Multilevel models (also referred to as hierarchical models or mixed models) are a class of statistical models that can be used when observations are not independent. Non-independence can occur when data is clustered due to the study design (e.g., data collected on households and their individual members or blocked agricultural studies) or when data is collected on the same observational units repeatedly under different conditions or over time (e.g., repeated measures or longitudinal studies).
The purpose of this workshop is to introduce the basic concepts, underlying statistical models, and estimation techniques commonly used when data are not independent. Cluster robust standard errors, GEE models, and linear mixed effects models will be covered. Multiple examples will be presented during the workshop to enable participants to recognize when and how such models can be applied in their own research.
The workshop is intended for participants who have the equivalent of two semesters of statistics and previous experience with ANOVA and linear regression. The workshop will be taught through a combination of lecture and hands-on demonstration in R.
- Cluster robust standard errors
- Generalized Estimating Equations
- Linear Mixed Models