Polling
On January 8th, according to SurveyUSA, Howard Dean would have been endorsed by Iowa with 25-34% of the vote. On January 20th, the day after the Iowa Caucuses, Iowa chose John Kerry, leaving Howard Dean with only 18% of the votes.
One possible read on this is that of the 100,000 to 125,000 Iowans that participated in the caucuses, somewhere between 7000 and 20,000 Iowans changed their minds between January 8th and January 19th. This is possible, of course.
But isn't the more rational conclusion to reach is that SurveyUSA screwed up? Perhaps its survey methods are flawed; perhaps its survey methods were corrupted; perhaps its statisticians are pranksters. Perhaps--and here's what I'm really getting out--there isn't really all that strong a correlation between what people say on the phone and what they do in the voting booth.
The Howard Dean "fall" is just an example. The news is riddled with many examples of supposed "sea changes" or "about faces" in public opinion that may never have happened. Isn't it more likely that for every time a whole bunch of people went from being adamantly opposed to Issue X to being adamantly for Issue X, there are ten more times where the pollsters just didn't find truth through the telephone in the first place? Isn't it more likely that Howard Dean was never the front-runner?
One possible read on this is that of the 100,000 to 125,000 Iowans that participated in the caucuses, somewhere between 7000 and 20,000 Iowans changed their minds between January 8th and January 19th. This is possible, of course.
But isn't the more rational conclusion to reach is that SurveyUSA screwed up? Perhaps its survey methods are flawed; perhaps its survey methods were corrupted; perhaps its statisticians are pranksters. Perhaps--and here's what I'm really getting out--there isn't really all that strong a correlation between what people say on the phone and what they do in the voting booth.
The Howard Dean "fall" is just an example. The news is riddled with many examples of supposed "sea changes" or "about faces" in public opinion that may never have happened. Isn't it more likely that for every time a whole bunch of people went from being adamantly opposed to Issue X to being adamantly for Issue X, there are ten more times where the pollsters just didn't find truth through the telephone in the first place? Isn't it more likely that Howard Dean was never the front-runner?
0 Comments:
Post a Comment
<< Home