can I include subjects in the pilot study in the main study??


New Member
Usually people do a pilot study to estimate the variance of the outcome variable and then calculate the sample size for the main study. My question is: can I include the subjects used in the pilot study in the subsequent main study? I know by doing this the type-I error will be inflated. But I can't perceive why this is the case, and how to adjust for the inflated type-I error.

Any input will be greatly appreicated!