Statistics Workshop: An Introduction to Bayesian Statistics for Psychological Research

Abstract

The use of Bayesian methods has grown quickly in recent years among researchers in the social, behavioral, and psychological sciences thanks to methodological and computational advances that have made their use feasible and practical. The Bayesian framework offers many advantages not available through conventional non-Bayesian approaches, among these include (1) the ability for researchers to incorporate their prior beliefs and expert knowledge into the analysis in a principled manner, (2) increased flexibility in terms of model specification, (3) usefulness in evaluating complex models with a small data set, and (4) straightforward interpretation of results and findings.

The goal of this workshop is to offer a gentle, intuitive, and practical introduction to the Bayesian framework. We do not assume previous exposure to Bayesian statistics; however, an understanding of basic statistical concepts is helpful. The workshop will be presented in two parts – Foundations and Applications. In Foundations, we will focus on developing the intuition underlying the Bayesian framework. In particular, we will discuss relevant concepts from probability theory, Bayes’ theorem (the main engine behind the Bayesian framework), and comparisons to frequentist approaches. In Applications, we will focus on the use of software to facilitate Bayesian analyses through hands-on examples with an emphasis on examples from psychology and interpretation of results. Although R will be the main tool used throughout the session, we do not assume prior knowledge of R and examples will be provided in multiple software packages (e.g., Mplus) to facilitate use across multiple software programs. In addition, the authors will develop a dedicated website that contains additional examples and resources to encourage workshop attendees to continue learning more about Bayesian statistics. Click here to access the workshop website.