Meta-analysis is a formal method for summarizing evidence across multiple studies addressing the same scientific question. The growing popularity of meta-analysis is reflected by the number of the contributed R packages that are now available with tools for meta-analysis (meta, rmeta, metafor, ...). The goals of this tutorial are to 1) give an overview of the theory of meta-analysis and its applications, 2) give a comprehensive review of contributed packages for meta-analysis in the R language and 3) to work through a number of hands-on examples of analyses, plotting, and reporting that commonly arise as part of a meta-analysis, using real data sets. Several special topics will also be covered including study selection and retrieval from the PubMed database (my package RISmed), meta-analysis of high throughput data (GeneMeta), and estimating effect modification of an individual patient data meta-analysis with aggregate data (my package ipdmeta).
Meta-analysis is a systematic and concise method for consolidating current knowledge on a scientific topic. The rapid development cycle of the R open-source software and its active community of developers make it an ideal environment for keeping pace with the latest methodological developments in meta-analysis.
By the end of this tutorial, attendees should have a firm understanding of current meta- analytic tools available (and not available) in R and have gained experience with the major uses of each.
Attendees are expected to have a basic working knowledge of R. Attendees should bring a laptop with the latest version of R installed. A list of packages to install for the tutorial will be sent to registered attendees prior to the conference.
This tutorial will be designed for research scientists who are users of R, but all interested useRs are welcome.