Poster #110 - Children’s Executive Functions and Academic Skills by English Use at Home

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Language, Communication

Integrative Statement

Executive functions (EF) play a key role in fostering children’s learning of literacy and math, and growing evidence suggests that these skills in turn enhance the development of EF (Welsh et al., 2010). Dual language learning (DLL) children generally exhibit EF advantages early in their development relative to monolingual children, an advantage that is believed to stem from DLL children having to manage and attend to multiple languages (Bialystok et al., 2011). Many DLL children (primarily Spanish-speakers) lag behind English monolinguals academically due to limited English proficiency upon entering school (Mancilla-Martinez & Lesaux, 2011). The EF advantages that DLLs exhibit early in their development represent a potential avenue to reduce the academic disparities among DLLs and English monolinguals. In this study, we examined the bidirectional associations among EF and children’s reading and math skills from kindergarten to 2nd grade, with the expectation that those associations would be stronger for DLL children than for English monolinguals.

Participants were 15,867 children (Mage = 67.50 months; 51.4% boys; 19% DLL) from the Early Childhood Longitudinal Study-Kindergarten (ECLS-K). Data were gathered in the fall and spring of kindergarten and 2nd grade. Measures of EF included the Dimensional Change Card Sorting task (Zelazo, 2006) and Numbers Reversed task in the Woodcock-Johnson III (Mather & Woodcock, 2001). Children’s reading skills were assessed via the letter recognition task in the Woodcock-Johnson III (Mather & Woodcock, 2001) and other literacy-related abilities such as interpreting questions. Math skills were assessed via a unique ECLS-K test of overall arithmetic knowledge.

Path analysis was used to test the bidirectional associations controlling for child gender and ethnicity. Separate models were tested by outcome of interest (reading and math skills). Multigroup model results (DLLs vs. non-DLLs) are reported because the multigroup models were significantly different from the models with the entire sample. As shown in Figures 1 and 2, the multi-group models fit the data well. Both reading model and mathematics model were significantly different by home language. In the model of reading skills (Figure 1), EF predicted reading for both groups from kindergarten to 2nd grade as well as concurrently within kindergarten and 2nd grade. However, bidirectionality of EF and reading was found only in DLL group, suggesting distinct associations by home language use. In the model of mathematics skills (Figure 2), EF predicted mathematics for both groups only within a school year, but not from kindergarten to 2nd grade. Mathematics at kindergarten predicted EF at 2nd grade only for DLL group. For the DLLs only, mediation analyses revealed that EF in 2nd grade mediated the relation between reading skills in kindergarten and reading skills in second grade (β = .10, S.E. = .04, p < .01), as well as the relation between math skills in kindergarten and math skills in 2nd grade (β = .15, S.E. = .06, p < .05). The discussion will focus on the similarities and differences found between DLLs and non-DLLs.

Authors

• Seunghee Han, University of Missouri
  Presenting Author

• Francisco Palermo, University of Missouri
  Non-Presenting Author