S4E17 Variable Transformations: Box-Cox, Fox in Socks
Quantitude
Nonlinear Transformations - How to Make an Interaction Go Away?
In psychology rarely if ever do we have a metric that we really interpret directly anyway. What worries me in my own work is that linearization of the predictors in the model can make an interaction go away. If you are being conscientious and you want your variable to be more normal in its shape, absolutely a nonlinear transformation can help with that. But you have to know you just changed the correlation structure among your outcome and all other variables in your model.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
