"Comparing Profiles of Mother and Teacher Rated Prosocial Behaviors: The Role of Gender and Ethnicity"
SRCD 2019 Biennial Conference
March 21-23, 2019
Title: Comparing Profiles of Mother and Teacher Rated Prosocial Behaviors: The Role of Gender and Ethnicity (Poster)
Session: Moral Development
Authors: Su Jiang, Ting-Lan Ma, Sandi Simpkins, Deborah Vandell, Nicole Zarrett
Abstract: Studies of prosocial behaviors have typically used a variable-centered approach to examine children’s propensity to help, offer comfort, and show concern for peers. There is some evidence (Veenstra et al 2008) that teachers and mothers use different criteria for rating children’s prosocial behavior. Less is known about how mothers and teachers’ perspectives of child prosocial behavior differentiate from each other and what contributes to the difference. This study took a pattern-centered approach to explore the profiles of mother- and teacher-report 4th grade prosocial behavior to understand the shared and unique perspectives of mothers and teachers, and to examine how gender and ethnicity associated with the differences between mothers and teachers’ perceptions of child prosocial behavior.
Data were drawn from the NICHD study of Early Childcare and Youth Development (n=877; 47% girls; 78% White, 10% African American, 7% Hispanic students). Mothers and teachers reported on nine aspects of children’s prosocial behaviors (e.g. friendly toward other children) at 4th grade (Ladd & Profilet, 1996). We conducted latent profile analysis on the 9 items in Mplus 8.0 using full information maximum likelihood estimation with robust SEs.
The four-class solution as the best fit for both mother and teacher reports based on multiple fit indices. The profiles consisted of children who are rated as highly prosocial on all 9 behaviors (labeled high prosocial) (n=621 for mothers, n=595 for teachers), highly friendly but less kind (labeled friendly but less kind) (n=166 for mothers, n=67 for teachers), highly kind but less friendly (labeled kind but less friendly) (n=37 for mothers, n=58 for teachers), and low in all prosocial behaviors (labeled low prosocial) (n=53 for mothers, n=157 for teachers).
Between mothers and teachers’ latent profiles on prosocial behavior, there is more consistency in high prosocial and low prosocial groups and less consistency in groups in between, for example, students who were rated friendly but less kind by mother overrepresented in teacher report low prosocial group. Chi-square tests suggest that girls and White students were overrepresented in the high prosocial group as rated by mothers and teachers. In contrast, African American students were overrepresented in the low prosocial group as rated by both reporters. Multinomial logistic regression tests were conducted to predict teacher report prosocial profile numbers (1-low prosocial, 4-high prosocial) using gender and ethnicity controlling for mother report profiles. Results suggest that comparing to mothers, teachers would rate males and African American students into low prosocial profile.
Mother and teacher reports of students’ 4th grade prosocial behavior demonstrated the consistent patterns of children’ prosocial behavior. However, the disparity of the composition of each profile between mothers and teachers’ perceptions of children prosocial behavior not only demonstrates mothers and teachers use difference criteria for rating children’s prosocial behavior, it also underscored the existence of the stereotypic gender roles that females are expected and believed to be friendlier and more prosocial, and the ethnicity bias that African American students are less prosocial than other ethnic groups.