An examination of the influence of early childhood care on academic achievement using quantile regression

Purpose: Early childhood education and care (ECEC) receives widespread support in the public domain; however, previous studies of ECEC have revealed mixed results. In some cases ECEC attendance is positively associated with... [ view full abstract ]

Purpose: Early childhood education and care (ECEC) receives widespread support in the public domain; however, previous studies of ECEC have revealed mixed results. In some cases ECEC attendance is positively associated with academic outcomes (e.g. Fox & Geddes, 2016), and in others ECEC attendance appears ineffective or negative (e.g. Coley et al., 2015) in terms of later academic achievement. The mixed evidence surrounding ECEC suggests the need for further investigation into its influence on educational achievement. The present study addressed this need by examining whether ECEC attendance influenced later school achievement not just at the means, but at multiple points along the distributions of achievement in a large sample of Australian twins spanning 3^rd through 9^th grade. Furthermore, this study examined whether ECEC attendance moderated the heritability and/or shared environmental influences of school achievement across the distributions of achievement. Methods: Participants were obtained from the Longitudinal Twin Study of NAPLAN (n=1517; 667 MZ, 850 DZ). Early childhood education and care attendance was measured as hours of attendance per week. Quantile regression was used to examine the influence of ECEC attendance across the distribution of reading in 3^rd, 5^th, 7^th, and 9^th grades. Next, a genetically-sensitive extension of quantile regression (see Logan et al., 2012) was conducted to examine ECEC attendance as a moderating influence on the heritability and shared-environmental influences of school achievement across the distributions of achievement. Results & Discussion: Results indicated a slightly negative, though mostly non-significant influence of ECEC attendance across the distributions of reading for all grade levels after controlling for ECEC type (e.g. family-day care, formal ECEC setting), father’s education level, and total number of months attending ECEC. Additionally, results indicated that heritability estimates were higher, and shared environmental estimates were lower, with ECEC attendance included as a moderator in 3^rd grade, with these trends disappearing by 5^th grade. These results suggest that, although ECEC attendance does account for a small portion of the shared environmental influence on reading in 3^rd grade, these influences appear to have no significant short-term or long-term sway on reading for Australian students. Additional results for numeracy and writing in 3^rd, 5^th, 7^th, and 9^th grades will be presented and discussed. 

References:

Coley, R. L., Lombardi, C. M., & Sims, J. (2015). Long-term implications of early education and care programs for Australian children. Journal of Educational Psychology, 107(1), 284.

Fox, S., & Geddes, M. (2016). Preschool-Two Years are Better Than One: Developing a Preschool Program for Australian 3 Year Olds–Evidence, Policy and Implementation. Logan, J. A., Petrill, S. A., Hart, S. A., Schatschneider, C., Thompson, L. A., Deater-Deckard, K., ... & Bartlett, C. (2012). Heritability across the distribution: An application of quantile regression. Behavior Genetics, 42(2), 256-267.