Poster #83 - The Relation Between Social Fearfulness and Prosocial Behavior: A Meta-Analysis

• In Event: Poster Session 01

Abstract

Prosocial behavior is a hallmark of social competence and is related to positive psychosocial outcomes, such as being more accepted by peers and experiencing fewer externalizing behaviors (Howes & Phillipsen, 1998; Layous et al., 2012). Yet whether children behave prosocially varies due to individual and situational factors. One individual-level factor that has been linked to prosocial behavior is social fearfulness, which refers to distress in social situations (e.g., shyness, social anxiety, behavioral inhibition; Eisenberg et al., 2006; Goldsmith, 1996). The findings on this relation, however, have been inconsistent, with many studies reporting negative links (e.g., Beier et al., 2017), but others finding null or even positive relations (e.g., Farmer & Rodkin, 1996). Thus, a synthesis of these research findings is needed, as well as an examination of factors that may moderate this association. We conducted a systematic review and meta-analysis on the relation between social fearfulness and prosocial behavior and tested the moderating role of situational and demographic factors on this relation.

Findings reported here are preliminary as only 32% of studies we have identified as relevant have been coded thus far. Final analyses will also include additional studies (e.g., articles found in updated searches). Our protocol was preregistered with the PROSPERO prospective systematic review database (CRD42022309600). Searches—the parameters for which are detailed in Table 1—yielded 2996 unique abstracts. We followed standard meta-analytic procedures for establishing reliability (90%+ agreement) between our 3 coders at each stage (i.e., abstract screening, full-text screening, and data extraction). So far, we have coded 50 relevant studies and 131 effects involving 19,627 participants. To avoid discarding relevant data, we extracted all effect sizes relevant to our research question and used a three-level meta-analytic approach to account for the nonindependence of effects within the same study (Cheung, 2014). We calculated an overall pooled correlation and assessed the amount of heterogeneity in our findings using the metafor package in R (Viechtbauer, 2010). We also tested several potential moderators: children’s age (sample mean age), gender (percent of sample that are girls), familiarity with the recipients of their prosociality (unfamiliar vs. unfamiliar), and the emotional intensity of prosocial situations (none/low vs. moderate/high).

Our meta-analysis yielded a small though significant pooled correlation of r= -0.076 (95%CI: -0.127, -0.025, p=.004). We observed a high degree of heterogeneity attributable to differences between studies (24%) and within studies (68%), indicating that moderator analyses were warranted. We found that older samples and those with a higher composition of girls showed weaker correlations between social fearfulness and prosocial behavior, though the latter was marginally significant (respectively, β=.0128, SE=.005, p=.005; β=.0011, SE=.0006, p=.075). We also found a marginal effect of familiarity with recipients (p=.069), but no effect of emotional intensity (p=.363). Findings for each level of these categorical moderators are listed in Table 2.

In sum, these findings demonstrate a link between social fearfulness and prosocial behavior, which may vary depending on individual and situational characteristics. We will discuss these findings in relation to current theorizing on social fearfulness and prosocial development.

Authors

• Sebastian P Dys, University of Waterloo

• Sarah D English, University of Waterloo

• Mya Dockrill, University of Waterloo
  Presenting Author

• Heather Henderson, University of Waterloo