Utrecht, Netherlands

Regression in R

when 2 February 2022 - 2 February 2022
language English
credits 0.5 EC
fee EUR 150

Linear regression is one of the most ubiquitous statistical methods. Most statistical techniques can be viewed as either special cases of linear regression (e.g., t-tests, ANOVA) or generalisations of linear regression (e.g., multilevel modelling, SEM, neural networks, GLM, survival analysis). In this course, students will learn how to apply linear regression techniques in R. We will cover (multiple) linear regression, categorical predictor variables, moderation, prediction, and diagnostics. Students will practice what they learn via practical exercises.

In the morning/early afternoon, new content will be presented via interactive lectures. In the afternoon, the students will practice what they learned via practical exercises. If the schedule permits, the students are also welcome to ask the instructor for advice on their own data analyses.

We will not cover basic R usage. Students should already know how to use R to read and write data, do basic data manipulations, run R functions, and work with the results returned by R functions.
Although we will briefly discuss the theory of the methods covered, we will primarily focus on applying these methods in R. So, students should already have some familiarity with the theory of linear regression.
We will not cover any generalisations of linear regression such as logistic regression, multilevel modelling, or SEM
.Participants should bring their own computer with both R and RStudio installed.

Course leader

Kyle Lang

Target group

Professionals seeking a master-level introduction to linear regression

For an overview of all our summer school courses offered by the Department of Methodology and Statistics please click here.

Course aim

After completing this course, students can:

Describe how to apply linear regression in R and choose the correct functions with which to implement a given analysis.
Write basic R scripts to do the following:
Run a multiple linear regression model
Manipulate the fitted model object produced when estimated a linear regression model
Incorporate categorical predictor variables into linear regression models using an appropriate coding scheme
Test for moderation using linear regression and conduct a simple slopes analysis
Generate predictions by applying a fitted regression model to new data
Calculate measures of prediction error to compare the performance of different models
Check the assumptions of the linear regression model via model diagnostics

Fee info

EUR 150: Course + course materials + lunch

Register for this course
on course website