9. BCN Advanced Statistics

Target Group

BCN PhD students who have completed the BCN Statistics Course or have a good understanding of basic statistical techniques (linear regression, ANOVA, t-test, etc.)
Credits
2 EC

Period
Start: September 15, 2026

Form
Advanced Statistics is offered as a regular course to students of the Research Masters of the Faculty of Behavioral and Social Sciences. BCN PhD students do not complete the full course, but rather select three weeks. During these three weeks, BCN PhD students attend lectures and complete the associated weekly assignments. There is no graded assessment.

We recommend one of two selections:
•         Weeks 2, 3, and 6 focus on Bayesian statistics.
•         Weeks 2, 4, and 5 focus on regression analyses.

More information can be found on Ocasys under Advanced Statistics (GMMSGE23).

Lectures
Don van Ravenzwaaij and Anja Ernst

Content
This course deals with a variety of statistical models. The underlying conceptual framework as well as their application will be discussed. Starting from basic techniques such as the t-test and linear regression, we study statistical inference in detail. Several techniques for estimation and hypothesis testing and for model comparison and validation will be discussed. The course covers a wide range of linear models, as well as the versatile class of generalized linear models. Special emphasis will be placed on Bayesian inference.

An important part of the course concerns learning the statistical software package R. Advanced familiarity with R, to the extent that the student can perform all statistical techniques that are either assumed known (e.g., ANOVA) or treated in this course, using R is a learning objective. As R will also be used in follow-up statistical courses in this degree programme, proficiency in R is not only important for passing this course, but for successful completion of all quantitative parts of the curriculum.

Learning outcomes
After the course, the students:
Understand repeated measures models and multivariate models.
Are familiar with various approaches to estimation and testing, including least squares, maximum likelihood and Bayesian inference.
Determine which statistical model is most appropriate for a given empirical question.
Are able to perform all statistical analyses learned in this course using the software package R.
Assess whether the required model assumptions are met for the data at hand.
Interpret the results derived from applying a statistical model to empirical data.
Reflect upon different philosophical positions concerning retrieving evidence from data by contrasting frequentist and Bayesian approaches for hypothesis testing and estimation.

Time Schedule
See Ocasys under Advanced Statistics (GMMSGE23).

Number of participants
Max. 5

Location
See Ocasys under Advanced Statistics (GMMSGE23).

Application and information
Send an email to: d.h.koopmans umcg.nl by indication if you choose Bayesian statistics or regression analyses.