ABSTRACT
The purpose of this chapter is to provide an introductory tutorial to the process of meta-analysis. Like in primary research, in meta-analysis the research question drives the data collection as well as the statistical analysis and interpretation. The process of integrating findings from several studies of the same phenomena is known as research synthesis. The chapter focuses on the most common effect size statistic in psychotherapy outcome research, the standardized mean difference. Effect sizes from the same study will typically be correlated with one another because they are based on the same participants and treatments. Treating these correlated effect sizes as the unit of analysis violates the assumption of independence of observations common to most statistical procedures, including meta-analytic statistics. Perhaps the most important information coded in a meta-analysis is the main outcome finding. Sensitivity analysis is concerned with exploring whether decisions made as part of the meta-analysis affect the results substantially.
