READING POLLS

When I hosted a daily radio show I talked often about the way to look at polls. In this election, perhaps more so than any other in history, we are bombarded by dozens of polls, national polls, state polls, and on and on. When reading these polls one should focus on the average number, not on the specifics of each poll. Thus, you can go to some sites like realclearpolitics.com and get the averages (right now in a three-way race they have Bush at 48.7%, Kerry at 45.8% and Nader at 1.4%). While this would seem to indicate a Bush victory, one needs to dig deeper to properly analyze the situation. Pollster Nick Panagakis of the National Council on Public Polls writes the following about how undecideds behave on election day:

How will undecideds vote on election day? Traditionally, there have been two schools of thought about how undecideds in trial heat match-ups will divide up at the ballot box. One is that they will break equally; the other, that they will split in proportion to poll respondents who stated a candidate preference. But our analysis of 155 polls reveals that, in races that include an incumbent, the traditional answers are wrong. Over 80% of the time, most or all of the undecideds voted for the challenger. In 41 cases, or 32% of the 127, the incumbent ended with less than his stated poll percentage. This means that about one in four of all 155 polls actually overstated the incumbent's percentage. Of the 127 challengers who gained more undecideds than did incumbents on election day, 78 gained 10 or more points over their stated poll percentage. Making allowances for factors stated above, most polls appear to estimate support for the incumbent. All or most undecideds end up with the challenger regardless of the size of the undecideds.

I have heard this stated by polling experts in the past, but this is the first time I've seen it stated by a researcher with the figures to back it up. If this is true, then Bush needs to increase his margin in the polls by at least 2 to 5 more percentage points to pull it out. Of course, what this researcher can't know is whether other factors (like the screening technology that allows many people to avoid calls from pollsters) has altered the sample of people being polled.