Polls may stifle voters’ interest

by Tracy Ellingson

Republican presidential hopeful Bob Dole says he will defy pre-election polls that show him trailing President Clinton, and win this year’s election.
Poll results, which have gained credibility with the public since the 1940s improved scientific methods of obtaining results were developed, are determined when pollsters ask a random sample of voters specific questions about election issues and candidates.
Each person interviewed is asked questions to determine the likelihood that they will vote. Those who indicate they will most likely not vote are thrown out of the final results.
Originally, journalists used polls to decide what issues to focus on in their coverage.
Rob Daves, an election pollster for the Star Tribune, said news journalists in general publish results to give voters a clear picture of how other voters are looking at candidates and issues.
“We try to tell our readers what is real at this point in time,” Daves said.
University political science professor Charles Backstrom, however, said polling results can affect how a person will vote in the election.
“The problem is that polling is supposed to be a testing instrument,” Backstrom said. “(Pollsters) are supposed to be just checking on the public, but the publication of the results possibly can affect the way people think about the election.”
Backstrom said when a poll shows one candidate several points ahead of the opposition, the results might convince voters who support the leading candidate that their votes are unnecessary. Conversely, the results might indicate to the opposition’s supporters that their votes would not help their candidate win. Some voters feel polls “have taken the spark out of elections and there’s no reason to vote because of that,” Backstrom said.
Backstrom added that poll results are scientifically accurate but can only reflect the opinion of the public at the time of the poll. Issues that arise after the poll may affect the final result on the day of the election.
“A change in the result wouldn’t mean that a poll was wrong,” Backstrom said. “It would mean the supposition under which people decided whether they were going to vote — or not — changed.”