Alan Abramowitz at the Emerging Democratic Majority weblog DonkeyRising does some number crunching and confirms my earlier hypothesis that all the recent poll movement is mere statistical noise:
It isn't just the tracking polls that seem to be fluttering randomly in the wind. An analysis of polls by six major media outlets (Time, Newsweek, CNN/USA Today/Gallup, CBS/New York Times, Pew, and Fox) that released polls during the first half of October and during the second half of October reveals that there is a strong negative correlation (-.84) between the early October results and the late October results.
In other words, the better Bush was doing relative to Kerry in the early October poll, the worse the was doing in the late October poll. Some of this might be explained by the well known phenomenon of regression toward the mean. If a poll's early October sample had either too many Bush supporters or too few Bush supporters just due to chance, it's late October sample should be closer to the true population mean. But these results went well beyond regression toward the mean. Every poll that was above the overall mean in early October was below the overall mean in late October and every poll that was below the overall mean in in early October was above the overall mean in late October. What this bizarre pattern suggests is that the movements of the major media polls in October, like the movements of the tracking polls, reflect sampling error and peculiarties of the polls rather than real change in the underlying preferences of the electorate.
Earlier, Abramowitz performed a similar analysis of the tracking polls, and found the same result:
So does it make sense to monitor the daily movements of these tracking polls? The answer is that if you’re hoping to learn something about real trends in support for the presidential candidates, it probably doesn’t make sense. That’s because there is no correlation between the day-to-day movements of the four tracking polls. In other words, they don’t move together—each poll’s movements are unrelated to all of the other polls’ movements. For example, the average correlation between the daily movements of the Zogby Poll and the daily movements of the other three polls is -.18. The average correlation for the Rasmussen Poll is -.07, the average correlation for the TIPP Poll is -.09, and the average correlation for the Washington Post/ABC News Poll is -.12. The combined average for all four tracking polls is -.11. These weak negative correlations mean that there is actually a slight tendency for the polls to move in opposite directions.
What these results indicate is that the day-to-day movements of the tracking polls are essentially random. Rather than reflecting real shifts in voter preferences, the day-to-day movements of the tracking polls are simply reflecting sampling error. This doesn’t mean that the overall results of these polls are wrong. In fact, the average margin between George Bush and John Kerry in the tracking polls has been very close to the average margin in other recent national polls. It just means that the day-to-day shifts in the tracking polls are probably not real and that the real level of support for George Bush and John Kerry within the electorate has not changed over the past few weeks: the presidential race has been very close since the beginning of October and it is likely to remain that way until Election Day.
So relax political junkies. Stop obsessing over the daily movements of the tracking polls and get a life! Follow the World Series. Follow your favorite college or professional football team. Follow the weather report. Follow something that is more real than the day-to-day movements of the tracking polls.
So retain your sanity and don't pay any attention to last-minute poll fluctuations. Instead, work hard and get out the vote!
Comments