|
ace: when you asked the folk at your meeting yesterday for their opinions about the iraq war, did you also ask them to characterize themselves politically? did you keep track of the responses? without that, what makes you think that this is a representative group? and representative of what? did you publish the results of your poll somewhere?
did you make a press release in which you basically distorted the results that you did obtain?
just wondering.
or maybe what is really important about a poll is that its results "feel right" to you--not its design, not its methodology, not whether the press version of the results are the same in any meaningful way as the poll results...
some thursday nights find me in a publick house with friends enjoying a fine malt beverage or 6. maybe tonight, if i go, i'll conduct my own "poll" and post the "results" here. it'll be just like your canvassing your meeting.
the trick is that it is not the case that the results you obtained from whomever your sample was are not interesting--they just dont operate with ANY of the same methods or implications of even a badly designed poll. and because that is the case, you really cannot use that information to legitimate the poll--particularly when the problems with it have already been pointed out in detail. if you want to defend the poll, defend its methodology, defend how it groups respondents, defend the way it is distorted in the press release.
__________________
a gramophone its corrugated trumpet silver handle
spinning dog. such faithfulness it hear
it make you sick.
-kamau brathwaite
|