[Question] Can I present both the results of log-odds and average marginal effects (logistic regression)?
Hello,
Health economics student here. I newly enter the field for my master degree.
I'm writing my master thesis and I'm running into an issue while trying to interpret my results.
Initially, I've decides to use OR (odd ratio). However, my supervisor told me the way I've interpreted it is not correct (probabilitied and chance). But he told me he didn't really know either how to interpret it correctly and had advised me to use logs odds instead!
It's ok, but I really wanted to have more "concrete results" that can be get by anyone. Log odds just show if the relation is negative or positive.
So, I've just heard about average marginal effects and that is literally What I was searching for during all this time. However, now I'm wondering if is common for scientific papers to use both log-odds and average marginal effects?
Do i need to create two tables? Or could I only keep the table with log odds and present the results of marginal effets in the text?
Thank you