Formalized Argument for Design Being Longer in QCD than RCT: Part 3

This is going to be my last post about my argument that if you cannot run an RCT, then you will have to spend more time in the design stage of a quasi-experimental controlled design (QCD). I’m going to make the argument mathematically, aiming to show how time and effort can act as substitutes for randomization. There’s a couple of pictures at the bottom…