Assessing Emergence of Ethnic-Racial Socialization: Longitudinal Measurement Invariance of African-American and Latino Children

• In Event: 3-066 - “What’s Measurement Got to Do with It?” Testing Ethnic-Racial Equivalence of Identity-Related Measures

Integrative Statement

Ethnic-racial socialization (ERS) in the family context is posited to directly influence the full range of a child’s developmental competencies (García Coll et al., 1996; Hughes et al., 2006); bolstering (and perhaps protecting) young ethnic/racial minority children’s development through practices that respond to the daily experiences of being a minority in a racially and socially stratified society. Identifying the manifestations and qualities of ERS practices can inform our understanding of the proximal contexts that link directly and indirectly to the development of competencies for children of color. However, scarce empirical work has been conducted on when these practices may initially manifest for families, and whether well-documented ERS practices of adolescence extend “back” to earlier periods of development, particularly early childhood.

This study addresses gaps in the ERS literature by investigating if and when children’s experience of ERS emerge during early childhood. We focus on the following main aims: First, we examine the extent to which different ERS strategies are reported across the early childhood period. Next, we determine if the prominent constructs of ERS can be (a) validly and (b) invariantly measured throughout the early childhood period. Finally, we explore whether latent level differences existed across early childhood when comparing groups by child gender and ethnicity.

This study draws upon data from the [Blinded] Project, a 4-wave longitudinal investigation of self-regulation, parent-child interactions, cultural home environs, and school readiness. The sample (Table 1) consists of 407 children (186 Latino, 170 African American, and 51 Multi-racial children); approximately 53% of the sample are male (n = 219). Parents were interviewed at home when the focal child was 30 months (W1) and 42 months (W2) of age and then once each in kindergarten (W3) and first grade (W4). Analyses used the cultural socialization, preparation for bias, and promotion of mistrust subscales of the Multidimensional Scale of Race Socialization measure (Hughes, 2003; Hughes & Chen, 1997) at W1-W4.

Using longitudinal, multi-group confirmatory factor analyses in a structural equation model framework, we tested (a) the empirical validation of an a priori common factor model, (b) the initial inquiries into the longitudinal and between-group measurement invariance, and (c) population heterogeneity of the sample. A 2-factor model was validated in this early childhood sample, and measured invariantly across all waves of the study (Table 2). Cultural socialization was consistently reported with a notable increase upon school entry, preparation for bias became apparent upon school entry, and promotion of mistrust was rarely reported. Longitudinal, multi-group invariance was generally supported between parents of boys and girls and African American and Latino families. Caregivers reported similar frequencies of ERS with boys and girls, however African American caregivers reported greater frequencies of ERS than Latino caregivers.

Findings will be discussed in light of the conceptual, substantive, and analytic insights this work provides the ERS literature and the themes of measurement across the symposium; including key developmental considerations when measuring ERS and how the field can work to link empirical inquiries of ERS across prominent periods of child and adolescent development.

Authors

• Mariah M Contreras, Tufts University
  Presenting Author

• Margaret Caughy, University of Georgia
  Non-Presenting Author

• Margaret Tresch Owen, University of Texas at Dallas
  Non-Presenting Author