Mixed Models

“Mixed Models” address datasets containing multiple measurements of the same individuals or of groups, a situation in which classical statistical approaches are biased. In this course, we begin with a summary of linear models and their limitations, then explain “Mixed Models”, their applicability, and usage. We will cover random intercept and random slope models in detail. Besides discussions on the interpretation and theory also ideas how to run the models with R and exercises will be provided.

Topics:

This introductory course on Mixed Models covers:

• Limits of linear regression in case of repeated measurements

• Introduction of mixed models

• Random intercept models

• Random slope models

• Interpretation and application of mixed models using R

Methods:

The course consists of theoretical lessons on mixed models, how to apply and how to interpret mixed models. Theoretical lessons will be followed by hands-on examples with best-practice solutions in R.

1. Understand the Limitations of Linear Regression for Repeated Measures

• Revise the knowledge in linear regression.
• Explain why classical linear models may be biased when applied to repeated measurements within individuals or groups.

2. Comprehend the Basics of Mixed Models

• Describe the structure and purpose of mixed models, focusing on handling grouped or repeated data accurately.
• Understand how mixed models are working without focusing on formulas.

3. Apply Random Intercept and Random Slope Models

• Differentiate between random intercept and random slope models and determine when each is appropriate.
• Fit and interpret random intercept and random slope models using R.

4. Interpret and Report Results from Mixed Models

• Analyze and interpret outputs from mixed models in R for practical applications.
• Communicate findings and insights gained from mixed models clearly in a research context.

Programming skills with R, e.g. course Introduction to R and knowledge of regression models, e.g. course Introduction to Statistics.