Pollsters collect information for a variety of clients—for example, politicians, media outlets, and interest groups. Often that information is collected and distributed publicly, such as through a news media organization. Other times, the information is used for private purposes, such as to plan campaign strategy based on voter preferences and behaviors.
Politicians and the political parties are among the most prominent users of private polls. In the years and months leading up to an election, candidate and their teams commission polls to find out how voters feel about the candidate and the campaign platform. Respondents' data tell candidates everything from how registered voters view candidates' appearance and personality to what voters think about key issues such as immigration, education, crime, and taxes. Candidates can then use the feedback to adjust their campaign focus or tailor their message to be most effective.
Sometimes candidates use focus groups for private polling. A focus group is a small group of people who represent a chosen subset of the population who discuss their beliefs and preferences in a moderated setting. A moderator trained by the polling organization manages the event. He or she asks open-ended questions, for example on the personal and professional attributes of the candidates and on a variety of issues that might be important during the campaign. Focus group members give candidates more in-depth information about these matters than can be obtained from a poll.
As an election approaches, candidates or their supporters might fund polling in another way: to persuade voters to vote for them and to dissuade voters from voting for the other candidates in the race. A push poll is an opinion poll that includes questions or statements aimed at "pushing" respondents away from the candidate or issue that the group paying for the poll opposes. For example, a respondent who indicates that she will vote for Candidate A might then hear five negative statements about that candidate before being asked the next question about the election. Public polling organizations do not use push polls. They are a tool of campaign efforts or groups that support a candidate.
Polls are a snapshot of public opinion on the particular dates when the poll was conducted. Polling organizations and public officials use tracking polls to see changes over time. A tracking poll is an opinion poll that surveys respondents on their beliefs and preferences over a period of time. Tracking polls are usually conducted by independent polling organizations. Pollsters use this data to track how people's reactions to the same politician or issue changes over time, with results indicating whether or not public opinion is affected by events. For example, Gallup has tracking poll results going back to 1980 on the question of whether respondents are satisfied with "the way things are going" in the United States. Political scientists correlate the results of tracking polls to gauge shifts in the public's thinking. In election years, nonpartisan tracking polls that measure people's feelings toward candidates can have an effect on candidates' ability to raise money, since potential donors are influenced by poll findings. Politicians may use tracking polls as they consider their position on issues.
A job-approval rating is a measure of how well the public thinks a politician is performing. Job approval is of great interest to every politician. The results of job-approval polls are regularly published by the media and can vary widely, even on a quarter-to-quarter basis. Current events and the president's public statements can produce these variations. For example, when the Watergate scandal began to make headlines in 1973, President Richard Nixon's approval rating dropped from 61 percent in the first quarter of the year to 44 percent in the second quarter. By the time Nixon resigned on August 9, 1974, his approval stood at only 24 percent. During the Persian Gulf War in 1991, President George H.W. Bush had an extremely high job-approval rating of 89 percent. By the summer of 1992, that rating had dropped to 29 percent due to economic troubles. It rebounded to 56 percent shortly before he left office in January 1993.
Besides tracking and job-approval polls, there are many other kinds of public polls, conducted by many different organizations. One major category is the issue poll, an opinion poll designed to measure the public's feelings on specific issues. For example, the Pew Research Center, a nonpartisan think tank, conducts a variety of issue polls in hopes of informing people of trends and changing beliefs. Major Pew polls have asked respondents to weigh in on the government's role in responding to national disasters, the importance of organized religion in their daily life, and the value they place on Internet freedom. The results of issue polls like the Pew surveys are widely known to politicians.
Interest groups also sometimes fund or conduct polls, usually to promote their own agenda within Congress. These interest groups may represent a certain segment of the population, such as retired persons or gun owners, or a certain industry, such as the oil or automobile industry. Many interest group polls are informal, or straw, polls. Those represented by the groups are contacted and encouraged to express their opinion on an issue currently before Congress, such as elder care or a new gas tax, by answering some questions. Often, respondents can take the survey more than one time, and little attention is paid to finding a random, representative sample of the population. The interest group behind the poll then releases results to media outlets, knowing that publication or broadcast of the poll will make it seem more legitimate than it is.
When surveying people, pollsters often ask respondents for identifying and personal information to gather additional categories of data. Age, gender, race, and ethnicity are commonly collected demographics. Additional demographics that may be collected include educational background, income, type of residential area (for example, urban, suburban, or rural), party affiliation, religion, and marital status. Analysts and political scientists examine how demographics inform and correlate public opinion. They may find that certain groups are more receptive to a candidate's message while others are less responsive. A candidate may take steps to reach out to the groups less favorably inclined in an effort to expand his or her appeal.