First things first, while I have no formal schooling in survey research back in the day I worked at more than a couple market research companies and political polling departments (including SRBI and NBC's Election Division) from being a lowly interviewer to Head Field Supervisor to Coding Director and I have to say something is rotten in the state of Denmark.
I'm certainly no expert on the subject, but it doesn't take an expert to know that if your sample sucks (no cell phones called which leaves out many younger, more Kerry-friendly voters who only use cell phones) and if you screen out people who have only recently registered (in the middle of a historically massive voter registration drive conducted by Dem-friendly groups) and if you weight the whole damn survey in favor of Republicans, you may just get flawed results.
Gallop's latest poll interviewed 12% more Republicans than Democrats even though nothing would suggest that there are 12% more Republicans than Democrats in the voting population.
A paranoid person might think we are being set up to think that everyone likes Bush and we should jump on the winning team or to dismiss any possible idea of thievery on a grand scale this time around because, after all, Bush was way ahead in the polls all along.
No comments:
Post a Comment