We are discussing Sara Steegen's outstanding paper on multiverse in our lab today (). I agree there are many benefits to using a multiverse approach. I am considering using it in my research. Do you know of any useful criticisms of this approach?
Rereading Sara Steegen's fantastic multiverse paper carefully ... ouch. "There are many different defensible ways to analyze the data, see e.g. numerous examples of the same first author who analyzed them differently in every single paper …" 🤦‍♂️
"Increasing Transparency Through a Multiverse Analysis" by Sara Steegen, Francis Tuerlinckx, Andrew Gelman, Wolf Vanpaemel (machine learning)
I haven't come across the idea of a "multiverse analysis" before. Interesting, although potentially timeconsuming