Variation in Iowa Polling


TWN reader Chris Brown shared this interesting and useful essay that probes significant variation in the polls emanating from Iowa.
As we try to understand the trends underway in the scramble towards January 3rd, this kind of steady analysis is valuable.
— Steve Clemons


5 comments on “Variation in Iowa Polling

  1. buck says:

    Several blogs posted comments on the ARG poll because it made a news splash. CNN even went so far as to claim that “Hillary is pulling away in Iowa”. TPM offered a link to the same story. But Charles Franklin, the cited expert, fails to make a point about shifting demographics. He correctly observes a clear trend of ARG polls to defy the average trend in favor of Hillary Clinton (and John McCain)–meaning, of course, that without the ARG results Hillary’s average numbers would have been even lower. But he fails to note that the latest poll was done in the run-up to a long holiday weekend for many. As Steve Benen points out, the trend this year is that more people–and mostly young people (under 40)–would have traveled away from home for the holidays. As such, many would have been absent to respond to the pollsters. If you look at the demographic breakdown of candidate support, Hillary gets the older vote, while the younger vote goes to Obama and Edwards. So the pre-Christmas polls likely would underestimate the Obama and Edwards support and overestimate the Clinton support. Gee, that’s exactly what the ARG poll shows, doesn’t it?!


  2. daniel says:

    “The clear result of the comparison above is that ARG generally showed a much better performance for Clinton than the trend in the first half of the year. From January through June, ARG usually had Clinton some 6-10 points above the trend estimate, with one exception in which ARG agreed almost exactly with the Clinton trend. During this period, ARG was the most discrepant of all pollsters from the Clinton trend.” Perhaps ARG is trying to lead voters?


Add your comment

Your email address will not be published. Required fields are marked *