Sample size...

#1
have found numerous on-line calculators and have read through some forum postings but am now more confused than ever.

I have been asked to calculate the sample size needed for a study looking at cellular changes after a type of operation. Where do I start when I have no idea what the "population" is or no idea the extent of the change that is expected ("could be anywhere from 5 - 60% difference").

Any help or guidance?

Thanks in advance
 
#2
You must first determine which statistical test you expect to use. The sample size calculation should be specific to that test. The size of the difference is a judgment call. Maybe enter 60% and see if the calculated sample size is reachable.