PD23-7: An Introduction to Bayesian Estimation and Missing Data Imputation for Educational Research

Session Type: Professional Development Course

Abstract

This introductory course introduces attendees to Bayesian estimation and missing data imputation. The course addresses a number of practical issues that researchers are likely to encounter in their research, including mixtures of categorical and continuous variables, interactive and nonlinear effects, and multilevel data structures. The primary goal of this course is to provide participants with the skills necessary to understand and apply cutting-edge missing data handling methods to their own data. Technical information will be presented in an accessible manner that is readily understandable by researchers who use, but do not specialize in, quantitative methods. Accordingly, the course will target practicing researchers (graduate students, professors, research professionals) who possess typical graduate-level statistics training, in particular, familiarity with multiple regression. The course will provide a mixture of lecture and computer applications. Presentation software will be used to deliver the course content, and attendees will be provided with extensive handouts. Estimation and imputation will be illustrated using the Blimp application, which is available as a free download at www.appliedmissingdata.com/blimp. Blimp was developed as part of an Institute of Education Sciences funded projects, and the program implements a number of cutting-edge approaches that are not available in other software. Course handouts will include computer code and scripts that attendees can use to analyze data.

Sub Unit

• Professional Development and Training Committee

Director

• Craig K. Enders, University of California - Los Angeles

Instructor

• Michael Woller, University of California - Los Angeles