List of polling firms

Public opinion polling is a form of gauging the public's responses to political, social, and cultural issues. National polling typically involves around 1,000 respondents answering a series of questions designed by a polling firm. These responses are then recorded and weighted to more accurately represent the population for which the poll was designed.

This page covers the fundamentals of polling as a practice and some of the polling firms covered by Ballotpedia.

Types of polling

Live-interviewer polling

Live-interviewer polling is conducted either in person or by phone. Paid or volunteer interviewers ask respondents a set of scripted questions to gauge the interviewees' opinions. This practice is sometimes referred to as traditional polling.

Online and interactive voice response polling

Online polling is conducted via the Internet using a variety of ways to obtain a sample of respondents. Some online firms, like SurveyMonkey, offer some respondents from other polls the opportunity to participate in a politics-specific poll for public opinion research. On the other hand, YouGov recruits interviewees through email and through non-political advertising on search engines.

Voice response polling—sometimes called robopolling—uses an automated script to ask interviewees questions. This method is sometimes used by Trafalgar Group, an Atlanta-based company that correctly predicted the outcome of the 2016 presidential election.

According to Nate Cohn of The New York Times, online polling is often scrutinized because it has more difficulty ensuring a randomized sample of respondents. In November 2015, Cohn wrote, "Random sampling is at the heart of scientific polling, and there’s no way to randomly contact people on the Internet in the same way that telephone polls can randomly dial telephone numbers. Internet pollsters obtain their samples through other means, without the theoretical benefits of random sampling. To compensate, they sometimes rely on extensive statistical modeling."

Methodological practices

Weighting

Weighting is the practice of adjusting a poll's sample to more accurately reflect the population aiming to be portrayed and to make the data more able to yield a statistical inference. According to the American Association for Public Opinion Research, there are three major types of weighting that public opinion polling firms engage in:

• Adjusting for probability: The responses of a survey are weighted to accurately represent the probability of choosing a respondent.
• Adjusting for sample design: One demographic is sampled more heavily to more accurately assess that group's opinions on a topic.
• Adjusting for demographics: After data is collected, it is adjusted so that it more closely reflects known aspects of the polled group, such as statistics from the U.S. Census.

Respondent selection

When conducting public opinion polls, there are two major methods of selecting respondents. Some firms—such as Iowa's Selzer & Company—work from lists of registered voters in a targeted geographical area. Some firms use voter rolls but only call those who are considered likely to vote, using factors like previous voting activity as a measurement. Others, like the Quinnipiac University Poll, use random digit dialing to contact participants, determining their voter registration and the likelihood of these respondents voting. According to Harry Enten of polling analysis site FiveThirtyEight, the difference in respondent selection means that "pollsters aren’t necessarily calling the same universe of people."

Rolls of voters

Many phone-based polls begin with rolls of registered voters, obtained from state election officials, in a targeted geographic area. The pollster then calls those voters who opt to give their phone numbers when registering. Pollster Amy Simon notes that proponents of using rolls of registered voters use the method under the hypothesis that "samples drawn from voter registration lists by definition consist of actual voters, while RDD studies rely entirely on respondents' self-reporting about whether they are in fact registered to vote."

Random digit dialing (RDD)

Random digit dialing (RDD) refers to when a polling organization uses a device to randomly generate phone numbers for participants and then asks if the participants are registered voters and are likely to vote in an upcoming election. According to FiveThirtyEight, firms that use RDD do so in an effort "to ensure that they wouldn’t be missing registered voters — especially in general elections — who decided not to make their phone numbers public information, or the newly registered, often young people."

Polling firms on Ballotpedia

The following organizations are examples of polling firms covered on Ballotpedia:

• Emerson College Polling
• Gallup
• JMC Analytics and Polling
• Marist Poll
• Marquette Law School Poll
• Monmouth University Polling Institute
• Public Policy Polling
• Quinnipiac University Poll
• Selzer & Company
• Suffolk University Poll

Footnotes