Meta-analyses aim to reach more precise conclusions than single studies by synthesizing effect sizes from multiple studies. Especially in research areas where sample sizes of single studies are rather small and sample estimates are therefore imprecise, meta-analyses can provide more reliable and accurate results. In addition, differences in effect sizes between studies can be used to examine the influence of study or sample characteristics on the effect size of interest in moderator analyses.
However, there are a number of possible degrees of freedom in conducting meta-analyses that are relevant to the interpretation of the results. In the spirit of Open Science, all relevant decisions should be documented transparently whenever possible, for example, using common publication standards for meta-analyses such as MARS or PRISMA. In addition, meta-analytic data should be as openly available and easily accessible as possible, thus enabling full replication of the meta-analysis. In particular, for regular updates of meta-analyses, it is desirable and efficient for the research community at large if existing data can be updated as straightforwardly as possible.
One possible approach for publishing meta-analytic data is the idea of Community-Augmented Meta-Analysis (CAMA). Here, the research community is provided with a repository to collect meta-analytic data cumulatively and collaboratively. At the same time, key meta-analytic outputs such as forest plots or meta-analytic estimates are made available on a graphical interface, making the results easily accessible. The functionality of such a system is presented using the platform PsychOpenCAMA, which provides such a system for psychology and related fields.
Speaker: Tanja Burgard is acting head of the research area Research Syntheses at ZPID and product responsible for PsychOpen CAMA.