I have done a lot with polling on this blog and with my own students. It used to be a great way to figure out who was voting for whom and recently has become a nightmare for pollsters and while not a return to the pre-WWII days, less of a reliable resource than before. This NYTimes article does a good job of explaining why, namely that with the prevalence of cell phones not everyone has the same likelihood of being contacted and the number of people who say they will vote and don't going up the models are no longer accurately able to account for this variable.
No comments:
Post a Comment