Quote:
Originally posted by Larry_OHF:
I don't beleive that survey. Afterall...they did not interview every
|
Errrrrm... yes, that's how surveys are conducted, isn't it? Of course you can't interview the entire population, but that doesn't mean there are NO valid conclusions to be drawn from such surveys. Of course the testers should take care that they pick representatives from ALL societal groups (unless the survey specifically researches a single group) and that they interview a large enough group of people. But if a survey is conducted in a scientifically sound way, you can definitely use it as a (tentative) guideline.
That said, of course it's good to be sceptical (as am I, see my post above) and question the specifics of the test like you do - like what kinds of people were interviewed, and how many of them. [img]smile.gif[/img]