Hi,
I've been using growthcurver package for growth curves experiment. I have the output data per well- but now I want to group wells together according to my meta-data (for instance three biological replicates of a treatment). What is the best option for it? average the output variable for instance? perform ANOVA?
Thank you!