Surveys can be conducted with a variety of outcomes in mind. One might be to learn about a product or service; it could also be used to find out a personal information. A popular reason to conduct surveys, however, is to prove pre-conceived notions; unfortunately, this sort of survey is the tool of marketers and turn-coat surveyors – integrity isn’t a strong suit of many surveys. For instance, if TMA conducted a survey of how satisfied readers were with our articles and, at the same time, offered a free headphone, to a ‘randomly selected’ individual, the survey would be flawed.
A recent one approached 3GS and Pre users in order to ascertain which one suffered higher return/exchange rates. I have issues with it on many levels and I hope that you will read along.