Kies de Nederlandse taal
Course module: 800897-B-6
Methodology for Premasters DSS
Course info
Course module800897-B-6
Credits (ECTS)6
CategoryBA (Bachelor)
Course typeCourse
Language of instructionEnglish
Offered byTilburg University; Tilburg School of Humanities and Digital Sciences; TSH: Department Cognitive Science and AI; TSH: Department Cognitive Science and AI;
Is part of
PM Data Science and Society
Convenant TSH
dr. R.J.C.M. Starmans
Other course modules lecturer
Academic year2020
Starting block
SM 2
Course mode
RemarksCaution: this information is subject to change
Registration openfrom 20/01/2021 up to and including 20/08/2021

After successful completion of the course the students:

  • can describe and explain the main paradigms, concepts, techniques, and problems in Methodology.

  • can apply these concepts and techniques to Data Science as a research area. 

  • can assess and evaluate the relevance of these concepts and techniques for Data Science as a research area. 

  • can identify and explain the several stages in the research process. 

  • can identify and assess the main threads to validity and reliability, particularly in research designs and data sets.

  • can analyze and compute performance measures and interpret graphs, tables and diagrams associated with it.

  • can identify and explain the differences between statistical learning and machine learning from a methodological point of view.

  • can evaluate the ethical aspects related to research in general and responsible and explainable AI in particular.


The core topics will be:

  • Introduction to methodology for Data Science: methodology between epistemology and science. Three perspectives on Methodology Four paradigms for scientific research. Data Science as a research area.

  • The Research Process, cycle, or roadmap and its several stages.

  • Explorative Data Analysis and Visualization

  • Validity and Reliability in Research. Many faces of validity and reliability, including cross- validation

  • Performance assessment and the Confusion matrix, including Bayes’ Theorem to compute performance measures

  • The curse of Causality: confounding, moderation, and mediation, including Simpsons paradox and Robinsons paradox.

  • Research designs and related threads to validity and reliability (RCT,  cross-sectional study, case-control study, cohort study)

  • Statistical Inference / learning 1 Tactics for Generalisation (external validity issues)

  • Statistical Inference / learning 2 Causality (internal validity issues)

  • Machine learning or how do new ML-algorithms affect research methodology and practice? Including such issues as the train and test paradigm, cross-validation and bootstrap, bias variance trade-off and overfitting.  

  • Responsible and explainable AI, including Ethics in empirical research

  • The Crisis of Science, including p-values wars, false positive rates, reproducibility crisis, “The End of Theory”, some classical fallacies.


Required materials

  • The course will use substantial parts of Paul R. Cohen, Empirical Methods for Artificial Intelligence. (1995), MIT Press. 

  • Throughout the course we will introduce papers, book chapters, and other documents, which will be classified as optional, recommended, or mandatory reading. All documents with the label mandatory reading, all slides of the lecturers / videos will be part of the exams.

  • We will be using IBM-SPSS to do analyses and simulations, showing the relevance of the methodological concepts. Students should have this installed this package before the course starts.

  • All further details will be made available at Canvas.


The course will be evaluated with two written exams, a midterm (30%) and a final exam (70%). The student should obtain at least a combined score of 5.5 to pass the exam. There will be one combined retake. There will be mandatory assignments that are part of the grading as well.

Contact person
dr. M. Postma
Timetable information
Methodology for Premasters DSS
Required materials
Recommended materials
Written (70%)

Midterm (30%)

Final grade

Kies de Nederlandse taal