Hello everyone,
I have a longitudinal quasi-experimental study to evaluate the impact of a program, in which I have several data collection waves (let us say three).
I have four broad categories of outcomes (math, language, musical, and psycho-social).
Since I have many outcomes, I understood I should adjust for multiple testing. I decided to use the Westfall-Young correction, with the package -wyoung-.
However, I am unsure of how to proceed.
Should I have one correction per broad category of outcome per wave (so 12 corrections in total)? Or one per broad category, all waves confounded (so 4 corrections total)?
To further complicate things, I then have subgroup analyses...
I would really appreciate any insight into this!
Thank you,
Julie
I have a longitudinal quasi-experimental study to evaluate the impact of a program, in which I have several data collection waves (let us say three).
I have four broad categories of outcomes (math, language, musical, and psycho-social).
Since I have many outcomes, I understood I should adjust for multiple testing. I decided to use the Westfall-Young correction, with the package -wyoung-.
However, I am unsure of how to proceed.
Should I have one correction per broad category of outcome per wave (so 12 corrections in total)? Or one per broad category, all waves confounded (so 4 corrections total)?
To further complicate things, I then have subgroup analyses...
I would really appreciate any insight into this!
Thank you,
Julie
