ABSTRACT
Research synthesis is a set of related methods that integrate the findings of separate empirical studies. It is a tool for understanding a body of literature and characteristics that enhance or diminish relationships of interest. Meta-analysis is an optional component of research synthesis that determines a quantitative estimate of the average relationship and explores the extent to which it varies across studies.
Research synthesis and meta-analysis help to process and understand large bodies of research that may be difficult to assess by looking at individual studies separately. In addition, they clarify when primary studies seem to present conflicting findings from one study to the next or across theories. Whereas primary research can provide insight into conflicting findings by examining the variation within a single study, research synthesis utilizes variation between studies in order to highlight the conditions under which a relationship is enhanced or diminished. These conditions might include characteristics of the study, sample, setting, or measures.
Most research synthesists agree that standard research synthesis methods include a comprehensive literature search, a systematic review of the literature, and, when appropriate, additional methods for meta-analysis. Standard meta-analysis methods include the creation of an average effect size, an estimate of the variability in effect sizes, and, when appropriate, moderator analyses to assess the extent to which theoretical or methodological factors (e.g., related to the participants, settings, interventions, etc.) explain variability in study findings. As in any field, methods are constantly being developed to enhance the quality of research syntheses. Key issues that have plagued the field include the challenge of finding unpublished research and detecting publication bias, missing data, the nature of the error model, and dealing with dependent tests of a hypothesis. State-of-the-art solutions have emerged in recent years to address these issues.
