I have a survey that was administered once to establish baseline, and again to follow up. The baseline was created from a random sample of the population (1,200 respondents). The follow up survey (~600 respondents) was administered to a randomly sampled panel of the baseline.
15 respondents in the follow up survey refused to give information for a variety of reasons, but did so for the baseline survey. For our analysis we have to do a basic extrapolation of the follow up survey results to the total population. This is done by dividing by ~600 and multiplying by the total population (somewhere around 10,000).
Methodology aside, I just need to know if I exclude these 14 surveys from the analysis of the results, does this affect the total population that would be used for extrapolation?