Introduction to sample size calculation

[4 half days] - [English]

Wondering how many patients or mice you have to include in your experiment? Do you need to justify a sample size to an ethics committee?

Sample size calculation consists in determining the number of observations necessary to be able to show an effect if it exists - in terms of probability (cf. Power).

Training aims
At the end of this training, the participant will be able to identify the information needed to calculate the sample size, understand the impact of a small change in the basic assumptions, choose the appropriate software to perform the calculation, find out how to access it and find help if necessary...

Prerequisites
This training course requires some basic statistical knowledge.

Content

• Concept of hypothesis testing (null and alternative hypothesis, one- and two-sided test, p-value, (non-)rejection of the null hypothesis)
• Type I and II errors
• Power of a test
• Effect size
• Loss rate
• Formalization
• Sample size calculation software (PASS and GPower)
• Power and effect size calculation a posteriori
• Examples and application exercises (means, proportions, correlations, regression, contingency table, ...)

Rate
This training course is recognized by the IABE, enabling participants to earn CPD points.
(Note that this is true for all SMCS courses.)

Tools used during training
G*Power

Methods and method families discussed
Planning
   Sample size calculation