The suburb in which you conduct this survey would have a large impact. For example there are (in my town) student suburbs, rich suburbs, bogan suburbs, young family suburbs, industrial suburbs with a few houses in, immigrant suburbs which started off as government housing and were bought out, woggy (Mediterranean nationalities) suburbs, Pommy Suburbs, Poor suburbs etc etc.StBishop said:
So yeah doing a survey in my current neighbourhood would result in a very high percentage of "Health freak" answers and "rich kid" answers and a few "student" answers. So things like those healthy microwave dinners, delicious shit you can't get without rich parents, and instant noodles.
In my old suburb you should expect a fuck tonne of instant noodles, and various types of asian foods, because I lived in a majority student/immigrant area and the large portion of international students who are Asian live in that suburb, the grocery stores reflect this.[/quote]
That is pretty much exactly what I meant when I said that "no true random sample is possible" left it in the quote above and everything. I think standing in a local mall and flagging people down is the best way (because people will travel a way out of their way to get to a mall) but even then your sample is going to be exclusively "people who go to the mall and have the patience to answer a questionnaire..." and still not an unbiased sample. In the case of preferred genres of literature that the Escapist is going to produce a considerably more biased survey than at a local grocery store and the grocery store bias can be handwaved almost entirely by saying "random sample from this neighborhood" and just verify that all questioned live within a certain radius of the store. (still isn't completely unbiased but by that point it becomes negligible)
