Is Bilingualism Associated with Executive Function in Early Childhood? A Propensity Score Analysis

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Attention, Learning, Memory

Abstract

In the past two decades, there has been great interest in whether bilingual individuals have superior executive function (EF) skills when compared to monolinguals. Overall, results of this body of research are mixed, with some researchers claiming to provide evidence of a bilingual advantage (e.g., Carlson & Meltzoff, 2008) and others arguing there is no consistent evidence (e.g., Paap et al., 2015). There are numerous potential explanations for the discrepancy in results across studies, as bilingual samples differ in many ways (e.g., language experience, age, demographics), different components of EF are measured across studies (and different tasks are used across studies to index the same component of EF), and there are differences in the comparability of monolingual and bilingual samples within and across studies. Differences across bilingual samples and EF constructs may represent true heterogeneity in estimates of the bilingual advantage, whereas differences in the comparability of monolingual and bilingual populations within and across studies may confound estimates of the bilingual advantage in the extant literature. To address potential confounds in the literature on the bilingual advantage, the purpose of this study was to use a comprehensive propensity-score matching approach to evaluate the potential for the bilingual advantage in EF among a large, nationally representative sample of Spanish-speaking kindergarteners in the U.S. This approach makes a positive contribution to the literature in two ways. First, it advances understanding of whether a bilingual advantage in working memory, inhibitory control, cognitive flexibility, and attention exists for Spanish-speaking kindergarteners in the U.S. Second, it provides a more precise, controlled estimate of the bilingual advantage in this context, as we use propensity-score matching to control for potential confounds across monolingual and bilingual children that could explain group differences in EF.
We matched bilingual and monolingual kindergarteners on a comprehensive set of child- and school-level covariates, and conducted sensitivity analysis to evaluate whether results were sensitive to observed and unobserved confounds. After matching groups, bilinguals had greater teacher-rated inhibitory control and attentional focus than did monolinguals; however, only the effect for inhibitory control was robust to sensitivity analysis. In contrast, no effects of bilingualism were observed for performance-based measures of working memory or cognitive flexibility. Although there was a small effect of bilingualism on teacher-rated inhibitory control, this study did not provide broad support for the bilingual advantage hypothesis. Our propensity score analysis revealed important differences between the matched and unmatched samples that limit generalizability to certain subgroups of Spanish-speaking children growing up in the U.S. Bilingual children who did not have a match had lower performance on measures of working memory and cognitive flexibility, came from schools with lower mathematics achievement, came from lower SES backgrounds, had less involvement in extracurricular activities, and experienced less book reading in the home, among other school-level factors that differed across the matched and unmatched samples. Therefore, any robust bilingual advantages in this study, which were limited to small effects on inhibitory control, are only generalizable to Spanish-speaking students in the U.S. who come from relatively advantaged backgrounds.

Authors

• John Marc Goodrich, University of Nebraska - Lincoln
  Presenting Author

• Natalie A. Koziol, University of Nebraska-Lincoln
  Non-Presenting Author

• HyeonJin Yoon, University of Nebraska-Lincoln
  Non-Presenting Author

• Sergio Leiva, University of Nebraska-Lincoln
  Non-Presenting Author