In small sample contexts, Bayesian estimation is often suggested as a viable alternative to frequentist estimation, such as maximum likelihood estimation. Our systematic literature review is the first study aggregating information from numerous simulation studies to present an overview of the performance of Bayesian and frequentist estimation for structural equation models with small sample sizes. We conclude that with small samples, the use of Bayesian estimation with diffuse default priors can result in severely biased estimates – the levels of bias are often even higher than when frequentist methods are used. This bias can only be decreased by incorporating prior information. We therefore recommend against naively using Bayesian estimation when samples are small, and encourage researchers to make well considered decisions about all priors. For this purpose, we provide recommendations on how to construct thoughtful priors.

Smid, S.C., McNeish, D., Miočević, M., and van de Schoot, R. (2019). Bayesian versus Frequentist Estimation for Structural Equation Models in Small Sample Contexts: A Systematic Review. Structural Equation Modeling: A Multidisciplinary Journal. DOI: 10.1080/10705511.2019.1577140

To promote transparency and replicability, syntax files are available at the OSF project page

PhD Student

Sannes PhD focuses on the use of informative priors in latent growth models with small sample sizes. She is interested in how prior knowledge can be used to compensate for small sample sizes.

Dan McNeish
Quantitative Methodologist working with the Center for Developmental Science and the Quantitative Psychology program
Further description is yet to come. Visit Dan's website for more information.
Visit website
Assistant Professor

In her current projects, Milica is focusing on optimal methods for data synthesis from non-exchangeable studies, on the consequences of specifying inaccurate priors in mediation models, and on issues that arise in applications of Bayesian mediation analysis with informative prior distributions in small samples.