Development of Mathematics and Working Memory in Early Grades: A Parallel Growth Analysis of ECLS-K:2011

• In Event: Poster Session 07
  In Poster Session: PS 07 Topic - Education, Schooling

Abstract

Recent research shows that early mathematics competencies are critical for future academic achievement (Jordan et al., 2009; Raghubar et al.,2010; Watts et al., 2014). The relationship between students’ mathematics achievement and working memory (WM) skills in early school years is well established in the literature (Bull et al., 2008; Bull &Scerif, 2001; Gathercole et al., 2004; Smedt et al., 2009). However, there is little research done to examine how mathematics development is related to WM development in early school years. Understanding not only how these two important cognitive domains are related at a given time point, but also how the development of these domains is related can provide more nuanced pictures of early academic development for young children which can help educators deliver more effective instructions to young learners. To fill the research gap, this study examined the mathematics and WM development patterns of students using data collected from the Early Childhood Longitudinal Study Kindergarten Class of 2010-2011 (ECLS-K:2011) across the nine measurement occasions from kindergarten to fifth grade.
Direct cognitive assessments of mathematics and WM were administered to a nationally representative sample of kindergarten students that was followed until the fifth grade. Both assessment scores are vertically scaled, allowing for the longitudinal analysis. To examine the relationship between the development of mathematics and WM, first, a series of latent growth curve models were estimated and compared for the best fitting model separately for mathematics and WM. For both domains, a Piecewise Latent Growth Curve Model with three segments of growth (from kindergarten to first grade, from second to third grade, and from third to fifth grade) and with two summer effects modeled (between kindergarten and first grade and between first and second grade) fit the data the best. Then, these two best fitting Piecewise Latent Growth Curve models for each domain were estimated simultaneously to examine the relationships between the growth parameters from mathematics and WM growth analysis.
Findings were consistent with the literature suggesting that students demonstrated non-linear growth, with much faster growth during the first two years of formal schooling in both mathematics and WM development compared to the later years in elementary school years. Furthermore, students’ WM skills in kindergarten were positively related to mathematics achievement in kindergarten as well as mathematics growth from kindergarten to third grade. However, students' kindergarten mathematics skills were negatively associated with WM development, suggesting that WM is a prerequisite for mathematics development but not vice-versa. This study also found a strong association between WM skills and mathematics achievement in early grades but weakening association as students get older (Best et al., 2011; Friso-van den Bos, 2013; Holmes & Adams, 2006). Findings from this study highlight the critical role of WM skills in developing early mathematics competencies especially in the first two years of formal schooling. Given these findings, supporting development in WM prior to kindergarten potentially can help students, especially students who are at-risk for not developing these prerequisite skills sufficiently by the time they begin kindergarten.

Authors

• Ayse Cobanoglu, Yale University
  Presenting Author

• Bitnara Jasmine Park, American Institutes for Research -District of Columbia
  Non-Presenting Author

• Mengyi Li, American Institutes for Research
  Non-Presenting Author

• Yuan Zhang, American Institutes for Research -Virginia
  Non-Presenting Author

• Xiaying Zheng, American Institutes for Research
  Non-Presenting Author