Capitalising on covariates in cluster-randomised experiments

Vanhove, Jan

doi:10.31234/osf.io/ef4zc

Back

Journal article

Capitalising on covariates in cluster-randomised experiments

BCU-LLM

Vanhove, Jan University of Fribourg

31.08.2020

Published in:

PyArXiv. - 2020, p. 1-45

English In cluster-randomised experiments, participants are randomly assigned to the conditions not on an individual basis but in entire groups. For instance, all pupils in a class are assigned to the same condition. This article reports on a series of simulations that were run to determine (1) how the clusters (e.g., classes) in such experiments should be assigned to the conditions if a relevant covariate is available at the outset of the study (e.g., a pretest) and (2) how the data the study produces should be analysed if researchers want to maximise their statistical power while retaining nominal Type-I error rates. The R code used for the simulation is freely accessible online, allowing researchers who need to plan and analyse a cluster- randomised experiment to tailor the simulation to the specifics of their study and determine which approach is likely to work best.

Faculty

Faculté des lettres et des sciences humaines

Department

Département de plurilinguisme et didactique des langues étrangères

Language