Meta-analysis in Second Language Research: Choices and Challenges

Meta-analysis is a statistical method that can appear complex and intimidating at first glance (Rosenthal & DiMatteo, 2001). But at its heart, a meta-analysis calculates the mean and variance of a set of numbers. The numbers are not individual scores, however, as researchers are accustomed to averaging, but instead are statistics reported across studies within a particular research domain, such as a set of study correlations, standardized mean differences between groups, or odds ratios. For at least 150 years, scientific researchers have engaged in the practice of averaging effects found across a set of studies or scientific observations; however, meta-analysis has developed relatively recently as a formalized statistical method for doing so (for information on the development of meta-analysis, see Borenstein, Hedges, Higgins, & Rothstein, 2009; Hunter & F. L. Schmidt, 2004; for early works on meta-analytic methods, see Cooper & Rosenthal, 1980; Glass, 1976; Hedges & Olkin, 1985; Rosenthal, 1978; F. L. Schmidt & Hunter, 1977).