Hi,
I have a lot of parameters coming from 1 behavioural experiment. Usually, if you have a lot of parameters, you should adapt your alpha for multiple testing by dividing by the number of parameters.

But, as my parameters are NOT independent (e.g. frequency and duration of 1 behaviour and parameter A, parameter B and parameter A divided by A+B), how do I treat my alpha?

If 2 parameters are dependent in one way or the other, do I treat them as 1 parameter?

Thanks!

Annetrude.