This is more of a series of related questions...
1. Generally speaking, what are the recommended methods/tools for power analysis when planning on using these models for repeated-measures type study designs (e.g., participant as random effect, plus a modest number of fixed effects)?
2. How applicable are other methods of power analysis (e.g., trying to extrapolate from say power analysis of simple two sample t-test, ignoring all other factors/effects)?
3. Is there a rule of thumb for the number of observations (or observations per participant) required for each fixed effect?
1. There are a few packages, e.g., http://www.bristol.ac.uk/cmm/software/mlpowsim/
2. Because of 1, you don't use these other ones. You could use design effect methods, but don't, use the packages for this.
3. No. Use packages like that listed in 1. Rules of thumb would depend on the circumstances, so input parameters related to your circumstances into packages like listed in 1.
I agree with the good sense in Daniel's post and would add this software
Optimal Design Software
The Optimal Design for Multi-Level and Longitudinal Research is a free stand-alone software program for performing statistical power calculations of impact evaluation studies, particularly group-randomized trials.
Raudenbush, S. W., et al. (2011). Optimal Design Software for Multi-level and Longitudinal Research (Version 3.01) [Software].
https://sites.google.com/site/optimaldesignsoftware/home
Also have a look into the simr-package: https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/2041-210X.12504
Best
Sven
Also, this just came across my feed (https://osf.io/6cer3/). Haven't read, but from the title looks like it might be relevant.
The introduction and tutorial in R by Kumle and Draschkow (2021) could come in handy. Here is the full citation.
Kumle, L., Võ, M. L.-H., & Draschkow, D. (2021). Estimating power in (generalized) linear mixed models: An open introduction and tutorial in R. Behavior Research Methods. https://doi.org/10.3758/s13428-021-01546-0
Good luck,
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics