If I collect some data for a new treatment, but don't collect a control group, can I use the data from an older study that has the control condition? If so, how would I integrate this into my analysis? and what is the name for this? historical control?

If I were to say that the primary benefit of randomization (and possibly blinding) is that it makes simple statistical models match the true data generating process quite well, would that be a provocative statement? Or would that be obvious? Is there a good reference for this line of reasoning?

#r and #statstodon people: do you have any resources for running sensitivity power analyses for mixed effects models? (e.g. for fixed samples to determine what is the smallest reliable detectable effect at a specific power level)

Great guide to thinking about the age-period-cohort problem by Julia Rohrer: https://osf.io/preprints/psyarxiv/8zmuv

New Substack Post - A secret third way: Likelihoodist statistics

I provide a quick tutorial on conducting research using the Likelihoodist approach. It's step-by-step + #R code

https://open.substack.com/pub/mzloteanu/p/a-secret-third-way-likelihoodist?r=3b457w&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true