Statistics II: Analysis of experiments using mixed-effects models

Content

The course will cover the following topics:

• block design, crossed design, hierarchical design and split-plot design
• analysis of variance
• linear mixed-effects models for normally distributed observations
• assumptions and transformation of data
• multiple comparisons.

Prior knowledge

Statistics I: Basic Statistics or equivalent

Objectives

The objective of the course is to give an overview of the basic principles behind design and analysis of replicated factorial experiments aiming at comparing experimental treatments. Examples are given within agricultural sciences and related fields of research. On completion of the course, the student will be able to

• describe basic principles in experimental design, such as replication, blocking and balance
• specify linear fixed- and mixed-effects models, including assumptions
• select an appropriate model for a given experimental design
• use the statistical software R for analysis of fixed- and mixed-effects models
• interpret and evaluate results correctly and draw reasonable conclusions
• clearly and concisely communicate results and conclusions.