Explicit and Implicit Self-Esteem Diverge From Each Other and Have Distinct Academic Correlates During Childhood

• In Event: Behavioral Consequences of Children’s Self-esteem During Childhood

Abstract

Self-esteem is an essential ingredient of personality in both adults and in children. Two kids of self-esteem have been identified and studied in the literature: explicit (conscious, self-reflective feelings of self-liking, self-worth, and acceptance; Zeigler-Hill, 2006) and implicit (associating oneself with positive or negative valence outside of awareness, e.g., me = good; Greenwald et al., 2002). In social psychology studies with adults, measures of explicit and implicit self-esteem have been found to be only weakly correlated, and they are widely considered differentiable constructs with unique behavioral correlates and outcomes (Greenwald & Farnham, 2000). This (potential) implicit–explicit differentiation is understudied in children.

In this cros-sectional study, we used both explicit and implicit measures of self-esteem in Grade 1 and Grade 5 children. This design allowed us to evaluate grade-related differences, gender differences, and the relations of both measures of self-esteem to achievement in math and language, two subjects that are gender-stereotyped (i.e., “reading is for girls” and “math is for boys”). Examining both types of achievement in the same children and in relation to their self-esteem offers an opportunity to obtain a targeted assessment of whether self-esteem relates to children’s acquisition of gender stereotypes about academic subjects during the time when such stereotypes are just forming.

Across eight Croatian elementary schools, 376 students (190 girls) participated in either Grade 1 (N = 186; Mage = 7.53 years) or Grade 5 (N = 190; Mage = 11.54 years). The explicit measure of self-esteem was Harter’s (1982) Self-Perception Profile for Children (SPPC). The implicit measure of self-esteem was a child-friendly adaptation of the Implicit Association Test (IAT) called the Child IAT (Cvencek et al., 2020). In the self-esteem Child IAT, children were tasked with sorting stimuli from the categories self, other, good, and bad as quickly as they could by pressing two response keys. Achievement in math and Croatian language subjects were obtained in the form of average grades from official school records for all participating children.

Results showed that Grade 5 students were significantly lower than average on explicit self-esteem (p < .001) but did not differ on implicit self-esteem (p > .05). Additionally, girls were significantly lower than boys on explicit self-esteem (p = .034), but higher than boys on implicit self-esteem (p = .015). Finally, children’s explicit self-esteem was positively related to better math achievement only (p = .002), and their implicit self-esteem was positively related to better language achievement only (p = .043).

In every way in which explicit and implicit self-esteem were compared, the two types of self-esteem diverged. We hypothesize that early emerging implicit self-esteem may not only form the groundwork for the development of explicit self-esteem, but may also be linked to the internalization of certain gender stereotypes, the formation of social identities, and to school performance. We will discuss how the research tools, empirical findings, and hypotheses offered in this study suggest the value of studying the origins and consequences of children’s self-esteem for school achievement, beginning at younger ages than have traditionally been targeted.

Authors

• Dario Cvencek, University of Washington
  Presenting Author

• Elizabeth A. Sanders, University of Washington

• Ružica Brečić, University of Zagreb

• Dora Gacesa, Faculty of Economics and Business, Zagreb

• David Skala, University of Zagreb

• Andrew N Meltzoff, University of Washington