I would like to ask you about the number of participants of the pilot study, is 40 persons are enough if the participants of the main survey are 385 persons?
First of all, Thanks for you. Your answer is really clear and very understandable. But, Could you explain what kind of formulas can use to calculate the size of sample for the pilot study if the goal of the test to reliable the survey items by Cronbach's alpha? @Heiko Breitsoh
The formulas I was referring to are relevant for power, not reliability.
Computing Alpha (or, more appropriately if your are using CFA, McDonald's Omega) will probably not be an issue with 40 subjects.
However, if your use CFA (which is probably appropriate and important), your model may not converge, and even if it does, you probably won't have enough power to detect any misfit between your model and the data.
It is really clear now. Thank you for giving me the opportunity to have a very useful conversation with you. .After a chat with you, I feel more confident and capable, and for that I will always be grateful.