Poster #127 - Multiple types of prosocial behavior in adolescence: A meta-analysis of PTM-R

• In Event: 2-110 - Poster Session 07
  In Poster Session: PS 07 Section - Moral Development

Integrative Statement

Prosocial behavior refers to voluntary acts intended to benefit others (Eisenberg et al., 2015). It is important to examine adolescents’ prosocial behavior because it is related to positive outcomes (Caprara et al., 2000) and has been shown to be a protective factor against problem behaviors (Carlo et al., 2014). Researchers argue that prosocial behavior is a multidimensional construct and there is a need to examine various types of prosocial behavior because each type may show different patterns of development (Eisenberg et al., 2015). Further, measures of prosocial behavior and study characteristics can impact our understanding of prosocial behavior, particularly the potential gender differences between girls and boys (Eisenberg et al., 2006; Fabes & Eisenberg, 1998; Hine & Leman, 2013). In this study, we aimed to examine gender differences in the Prosocial Tendency Measure - Revised (PTM-R; Carlo et al., 2003), which is the most often used prosocial measure designed to capture six types of prosocial behavior during adolescence (i.e., compliant, public, anonymous, dire, emotional, and altruistic).
Article selection processes are presented in Figure 1. Articles were selected using the following criteria: use of PTM or PTM-R, inclusion of both males and females, adolescents’ age between 10 at 18 years, written in English, and location of the participants within the U.S. Two researchers independently reviewed the full-text of 37 articles. We temporarily excluded 19 articles that did not include necessary information (e.g., gender information, effect sizes). We plan to contact authors for missing information. A total of 18 articles (10 studies and 70 effect sizes) were independently coded by two reviewers and used for preliminary analyses. Good reliability was established for the reported effect sizes and characteristics (s = .70-1.0, ICCs = .79-1.0). If reported, seven types of prosocial behavior (i.e., overall, compliant, public, anonymous, dire, emotional and altruistic) were coded as separate effect sizes. We conducted a three-level random-effects model using structural equation modeling (Cheung, 2014) to account for the clustering of effect sizes within studies.
Preliminary results suggest there are mean-level differences across various types of prosocial behavior between adolescent girls and boys, although these differences are small in magnitude, absolute d = .24, 95% CI [.17, .30], level 2 2 = .0113 and I2 = 40.28%, level 3 2 = .0034 and I2 = 12.29%, Q (69) = 142.52. Effect sizes did not significantly vary between the six types of prosocial behavior. There were slightly larger gender differences using overall prosocial behavior (absolute d = .33), followed by the altruistic subscale (absolute d = .30), and the smallest effect size was found for the anonymous subscale (absolute d = .14). The mean age of the sample did not explain differences between studies (p = .120). We will rerun our analyses after the remaining 19 articles are coded and include additional predictors (e.g., racial/ethnic composition, reporter type, measurement reliability). Together, our initial findings suggest that gender is an important factor to consider in the research of adolescent prosocial development.

Authors

• Emi Hashi, Arizona State University
  Presenting Author

• Sonya Xinyue Xiao, Arizona State University
  Non-Presenting Author

• Kevin M. Korous, Arizona State University
  Non-Presenting Author