Carrying out a Pilot Study
Carrying out a pilot study is absolutely essential. You will be able to test out your proposed methodology, learn the problems (there are always problems!), and become able to estimate the population mean and standard deviation for your sample size calculation.
Without a pilot study you run a grave risk of not realizing a problem before you are half way through the research project. Failure of a pilot project is a positive thing. Failure in the real project is a disaster.
A pilot project should aim to test:
The system for patient identification
The rate at which you are likely to recruit patients
The practicality of the intervention
The data collection system
The mean and standard deviation of the primary outcome measure
From this list it can be seen that you are not trying to do any sort of statistical analysis on the results of the pilot study. It is more to test the systems that you are putting in place - can you identify the patients, can you perform the intervention and can you collect the results.
