Polling may affect elections, not just gauge them

    50

    By MELINDA SEMADENI

    By crunching numbers and correlating them with data, pollsters can play an integral role in shaping the public opinion they attempt to measure. Pollsters are trained to ask questions, but recently some of their results are making some Americans distrustful of their research.

    In an era where public opinion can be measured with the touch of a button, a click of the mouse or by phone, a growing number of the American electorate are becoming wary of public opinion and market research.

    However, researchers’ methods raise questions about polling’s influence on voters, politicians and the media.

    BYU political science professor, David Magleby, said polling affects voters first and foremost in the selection of candidates.

    Magleby said he believes another impact of polling is inappropriate application in the early stages of a race, when people don’t have enough information to make a decision. He said these polling results could discourage them from becoming involved in the election.

    “Polling is also influential when an unknown candidate competes against a well-known candidate for the nomination,” Magleby said.

    Maury Giles, a political and public affairs research manager for Wirthlin Worldwide, an opinion research firm, said their company tracks public opinion by measuring how salient the issue is on three levels.

    “The first level is general, how important is the issue to them overall, as a country,” Giles said. “Second, how important it is to them personally on a day-to-day basis, and the third level is voter determinant or what issue has the most influence on how they vote.”

    The three levels vary and often don’t all match, even for the same person, Giles said. Wirthlin tries to measure an individual’s attitude and how their attitude affects their behavior and is driven by their anchoring values, he said.

    “In the mind of the voter or general public, they don’t differentiate between issue position and leadership imperatives, they are intertwined. Candidates must communicate leadership imperatives through their position of the most salient issues,” Giles said.

    “The American electorate feels the country and the economy are going in the right direction,” Giles said. “So while they may personally dislike the president, they approve of the job he is doing in office, because they have seen his ability to communicate issues like welfare and education.”

    Magleby said most politicians are either delegates or trustees. Some choose to mirror their constituents on every issue, while others try to led public opinion on certain issues.

    The recent House Judiciary Committee’s impeachment hearings is one example, according to Magleby, of some politicians following constituency’s opinions and others acting upon their own conscience.

    “If we were merely to follow existing public opinion, things like the civil rights revolution and extending the franchise to women, would never have happened,” Magleby said. “It took politicians who were willing to risk offending whatever the current opinion was to try to transform, educate or change the public opinion.”

    Another problem, Magleby said, is the area of public opinion called selective perception.

    “People at all levels from ordinary citizens to politicians have a tendency to want to read into these things what they want to see,” Magleby said.

    Magleby said one illustration of selective perception is in the results of the midterm elections. November’s elections left Republicans mystified that voters were not incensed with the president and Democrats seeing the results as an exoneration of the president’s actions.

    “Both sides have to worry about the tendency to read too much into it from their point of view, when the reality is that it was neither,” Magleby said.

    Polling affects the media when they don’t understand how to interpret the information connected with the polls, Magleby said. He said he believes such methods can disrupt elections and benefit people with high-name identification or lots of money.

    “Another problem is when polling is done on the cheap and the media outlet interviews the wrong universe of people,” Magleby said. “So instead of interviewing

    registered or even more appropriately likely voters, they interview anybody who answers the phone or the door. A lot of those people don’t vote so they do projections based upon an inappropriate universe.”

    Giles said there are differences between media polling and industry polling.

    “The media love to jump on and sponsor polls, it gives them good talking points, but because of their constraints they don’t always follow stringent sampling and scientific methodology,” Giles said. “The industry lives and dies by their numbers.”

    Howard Christensen, professor of statistics at BYU, said consumers should look for several things when evaluating a poll.

    “First, voters need to identify the polling organization and decide if it is irresponsible,” Christensen said. “Voters can also look at what kinds of questions were asked and how they were asked, as well as looking at looking at the size of the sample to see if it is representative of the population.”

    Voters can check who sponsored the poll and what kind of return they were expecting from it, Christensen said. He encourages the public to be skeptical about polls.

    While there are not stringent government regulations on the polling industry, the American Association of Public Opinion of Research and its members subscribe to a basic code of ethics.

    According to AAPOR’s web site (http://www.aapor.org/ethics/), research professionals from more than 1,400 research firms adhere to the code. AAPOR’s code delineates acceptable ethical practices as they relate to the public, clients, and respondents. The code also sets the standard for minimal disclosure on public opinion or market research.

    Print Friendly, PDF & Email