A Quantitative Synthesis of Early Language Acquisition Using Meta-Analysis

Abstract

To acquire a language, children must learn a range of skills, from the sounds of their language to the meanings of words. These skills are typically studied in isolation in separate research programs, but there is a growing body of evidence that these skills may depend on each other in acquisition (e.g., Feldman, Myers, White, Griffiths, & Morgan, 2013; Johnson, Demuth, Jones, & Black, 2010; Shukla, White, & Aslin, 2011). We suggest that the meta-analytic method can support the process of building theories that take a systems-level perspective, as well as provide a tool for detecting bias in a literature. Here we present meta-analyses of 12 phenomena in language acquisition, with over 800 effect sizes. We find that the language acquisition literature overall has a high degree of evidential value. We then present a quantitative synthesis of language acquisition phenomena that suggests interactivity across the system.