For many empirical questions, it can seem that science has generated both too much and not enough evidence. Too much, because there are thousands of relevant published studies. Not enough, because single studies are rarely definitive, and different studies yield conflicting results. In such situations, a method called meta-analysis—the use of statistical techniques to combine and analyze data from multiple already-published studies—comes to the rescue. Meta-analysis provides an assessment of what conclusion some complex body of evidence supports. Philosophical discussion of meta-analysis so far has focused on one question: Is meta-analysis an objective way of combining evidence, or do the choices and judgments researchers make as they conduct meta-analyses compromise objectivity? In this talk, Kovaka argues that meta-analysis has another, under-explored epistemic function. It allows scientists to better understand the causes and consequences of variation in a body of evidence. This epistemic virtue is crucial in the life sciences, and the benefits to scientific understanding are largely independent of the outcome of the debate about the objectivity of meta-analysis.