Missing Data Analysis with Blimp

1 Introduction

This guide will teach you how to use Blimp for missing data analysis. Blimp is an easy-to-use Bayesian estimation software program that is freely available at https://www.appliedmissingdata.com/.

Blimp is also installed on the SSCC’s Linux servers.

1.1 Scope of this Book

This book has three purposes:

1. Introduce the basics of Bayesian estimation, particularly where it relates to missing data analysis.
   • This is the next chapter, Bayesian Basics.
   • You will typically need to take one or more graduate-level Bayesian-specific courses to thoughtfully formulate and make sense of a Bayesian analysis, in addition to several other courses in (frequentist) statistics. Consider the next chapter on Bayesian Basics as a review if you are familiar with Bayesian statistics, or as a starting point if you are new to it. Those seeking to learn more are encouraged to consult the books in the Resources section below.
2. Demonstrate Blimp’s capabilities through a series of examples.
   • This is the second part of the book, Modeling with Blimp.
   • The team behind Blimp has already made excellent tutorial videos and a user’s guide with examples available (see links in the Resources section below). Craig Enders is a master teacher, and you should make use of his and his team’s wonderful resources to learn more about Blimp. The purpose of this book in the context of these existing resources is to distill and demonstrate models that are particularly useful for researchers in the social sciences.
3. Discuss how to incorporate Blimp into an analysis workflow with data wrangling and parameter reporting.
   • This is the third part of the book, Workflow.
   • This book assumes you already have data wrangling skills in some other language (Stata, R, Python, etc.). If you would like to gain these skills, see our online guides and sign up for our training.

1.2 Resources

Blimp:

• Papers, presentations, and workshop materials
• Tutorial videos
• User’s guide

Books (all available online through the UW-Madison library or elsewhere):

• Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2015). Bayesian data analysis (3rd ed.). Chapman and Hall/CRC. https://doi.org/10.1201/b16018 (available at https://stat.columbia.edu/~gelman/book/BDA3.pdf)
• Jackman, S. (2009). Bayesian analysis for the social sciences. Wiley. https://doi.org/10.1002/9780470686621
• Levy, R., & Mislevy, R. J. (2016). Bayesian psychometric modeling. Chapman & Hall/CRC. https://doi.org/10.1201/9781315374604

Articles:

• Du, H., Enders, C., Keller, B. T., Bradbury, T. N., & Karney, B. R. (2022). A Bayesian latent variable selection model for nonignorable missingness. Multivariate Behavioral Research, 55(2-3), 478-512. https://doi.org/10.1080/00273171.2021.1874259
• Enders, C., (2023). Missing data: An update on the state of the art. Psychological Methods. https://doi.org/10.1037/met0000563
• Keller, B. T. (2022). An introduction to factored regression models with Blimp. Psych, 4, 10-37. https://doi.org/10.3390/psych4010002