Exactly my point they are looking at a volume number to get a representative sample not a cross section. If you don't answer do they call someone else of the same age, base, seat and commute status? The answer is no, they move on to the next randomly generated phone number with the assumption the database they are provided is equally valid for all individuals. There is no time of day adjustment for those on international layovers or even those on the west coast. Sampling hours, availability of the subject, accurate phone numbers and other factors all add error to the sample.