported favoring daughters more than sons (b = -.12, se = .04, p < .001).
Although the effect sizes in our analysis were generally small, our findings have implications for clinicians working with families struggling to manage or resolve issues associated with parental favoritism. Parents may need to be aware of an inclination favor daughters over sons and the implications that may have for their children’s development.

Authors

• Alexander C. Jensen, Brigham Young University
  Presenting Author

• Alexandra Burns