Doing Meta-Analysis in SLA: Practices, Choices, and Standards

Abstract

Meta-analysis is a statistical method used to synthesize the cumulative effect of an interventional treatment or a certain relationship that has been subjected to empirical investigation. Notwithstanding the increasing popularity of meta-analysis in second language acquisition (SLA) research, there is considerable methodological inconsistency in the 22 SLA meta-analyses published to date. Also, given the highly technical nature of some aspects of this statistical procedure and the specialized knowledge associated with it, it is a daunting task to conduct a meta-analysis and understand or interpret meta-analytic results. This paper seeks to solve the conundrum surrounding meta-analysis by explicating its basic procedure and technical aspects in an accessible manner. Utilizing SLA meta-analyses as examples, it discusses the methodological options of each stage of a meta-analysis, makes recommendations about how to handle some key issues, and points out the pitfalls that might undermine the robustness and credibility of meta-analytic findings.