Thursday 18 February 2021, 16:00-17:00, online.
Klaas Wardenaar: Power analysis for multilevel ESM studies.
Diary studies are increasingly being used in psychiatry, but designing a diary study remains a challenge. This is also true for determining the optimal sample size: how many persons do and how many assessments per person do we need to test our hypotheses with sufficient power in multilevel analysis? Due to the complexity of the design and analyses there are no simple formulas or rules of thumb for this situation. In this presentation I will first explain the reasons for this. after that, I will show how changes in the number of persons and the number of assessments per person affect the power of multilevel diary studies. Finally, I will show some examples of simulation methods to illustrate how sample size can be properly estimated a priori.
The scripts will be added to the google drive soon.