Advanced Analytics Workshop
Short course series: Fundamentals of linear mixed models for designed experiments
September 19-20, 2026
164 Justin Hall, Manhattan, Kansas
Instructors
- Josefina Lacasa, PhD, Assistant Professor, Department of Statistics, Kansas State University.
- Claudio Dias da Silva, Graduate Research Assistant, Department of Plant Pathology & Department of Statistics, Kansas State University.
Overview: Linear mixed models are widely used for analyzing data generated by designed experiments. However, figuring out how to model a given dataset requires a careful understanding of the data generating process and a basic intuition of how mixed models work. This workshop series aims to help practitioners gain understanding and develop the intuition for the most common assumptions in mixed models.
Target audience: K-State faculty, research scholars and graduate students interested in the applications of mixed models for modeling data generated by designed experiments.
Software and computer requisites: Since model applications will be demonstrated using R software, prior experience using R software will be convenient but not required. Likewise, attendees are encouraged to bring their laptops, but will be able to follow the content regardless.
Email questions to Dr. Josefina Lacasa at lacasa@k-state.edu.
Saturday, September 19 | 9:00 a.m. - 4:00 p.m.
Fundamentals of linear mixed models for designed experiments
- Introduction to the intuition behind mixed models.
- How to build a statistical model.
- Fitting a mixed model to experimental data.
- Model diagnostics.
Sunday, September 20 | 9:00 a.m. - 12:00 p.m.
Generalized linear mixed models
- Modeling data with non-normal responses.
- Model diagnostics for non-normal GLMMs.
- Gelman, A. and Hill, J. (2006). Data Analysis Using Regression and Multilevel/Hierarchical Models (1st ed.). Cambridge University Press. [link]
- Stroup, W.W., Ptukhina, M., & Garai, J. (2024). Generalized Linear Mixed Models: Modern Concepts, Methods and Applications (2nd ed.). Chapman and Hall/CRC. [link]