Polls and Politics: The Dilemmas of Democracy

6. When Push Comes to Shove: Push Polling and the Manipulation of Public Opinion

6 When Push Comes to Shove: Push Polling and the Manipulation of Public Opinion Matthew J. Streb and Susan H. Pinkus With each passing election cycle,polling has become a more integral part of candidates’ campaigns and has played an increased role in the media’s reporting of voting patterns. As the importance of polling to candidates, the media, and researchers increases, citizens must be aware of the quality of the polls being conducted. Many organizations are—by intent—failing to run scientific, reliable polls. This chapter focuses on one of the more devious, fraudulent, and increasingly popular ways of polling: push polling. We begin by explaining what push polling is, and then discuss the increased prominence of push polls and the attempts of polling organizations and legislatures to limit their use. We go on to mention several other polling practices that citizens must be aware of, and we close by arguing that these types of polls undermine a healthy democracy by reporting inaccurate information and raising cynicism about polls and politics in general. What Is a Push Poll? The term “push poll” is actually inaccurate, because a push poll is not a poll at all. The NCPP, a watchdog organization for the polling industry, referred to push polls as “masquerading as legitimate political polling,” when in fact “they are political telemarketing.”1 In a push poll, respondents are presented with hypothetical, sometimes blatantly false information under the guise of legitimate survey research. 95 Pollsters who conduct push polls are using unethical and unfair tactics to boost their candidate in the eyes of the voters or to create a negative impression of the opponent. They are using a deceitful method to “push” a person towards their candidate and away from a candidate that the respondent might have initially supported. The following is an example of a push poll mentioned in the Sarasota Herald Tribune.2 Caller: Are you planning to vote in the Smith-Jones race for senate? You: Yes. Caller: And whom do you plan to vote for? You: Jones. Caller: Tell me if your vote would be affected definitely, possibly or not at all if you knew that Jones had been charged with passing bad checks. You: Uh, possibly. Caller: And, if you knew that Jones had been arrested for drunken driving? Would that affect your vote definitely, possibly or not at all? You: Oh, definitely. Caller: Thank you very much. We appreciate you taking the time to help us. If you had said you were going to vote for Smith, the caller would have thanked you and hung up without asking any other questions. As you can see from the phone call, the interviewer did not say Jones was charged with kiting checks or was even arrested for drunk driving. They were hypothetical questions, but clearly left the impression that Jones actually did these things. The sponsor of this poll was hoping that the respondent believed that Jones had a problem. The callers even thanked the respondent for participating, making it seem like a legitimate poll. Push polls have several distinct features that might indicate to a respondent that they are participating in one. In a letter decrying the use of push polls, Bill McInturff, a partner at Public Opinion Strategies , a Republican polling/consulting firm, distinguished between push polling and legitimate polling. McInturff wrote: The rapid rise in the use of “push polling” as a campaign tactic has lead to significant confusion between advocacy polling and legitimate survey research. The differences between push polling and survey research could not be more dramatic: 96 Matthew J. Streb and Susan H. Pinkus —Every survey research firm opens its interviews by providing the name of the survey firm or the telephone research center conducting the interviews. Most push polls provide no name of a sponsoring organization. —Survey research firms only interview a limited sample of people that attempts to mirror the entire population being studied. Push polls contact thousands of people per hour with the objective of reaching sometimes hundreds of thousands. —Survey research firms conduct interviews of between five minutes for even the shortest of tracking questionnaires to over 35 minutes for a major benchmark study. Push polls are generally designed to be 30 to 60 seconds long. —Survey research firms use different questionnaire design techniques to assess how voters will respond to new information about your candidate and the opponent. Push polls are designed solely...