Instant Debate Poll- Evaluate Carefully
The candidate debate is now a permanent fixture in a presidential campaign and an increasingly common feature of lesser campaigns. And so, similarly, is the post-debate poll.
Here's the scenario. The debate has just ended. Before the candidates can step back from their podiums, the pollsters are on the telephone asking who won. Won what? Who had the better approach to the critical issues? Who exhibited more confidence? Who took the more negative approach? Who had the "best" sound bite? Actually, the basic interest is in none of these. The interest, as though it were a sporting event, is simply who won. So the pollsters ask, and people oblige with their answers. Even worse are the post-debate polls that actually begin before the debate is concluded, so results can be reported within minutes of the debate being over.
Whatever the instant answers, they come from the relatively small proportion of the public who saw and heard the debate and have had no time to give it any thoughtful consideration. Furthermore, since the poll must be completed very quickly, the sample of people interviewed may not be representative of all viewers - and certainly not of all voters. The results of such instant debate polls may be more misleading than enlightening. The results probably are more reflective of previous candidate preference than of debate performance. Most viewers think the candidate they support "won."
An interesting case was the Carter-Ford debate in 1976 in which Ford mistakenly said that there is no Soviet domination of Eastern Europe and that he didn't believe the Poles consider themselves dominated by the Soviet Union. The instant post-debate polls showed the candidates about even with respect to who won. Follow-up polls after the media made the public aware of Ford's misstatement shifted markedly in Carter's favor.
Keep in mind that the instant post-debate poll: (1) measures only top-of-head reaction to the debates, (2) does not measure the debate's effect on candidate preference, and (3) applies only to those viewers who were contacted and participated. Remember that who won the debate may have little or no influence on candidate preference.
This is not to say that all post-debate polls are of no value, but they are of little use without some context. For example, a pre- and post-debate approach can show the debate's immediate impact on candidate preference among viewers. Then a subsequent poll taken after voters have been exposed to the media's coverage of the debate adds more insight as to the influence of the debate on candidate preference among all viewers. Additionally, look for polls that allow respondents some opportunity to express their views in their own words, as these can provide useful insight.
The basic message is that all post-debate polls are not created equal. Be alert as to when the poll was conducted - during the debate, immediately after the debate, or sometime later.
For more information about this and other polling issues, contact the NCPP Polling Review Board Members.