Hi there,
I am designing an experimental with several groups and wish to reliably determine sample size a priori for each group where I have an effect size set at 0.25 and power of 0.8.
I have a total of 16 groups where I have the following conditions (which are named this for confidentiality purposes):
Treatment A (2 levels +/-)
Treatment B (2 levels +/-)
Treatment C (2 levels +/-)
Timepoint (2 levels)
I've tried calculating sample sizes in g power and R (pwr package) and both give me different results. I am relatively new to this way of designing experiments and I wish to get as much advice as possible. Basically we want to determine whether our treatments influence our dependent variable at either time-point so should I be looking at running power calculations based on ANOVAs?
Any advice or help would be much appreciated!
Thanks in advance,
David