WIAS Course Design of Experiments

Course schedule

Dates Start time End time Location Coordinator registrations
16-17-18 December 2020 online course Marianne Bruining 24 / 24 Add to waiting list

Course description

Course description

The aim of this course is to provide an understanding of the statistical principles underlying experimentation. A proper set-up of an experiment is of utmost importance to be able to draw statistically sound conclusions.

The role of sample size, randomization and the reduction of unwanted noise factors will be highlighted. The way errors propagate will be discussed. The difference between experimental unit and measurement units and consequences for statistical analysis will be discussed.

Examples of basic designs (CRD, RCBD, BIBD), but also more advanced designs will be discussed. Lectures will be interchanged with computer practicals, using R. The final half day will be devoted to discussion of the own experimental designs of the participants.

Learning goals

  • Understanding the principles of experimental design: the role of sample size, randomization and reduction of noise factors in the efficient set-up of experiments
  • Experimental units vs measurement units, and consequences for statistical analysis
  • Examples of commonly used experimental designs, and some more advanced designs
  • Basic understanding of propagation of errors (variances)

Target group / group size

PhD candidates / 10 to 24 participants

Prior knowledge required

Basic knowledge about statistics: summary statistics,  parameter estimation, hypothesis testing, t-tests, simple regression, one-way ANOVA (will be refreshed)


Preparing a short presentation of own experimental design

Other course info

The course duration is 2.5 days and the course is given twice a year. Self study hours: 0.5h. Credit points: 0.8 ECTS

Course programme

Day 1:

09.00-10.45 Session 1: Refresher statistics: summary statistics, t-tests, regression and ANOVA; statistical software R; basic design: three R’s & terminology

11.00-12.30 Session 2: R1: Replication: power and sample size

12.30-13.30 Lunch

13.30-15.15 Session 3: Power and sample size (cont), principles of error propagation

15.30-17.00 Session 4: R2: Randomization

Day 2:

09.00-10.45 Session 5: R3: Reducing noise factors:  blocking and covariates; ANCOVA

11.45-12.30 Session 6: Special designs: CRD, RCBD, BIBD

12.30-13.30 Lunch

13.30-15.15 Session 7: Special designs: split-plot design; mixed models

15.30-17.00 Session 8: Special designs: Latin-squares, repeated measures situation, cross-over design; alpha design

Day 3:

9.00-10.45 Session 9: Apply to own case (groups of 4)

11.00-12.30 Session 10: Apply to and discuss own case (groups of 4); presentation 6x15 min