Module 7: Experimental Design

Dates - Venue - Description - Schedule - Target audience - Exam - IMPORTANT: Incorporation in DTP and reimbursement by DS
Course prerequisites - Teacher - Course material - Fees - Enrol


6 Tuesday evenings in April and May 2019: April 23 and 30, May 7, 14, 21 and 28, 2019, from 5.30 pm to 9.30 pm.
Please note: the deadline for UGent PhD students who want a refund to open a dossier on the DS website (Application for Recognition) is March 21, 2019.


Faculty of Science, Site Sterre, Krijgslaan 281, building S9, Ghent, pc room 1.1 - Alan Turing, 1st floor (except last class).


Study design is arguably one of the most important aspects of any empirical research project. Its impact on the correct statistical analysis to follow and on the conclusions that can be drawn should not be underestimated. Timely input of statistical knowhow and experience can do wonders at this stage. We illustrate this fact through several examples in this course and give participants the basic tools to set-up cost-efficient study designs.

The course content is close related to the theory and practice of linear statistical model (such as regression analysis and analysis of variance). Although the design phase of a study appears prior to the experimentation phase, data collection phase and statistical analysis phase, a design is constructed is constructed in function of the data analysis that will follow. The design is not only evidently related to the particular research question, but it is also entangled with intended data analysis method. A good knowledge of the theory of linear statistical models is therefore of utmost importance.

The key role of experimental design in scientific and operational research is evident. First of all, a good design must allow for correct interpretation and relevant results of the statistical analysis following the experiment. For example, unbiased estimation of interesting effect sizes and their standard errors is of interest. Second, efficiency in terms of cost versus precision may be considerably increased by choosing an appropriate design, an appropriate sampling scheme. Questions that may arise from this point of view are the following. How many experimental units do I need to sample? What levels of a predictor of interest do I need to consider?

The aim of this course is not only to teach important concepts related to designing studies, but also more generally to broaden the understanding of the relation between experimentation and statistical inference. In particular, the following topics will be addressed:

  • General concepts: sampling, randomization, blocking and stratification, (selection) bias, confounding, orthogonality.
  • Sample size calculation: exact and asymptotic methods, approximation methods using simulation.
  • Optimal designs: methods based on the Fisher information matrix (for example G, D, A optimal designs), designs for parameter estimation versus prediction.

Next to theory sessions, hands-on exercises will be worked out behind the PC using the statistical software package R.

Target audience

This course targets professionals and researchers from all areas who are involved in designing experiments and/or are interested in a better understanding of how the design and the sampling of an experiment is related to the data analysis.


Participants can, if they wish, take part in an exam. Upon succeeding in this test a certificate from Ghent University will be issued.
Please note: For UGent PhD students it is no longer necessary to succeed in this exam to be able to incorporate the course in the DTP.

Incorporation in DTP and reimbursement from DS for UGent PhD students

As a UGent PhD student, to be able to incorporate this course in your Doctoral Training Program (DTP) and get a reimbursement of the registration fee from your Doctoral School (DS) you need to follow strict rules: please take the necessary action in time. The deadline to open a dossier on the DS website (Application for Recognition) for this course is March 21, 2019. Please note that opening a dossier does not mean that you are enrolled. You still need to enrol via the registration form on this site.

Course prerequisites

Participants are expected to have an active knowledge of

  • R (although a lot of exemplary code will be provided),
  • basic statistics (probability, hypothesis testing,…),
  • linear regression and ANOVA,
  • basic matrix algebra.


Dr. Karel Vermeulen obtained his Master in Applied Mathematics, Master in Statistical Data Analysis and his PhD in Statistical Data Analysis at Ghent University and is currently doctor-assistant at the Department of Data Analysis and Mathematical Modeling at the Faculty of Bioscience Engineering (Ghent University). He teaches probability theory and statistics courses to Bachelor students Industrial Engineer in Bio-Sciences and to Bachelor students in Bio-Science Engineering but also the course Experimental Design to Master students in Bio-Science engineering and to students in the advanced Master in Statistical Data Analysis. In the past he was also involved in many other teaching duties for both applied and methodological courses.

Course material

  • Syllabus.
  • Recommended handbooks (optional, for your own interest):
    • Book 1: “Optimal design of experiments: a case study approach”, P. Goos and J. Bradley, (2011), John Wiley & Sons, ISBN 978-0470744611.
    • Book 2: “Design and analysis of experiments”, D.C. Montgomery, (2008), John Wiley & Sons, ISBN 978-8126540501.


Different prices apply, depending on your main type of employment.

EmploymentModule 7Book 1Book 2Exam
Industry/Private sector1 900 75 60 30
Non-profit, government, university outside AUGent2 405 75 60 30
(Doctoral) student outside AUGent2 315 75 60 30

1 If three or more employees from the same company enrol simultaneously for this course a reduction of 10% on the module price is taken into account.

2 AUGent staff and AUGent doctoral students who pay through use of an SAP internal order/invoice can participate at these special rates.

Enrol for this course