This intermediate level course introduces the fundamental concepts of regression analysis. A unified approach to regression is taken, which emphasizes the common features of regression for quantitative and discrete response variables.
The course emphasizes the proper formulation and interpretation of regression models and uses several data examples from various health science fields to illustrate the concepts. Specific types of models illustrated by example will include linear regression, analysis of variance, logistic regression, relative risk regression, and Poisson regression.
The course also emphasizes the assumptions of regression analysis and the impact of violations of the assumptions on inference. As one example of this, robust variance estimates will be used to minimize assumptions about the response variance. A brief introduction to regression analysis of multi-level clustered data will also be given.
Assigned readings will be included in the course materials. The course will include an afternoon computer lab session where the students will gain experience in performing and interpreting various types of regression analysis using statistical software.
This course is equivalent to the Erasmus Summer Programme course Regression Analysis (ESP09) in August.