About this srcd poster session

---

Panel information
Panel 17. Race, Ethnicity, Culture, Context

Abstract

---

Latino/a/x late adolescents' sociocultural and family contexts can provide opportunities for accumulating funds of knowledge about higher education through interactions with family members, teachers, and friends (Hogg, 2011). Funds of knowledge are multi-faceted areas of knowledge (e.g., social, cultural, education) and are an important resource for wellbeing, skills, and experiences (Gonzalez et al., 1995; Moll et al., 1992). Considering contextual promotive resources (e.g., family; García Coll, et al., 1996) for funds of knowledge has rarely been examined during late adolescence in Latino/a/x populations, where there is a strong cultural emphasis on interdependence and cohesive family relationships (Knight et al., 2011). In this study, we aimed to establish the validity of a funds of knowledge measure drawing on ratings from Latino/a/x late adolescents. Framed within cultural-ecological perspectives (García Coll, et al., 1996), we aimed to establish the factorial validity and describe dimensions of Latino/a/x late adolescents' funds of knowledge.

To address the first study goal we conducted Bayesian confirmatory factor analysis (CFA) to examine the measurement model of funds of knowledge. We expected that the 3-factor measurement model would reveal support for dimensions of family, school, and friends. To address our second goal of describing the frequencies of funds of knowledge dimensions, we conducted one-way multivariate analyses of variance (MANOVA) with dimension as a within-individual (repeated) factor. Participants were part of a larger longitudinal study of Latino/a/x late adolescents recruited from the Southwestern U.S.(Mage = 18.10, SD = 0.41; 64.4% female; 84.7% Mexican). The Bayesian CFA was conducted in RStudio (R 4.3.2), and missing data were accounted for using Bayesian estimation. The MANOVA were conducted in SAS 9.4. Model specification and item details are reported in Tables 1 and 2.

With regards to the measurement model, the numerical checks, PSRF < 1.01, and visual inspection of the MCMC diagnostic results (e.g., trace plots, autocorrelation, running mean) supported evidence of convergence for all model parameters. Model data fit provided tentative support for the hypothesized 3-factor structure as the basis for a model, with potential misfit suggested by the χ2, SRMR, and RMSEA (Table 1), and local fit (e.g., covariances). All standardized factor loadings were above .40, and none of the HPD intervals included zero (Table 2). The results of the CFA provide initial evidence of the 3-factor structure representing the dimensions of family, school, and friends. Turning to the second study goal, the MANOVA was significant, F(2, 202)=30.05, p<.001, and the follow-up contrasts revealed all three dimensions differed from one another, adolescents endorsed funds of knowledge from friends more strongly than family, F(1, 203)= 5.86, p<.05, and school, F(1, 203)= 58.05, p<.0001, and family more strongly than school, F(1, 203)= 25.62, p<.0001. This study advances our understanding of the dimensions of funds of knowledge and the extent to which Latino/a/x late adolescents endorse different sources of funds of knowledge. Discussion will address the utility of funds of knowledge for developmental research, in addition to future directions for investigating the psychometric properties of the measure.

Author information

Author	Role
Karina Cahill, Arizona State University	Presenting author
Leah Doane, Arizona State University	Non-presenting author
Roy Levy, Arizona State University	Non-presenting author
Nancy Gonzales, Arizona State University	Non-presenting author

⇦ Back to session