In recent years, discourse in psychology has been substantially affected by the 'replication crisis' - the realization in our field that many of its most prominent studies do not reliably replicate. Although large scale replication efforts and highly publicized failures to replicate have been the most notable ways of assessing and showcasing the reproducibility of psychological research, meta-analyses can be a useful option for evaluating the rates of replicability in a population of studies in the literature. However, since their inception, meta-analyses have also been the subject of censure. There is the issue of publication bias causing inflated effects, the problem of the inclusion of the results of under-powered studies and the overlooked complication of heterogeneity. In this talk, Dominic and Sammy show how meta-analyses can be improved, including the use of p-curve analyses and continuously cumulating meta-analysis. They then go on to discuss how meta-analyses can serve as diagnostic tools for assessing statistical power, selective reporting bias, and between-study heterogeneity and ultimately, how meta-analyses can themselves be used as way quantify the degree of replicability in psychological science.