This course gives a practical introduction to the use of multiple linear regression in the analysis of continuous outcomes.

In simple linear regression a continuous outcome (e.g. blood pressure, salary) is predicted using one variable by searching for the line that best fits the data. In multiple regression we extend this idea to the context where 2 or more predictors are used to predict the outcome. In case of a categorical predictor multiple regression is often referred to as ANOVA or ANCOVA.

The course starts with an online self-study through a starter's package, followed by two in-class lectures. Participants are expected to complete the starter's package by the first in-class lecture. The software package SPSS will be used in the exercises.

For this course, we offer the possibility to take an exam. For the PhD students in the faculties IOB and Applied Economics, this is a requirement to obtain a credits for these courses, but people from other faculties are allowed as well. If you are interested in taking the exam, check the wants-to-take-exam-box in the registration form. Participating in the exam costs 10€, which is deduced automatically from your educational credit.

Course contents

  • Starter's package (to be completed by first lecture)
    • SPSS demo 
    • Theoretical basis of the simple linear regression model
    • Exercise simple linear regression
  • Lecture 1
    • Q&A on starter's package + feedback on exercise
    • Dummy coding of categorical variables
    • 2-way ANOVA and ANCOVA
    • The multiple linear regression model
  • Lecture 2
    • Including higher order and interaction terms in a multiple linear regression model
    • Checking the model assumptions and troubleshooting if model assumptions are not met
    • Model building (forward, backward)
    • How to deal with problems like multicollinearity 

Target audience / prerequisites

Students should be familiar with the concepts and techniques that are taught in the course "Basic Principles of Statistics". In particular, simple linear regression and one-way ANOVA are assumed to be known.

Instructors

Ella Roelant and Erik Fransen

Prices

PhD student UAntwerpen : € 50 

UA-affiliated : € 90

Academic non-UA : € 160

Publicand non-profit sector : € 250

Private sector : € 500

Time and place

The next course will take place in June 2023.