12 Apr module 11: basic regression analysis
- module 11: basic regression analysis
14/09/2020 - 18/09/2020
1:45 am - 3:15 pm
Mon 14/09: 1:45 pm - 5:00 pm | online
Tue 15/09: 1:45 pm - 5:00 pm | online
Wed 16/09: 1:45 pm - 5:00 pm | online
Thu 17/09: 1:45 pm - 5:00 pm | online
Fri 18/09: 1:45 pm - 3:15 pm | online
week 2 | register now
Linear regression is used to predict how a continuous dependent variable (e.g., anxiety levels, temperature, attitudes) is affected by one or more predictors of any type. It is a basic statistical technique, used in many areas of research. Most of us will already be familiar with the concept of drawing a line through a cloud of data points. In this course we will start from the very beginning, including preparing your dataset for analysis and performing an exploratory data analysis. Next, we will recapitulate the basic concepts of linear regression, slowly expanding the model, from simple models with single predictors to models with multiple predictors and interaction effects. During the entire course, formulas and technicalities will be kept to the minimum. The focus of the course will be on interpretation and practical use of these models. All issues discussed will be illustrated in the five hands-on sessions of this course. Hands-on practical sessions will be organized separately for uses of different software packages (R or SPSS, to be chosen by the participant).
At the end of this module, participants should be able to:
- prepare and explore data for data analysis
- perform and interpret simple and multiple linear regression, including interaction effects
- assess the assumptions of the model
- select the most appropriate model for the data
- understand specific regression models and relations with group comparison techniques like ANOVA, t-tests and ANCOVA
- report on the different analysis steps and study results in a clear and correct way
The module is taught in 9 sessions of 1.5 hours, 5 of which are foreseen as hands-on computer labs (parallel sessions per software package). You will need a laptop with R (or Rstudio) installed on it.
Any good book on basic linear regression will suffice. We recommend for example the early chapters of Kutner, Nachtsheim, Neter and Li (2005) Applied Linear Statistical Models. For SPSS-users, the regression chapter of Field (2009) Discovering Statistics using SPSS (3rd ed.) is a good way to start.
Limited knowledge of basic statistics (meaning: you have had a course in statistics before (for example in your bachelor or master), but you do not know all details anymore). Basic knowledge of the software package of your choice (R or SPSS): importing data and basic data manipulation. Especially for the package R a basic knowledge is necessary!
Dr. Jef Vlegels is a sociologist and statistician working at Ghent University as a post-doc (CHEGG research group) and as a lecturer at Flames training network for Methodology and Statistics. In his research he specializes in quantitative methods for the social sciences, with a specific focus on Social Network Analysis. He is an experienced lecturer in methodology and statistics, with previous experience in introductory courses as well as advanced specialized courses for Bachelor, Master and PhD students.
Cristina Cametti is the Flames coordinator at KU Leuven. She holds a Master degree in Statistics (KU Leuven) and in Crime and Security Sciences (Università Cattolica del Sacro Cuore – Milan). Her interests focused mainly on teaching statistical software courses (R, Python, SAS), on multivariate data analysis and data cleaning techniques.