Hello Ioana!

I'm not aware of guidelines that would answer your question but a closer look to the concept of test power might help you:

Test power is influenced by the following factors:

1. Statistical significance criterion (increasing alpha, increases your power)

2. Strength of effect (it's easier to detect a strong effect)

3. Sample size (test power increases with sample size)

Since you are planning a proof of concept study, you probable have no idea about the strength of effect. Addionally, you already know that your sample size will be low. What you could do, is changing the statistical significance but I question whether this would be necessary.

I assume that the reason for the proof of concept study is to find out what the strength of effect might be. Hence, your results of the proof of concept study would be a perfect starting point for a power analysis.

Best regards,

Robert

Thank you for the resources! I know the general things about and I am not looking to conduct a study, I am re-analysing somebody else's triaks. So I am actually looking for a more sophisticated resource if possible applied to RCTs, where there are a lot of proof of concept trials done.