“Polling” can be a loaded term. Since the early days of polling, the field has had to defend its reputation against the damage done by individuals and organizations who manipulate or even fabricate poll findings to bring the force of public opinion to their agenda. Some polling organizations release partial results, or only release numbers favorable to their client. Some conduct polls with biased question wordings.
A push poll is not a poll at all. The goal of a push poll is not to gather information from respondents, but to push information to them. Push polls use biased language and leading questions to convince respondents to vote for – or against – a candidate or referendum. In this way, it is the equivalent of a marketing call. However, by describing the calls as research, a push poll can impart a veneer of impartiality to the messaging. In the most egregious cases, push polls can be used to start whisper campaigns, spreading false rumors to thousands of potential voters.
A push poll should not be confused with a legitimate poll that includes message-testing. Legitimate polls can, for example, provide negative information about an individual or a bill in order to gauge public response to that information. But the purpose is to measure the effect of that information on the respondent, not to communicate information directly to the public.
“Frugging” – fundraising in the guise of research – is another practice that uses the appearance of research to give legitimacy to another activity, in this case a request for donations. Closely related to frugging is “sugging” – sales in the guise of research. As with push polls, no data is actually collected in a frugging or sugging campaign.
Professional polling organizations stand united in their condemnation of these practices. Ethical pollsters fear that respondents who do not recognize a fraudulent poll for what it is will assume that all pollsters write biased questions or use their questions to open a sales pitch. These perceptions can damage confidence in survey research overall. So the polling profession makes education about legitimate and illegitimate polls an important priority in their outreach to the broader public.
How can you recognize a call from a purported research poll that is in fact intended for a wholly different aim? First, legitimate polling organizations will identify themselves at the very beginning of the poll. No legitimate poll will ever ask respondents for a donation or purchase. Finally, demographic questions about respondent age, education, gender or characteristics are of utmost importance to real research. Not only do these questions offer the opportunity for deeper analysis, many demographics are used to weight polls to Census benchmarks. Organizations that intend to sway or sell often do not bother to collect demographic information.
To learn more, see the American Association for Public Opinion Research, What is a Push Poll.
See also: LINK TO HOW TO READ A POLL
noun [countable]
an opinion poll (= an attempt to find out what people think about a subject by asking them questions about it) whose real purpose is to influence people's opinions rather than to collect information about them
'A Bedford resident complained to the state attorney general's office about a campaign telephone poll from Hawkins that was critical of Sanborn. Sanborn called such push polls a "dirty way" to get elected.'
Concord Monitor 11th September 2012
'The technique of push polling is part of the election battle being fought on the ground in the swing states where the margins of victory have been narrow in past elections.'
The Guardian 3rd October 2008
With the US presidential election now only weeks away, supporters of particular candidates are becoming increasingly zealous in their attempts to influence the outcome of the vote. Among the range of tactics employed in an effort to convince potential voters that their man (or woman) is the one to go for, is the push poll – political advertising in a cunning disguise.
the essential difference between a push poll and a valid opinion poll, is that the latter merely attempts to elicit opinions, whilst the former aims to change and influence them
In essence, a push poll is something approaching a smear campaign (= an attempt to damage someone's reputation by giving false information about them) masquerading as an opinion poll (= a survey used to gather information about general opinions on a particular topic). A push poll typically involves phoning very large numbers of voters and, in the guise of doing a survey to find out how certain pieces of 'information' may affect their voting preferences, feeding them false or damaging information about a particular candidate. The idea then is to 'push' them away from any rival candidate and steer them towards the candidate that the caller supports. Crucially then, the essential difference between a push poll and a valid opinion poll, is that the latter merely attempts to elicit opinions, whilst the former aims to change and influence them. In a push poll, large numbers of respondents are contacted, but little or no effort is made to collect and analyze response data because no-one is really interested in this information once the phone call is over, the main aim having been to negatively influence the targeted candidate.
The derived term push polling is sometimes used to refer to this as a technique, and one of the most notorious implementations of it occurred in the US Republican Primaries of 2000 (in the USA, the Primaries are the selection process by which voters within a particular party choose its presidential candidate). Allegedly, supporters of George W. Bush used push polling to sabotage the campaign of rival John McCain, reportedly asking if voters were more or less likely to vote for McCain if they had known that he had fathered an illegitimate (black) child. In the 2008 presidential election, some Jewish voters were targeted by a push poll attempting to damage Barack Obama by false accusations of a link to the Palestinian Liberation Organization. As we speak, allegations of push polls are rife as campaigns for candidates in the 2012 presidential election enter their final phase.
Background – push poll
The expression push poll first emerged in the mid-nineties in the USA, and is mainly associated with US politics, though there's some more recent evidence for its use in Australia. The technique's origins are often linked to the aggressive tactics associated with the late Lee Atwater, an American political consultant and strategist to the Republican Party.
The word poll dates right back to Middle English, when it originally meant 'head', and later, by extension, 'an individual person among a number'. This in turn developed into the sense of 'the number of people determined by counting of heads', which by the 17th century had morphed into simply 'the counting of heads or votes'.
by Kerry Maxwell, author of Brave New Words
Last week …
Read last week's BuzzWord. skeuomorphic.
“Push polls” — which are not really polls at all — are often criticized as a particularly sleazy form of negative political campaigning. Voters pick up the phone to hear what sounds like a research poll. But there is no effort to collect information, which is what a legitimate poll does. The questions are skewed to one side of an issue or candidate, the goal being to sway large numbers of voters under the guise of survey research. But the fact that a poll contains questions with negative information about one or more candidates does not make it a push poll. Campaigns regularly conduct genuine surveys that test campaign messages and advertising, including negative content. Push polling is so incompatible with authentic polling that the American Association for Public Opinion Research (AAPOR), the American Association of Political Consultants (A.A.P.C.), the Council for Marketing and Opinion Research (CMOR) and the National Council on Public Polls (N.C.P.P.) have all denounced the practice. “Survey research organizations are always concerned about establishing a good rapport with respondents, both in order to complete the interview and more generally to maintain a positive image for the industry,” said Michael Traugott, a professor of communication studies and political science at the University of Michigan. “Having a bad experience with something that seems like a very biased poll is harmful to both these interests.”
Image
The A.A.P.C. website states that advocacy phone calls, which skip the polling pretense and simply try to persuade voters, are a legitimate campaign practice, but the organization condemns calls that deceive by pretending to be a poll or by presenting deceptive information or by not accurately identifying the sponsor.
A legitimate survey will identify the call center, although it often does not mention the candidate or political party sponsoring the research because that could influence the results. The survey will contain more than a few questions and generally will ask about more than one candidate or mention both sides of an issue. Demographic questions, such as those on age, race and education, will come at the end of the questionnaire. And the number of respondents to legitimate surveys will normally be between 500 and 1,000.
“Good message testing includes pro and con statements about both your candidate and his or her opponent,” said Nancy Belden, partner of Belden Russonello Strategists. “You need to explore the strength and weaknesses on both sides.”
Ms. Belden’s firm provides message development and communications consulting for nonprofits, political campaigns and other clients. She said that people can sometimes be confused when hearing negative questions and think surveys conducted by political consultants are push polls. She said that “the pro and con options in a good questionnaire are truthful; they may test the best way to say something, but they don’t lie. “
By contrast, with a push poll it is often very difficult to find out what organization is behind it. Only a few questions are asked, all about a single candidate or a single issue, uniformly negative or rarely positive. Time is not wasted asking demographic questions because there is no analysis being done. And the number of people called is very large, sometimes several thousand.
Ms. Belden called the practice an “ugly campaign technique,” adding: “In addition to being contrary to our codes of ethics, it’s not research. It doesn’t further what we are hired to do.”