Linear regression addresses how a continuous dependent variable is associated by one or more predictors of any type. The fact that many practical problems deal with continuous outcomes (e.g. income, blood pressure, temperature, affect) makes linear regression a popular tool, and most of us will be familiar with the concept of drawing a line through a cloud of data points.
The first two sessions of this module introduce the conceptual framework of this method using the simple case of a single predictor. Formulas and technicalities are kept to a minimum and the main focus is on interpretation of results and assessing model validity. This includes confidence statements on the predictor effect (hypothesis tests and confidence intervals), using the regression model to predict future results and verification of model assumptions.
In session 3 and 4 we allow for more than one predictor leading to the multiple linear regression model. We focus on either explanation or prediction. How to come to a parsimonious model starting from a large number of predictors will be discussed in detail. In these complex linear models special attention will be given to interpreting individual predictor effects, as they critically depend on other terms in the model and underlying relations between predictors (confounding).
In the last session a more elaborate data analysis is discussed. We touch on problems where linear regression is not appropriate and replaced by related approaches such as generalized linear models and mixed models.
Different features will be illustrated with case examples from the instructors practical experience, and participants are encouraged to bring examples from their own work.
Hands-on exercises are worked out behind the PC using the R software.
Schedule:
5 Thursday evenings in February and March 2025: February 6, 13, 20, 27 March 13, 2025 from 5.30 pm to 9.30 pm. Each lecture is followed by a hands-on practical session.