Poster #31 - Domain-General Processes and Science Achievement in Multilingual and Monolingual Students in Early Elementary School

• In Event: Poster Session 06

Abstract

Job growth in science, technology, engineering, and math (STEM) is outpacing job growth in all other fields; however, linguistically diverse minorities are underrepresented in the U.S. STEM workforce (Pew Research Center, 2018). Differences in STEM achievement can be traced to kindergarten (e.g., Morgan et al., 2016), indicating that more attention on early STEM achievement is needed to increase representational equity. The empirical literature focused on STEM achievement among multilingual students in early elementary school in the U.S. is almost non-existent. Prior research (Barnard-Brak et al., 2017; Hwang, 2020) guided by domain-general theories of development (e.g., Vygotsky, 1962, Pascual-Leone & Johnson, 2021), suggests that science achievement is distinct from math and reading achievement during early elementary school. However, Barnard-Brak et al. (2017) did not focus on multilingual students specifically and the science measure used in Hwang (2020) was a measure of general knowledge that included science and social studies items. Research also suggests that language proficiency and executive functioning (EF) support science achievement (Koerber & Osterhaus, 2019; Mayer et al., 2014; van der Graff et al., 2016), though such studies were conducted internationally and limited to concurrent relations. To better understand science achievement, its associations with math and reading, and factors that support science achievement among multilingual children in the U.S., more research across early elementary school is needed.

The present study used the public use file of the Early Childhood Longitudinal Study, Kindergarten class of 2010–2011 (ECLS-K: 2011) and included multilingual (n=1,022) and English monolingual (n=12,343) children. Science achievement was measured annually in the spring of kindergarten, first, and second grade. Language proficiency and EF was measured during the spring of kindergarten. Consistent with Barnard-Brak et al. (2017) and Hwang (2020), science was separable from math and reading achievement for multilinguals and English monolinguals, though highly correlated (see Table 1). Multigroup path analyses and tests of measurement invariance indicated that kindergarten language proficiency and EF supported science achievement similarly across early elementary school for multilingual (R2 = 0.24-0.39) and English monolingual (R2 = 0.25-0.34; see Table 2) children, even when accounting for the effect of SES statistically. This pattern of results was cross validated using a subgroup of multilingual English language learners (ELLs) (n=524) and a subgroup of English monolinguals who were not ELLs (n=294); thereby strengthening the inferences that can be drawn from this study.

The results from the path analyses coupled with high correlations among science, math, and reading achievement suggest that science instruction that takes advantage of domain interconnectedness may reduce redundancies and help children connect with science content. Moreover, based on a clear body of research that indicates oral language is critical for learning, we posit that intentionally organizing instruction such that science, math, and reading instruction that takes advantage of children’s oral language may help reduce redundancies and increase instructional time dedicated to science content, thereby contributing to increased learning. At the very least, integrated instruction and cross-domain effects is a promising new area of research (Cabell & Hwang, 2020; Cervetti & Wright, 2020).

Authors

• Matthew E Foster, University of South Florida
  Presenting Author

• Sara A Smith, University of South Florida

• Trina D Spencer, University of South Florida