Introduction to Multiverse Analysis: Improving Transparency and Robustness in Psychological Research

Researchers routinely face many reasonable choices when preparing data and conducting statistical analyses. These analytic decisions can meaningfully influence scientific conclusions, contributing to concerns about reproducibility and researcher degrees of freedom. Multiverse analysis provides a systematic approach for examining how conclusions vary across a range of defensible analytic decisions. This training session will introduce the conceptual foundations of multiverse analysis and explain when and why it should be used in psychological research. Participants will learn how to identify researcher degrees of freedom, construct a principled multiverse, summarize and visualize results, and interpret robustness across alternative analytic decisions. Using practical examples and demonstrations in open source tools, attendees will learn how multiverse analyses can complement preregistration and other open science practices by providing a more complete picture of the robustness of research findings.

Accessibility

Auto-captioning will be enabled, and you will be able to access a transcript of the recording two to three weeks after the webinar.