Polling Research Paper

Academicians and practitioners engaged in opinion polling have preferred to call themselves ‘survey researchers,’ in contrast to ‘pollsters,’ who carry out nonacademic polls to predict election results, or report the mass public’s or others’ opinions and attitudes in the news media, or who work as consultants doing proprietary research, such as market research, for political candidates, political parties (including polls for their presidents), or other clients (Converse 1987, Moore 1992, Jacobs and Shapiro 1995). Polls and surveys are synonymous, but their quality (validity and reliability) can vary widely.

1. ‘Straw Polls’

The use of polling had expanded greatly worldwide by the late twentieth century. Its origin and impetus occurred in the United States beginning with election ‘straw polls’ in the nineteenth century, and developing into modern-day political polling and academic surveys used widely to study social, economic, and political phenomena. Polling was arguably a major part of the rise of quantification in society consistent with Max Weber’s writing about the role of calculation and rationality in social life (Herbst1993). The earliest straw poll counterparts of modern opinion surveys occurred in the 1824 presidential election, and the use of different types of these in politics and journalism increased until the emergence of more consciously scientific polling in the 1930s (Smith 1990).

These polls took many different forms, including oral counts, hand-raising, paper ballots, and writing preferences in ‘poll books.’ The participants were ‘self-selected’ in ways that were likely to produce a problem known as ‘selection bias,’ in which the results of these polls would be influenced by the characteristics of those who were able to participate, which might be different from the electorate or population to which the polls’ results would be generalized. Reporters and partisans nonetheless treated these polls as indications of candidates’ election chances.

2. ‘Scientific Polls’

The problems associated with the overall representativeness and accuracy of such polls became clear publicly, when the accuracy of the Literary Digest’s paper ballot poll, which had predicted the winners in US presidential elections from 1920 to 1932, was called into question by George Gallup and other new and more scientific-minded pollsters, statisticians, and social scientists. Gallup was the most widely known, if not the most important figure (along with Elmo Roper, Archibald Crossley, Paul F. Lazarsfeld, and Hadley Cantril) in ‘scientific polling,’ and approached polling with a democratic philosophy (Gallup and Rae 1940). While the first national Gallup election poll, done through what became standard in-person interviews did not predict the margin of President Franklin Roosevelt’s 1936 election victory over Alfred Landon accurately, it triumphed over the Literary Digest in predicting the winner.

The Gallup poll represented the actual electorate far better than did the Digest, confirming the superiority of scientific sampling methods. In principle, initially, these methods allowed each potentially eligible adult voter (or eligible respondent, given the research objectives of a particular poll) an equal probability of selection, which was not the case in the Digest or other straw polls that allowed self-selection (Squire 1988). The early scientific polls involved sending interviewers to randomly-selected locations and then allowed interviewers largely to select their respondents to fill ‘quotas’ based on population characteristics to be represented in the poll. This led to selection biases.

These and other problems in the early polls, including ceasing to interview weeks before an election, were debated widely among pollsters and social scientists when the polls predicted incorrectly that Thomas Dewey would defeat President Harry Truman in the 1948 presidential election (Mosteller et al. 1949). Gallup and other pollsters improved their sampling methods, and the pollsters and academic researchers, who were beginning to engage in large-scale public opinion and voting studies at this time, engaged in a concerted and continuing effort to improve and defend their research (see, e.g., Sheatsley and Mitofsky 1992, Mitofsky 1998).

3. Academic Research

University-based survey research had its beginnings in 1939–40 with sociologist Paul F. Lazarsfeld’s Office of Radio Research, later renamed the Bureau of Applied Social Research. This history is described by Rossi (1959), Natchez (1985), and especially Converse (1987). Lazarsfeld and his colleagues in The People’s Choice (Lazarsfeld et al. 1944) did the first sophisticated survey study of presidential elections in 1940 in Erie County, Ohio. This was followed by Voting (Berelson et al. 1954), a study of the 1948 election in Elmira, New York. Two of the largest and most widely known research centers that developed the capacity to do long-term national surveys are the National Opinion Research Center (NORC), which was established in 1941 at the University of Denver and moved to the University of Chicago in 1947, and the University of Michigan’s Survey Research Center (SRC, Institute for Social Research (ISR)), established in 1946. With the publication of The Voter Decides (Campbell et al. 1954), the funding and locus for national election surveys shifted to Michigan’s SRC, which led to the publication of The American Voter (Campbell et al. 1960) and the continuing series of American National Election Studies and studies based upon them (Nie et al. 1976 1979, Miller and Shanks 1996).

4. Further Expansion

Political polling and market and academic survey research increased after 1948. The early and subsequent public pollsters who provided their survey results to the mass media and other subscribers (for example, Louis Harris and Daniel Yankelovich, among others) had used their political polling as advertising to generate business for their proprietary market and other types of research. Market researchers and other pollsters expanded their work into polling and consulting for political parties and candidates (Moore 1992). Polling experienced its greatest expansion beginning in the 1970s, when it was found that sufficiently representative and accurate polling could be done by telephone (based on experiments comparing polls conducted by phone with those done in-person), eventually leading all the major pollsters to do most of their polling by phone. The most important national news media, which had relied exclusively on polling by public pollsters, began sponsoring and conducting their own polls, establishing their own, often joint, polling operations or making arrangements to contract-out polls in which they controlled the content and reporting of results (Gollin 1980, Mann and Orren 1992).

5. Methodology And Technology

While social scientific polling depends on an understanding and use of scientific sampling to draw inferences about a target population (e.g., national or local, the entire adult population or particular subpopulations, ordinary citizens or elites) polls can be conducted in different ways depending on the research goal, the kind of survey questionnaire to be administered, and the available resources (money and time). In-person national polling (typically randomly sampling geographic locations, then households in these locations, and then random sampling within households) is the most expensive and time-consuming because of travel costs, though these can involve longer surveys and the use of visual aids and secret ballots for responses to sensitive questions. In contrast, election ‘exit polls’ that are used to project election results and provide the basis for more indepth statistical analysis of voters’ behavior, are short and efficient surveys, based on random samples of voters exiting voting places. Exit polling has occasionally been controversial due to the ‘early’ reporting of results that might affect the behavior of those who vote later in the day.

Telephone polling (typically using Random Digit Dialing, RDD) is less costly, as it eliminates travel costs and is easier to monitor and validate than in-person interviews; however, it excludes households without telephones, and it has higher rates of nonresponse due to households that are not reached (no answers, busy signals, answering machines, and refusals to participate); it is easier to make contact with respondents and establish rapport through in-person contact. Mail surveys are the least expensive but produce the lowest response rates unless extensive follow-ups are made (Dillman 2000). Low response rates, with their increasing potential for selection or non-response bias, has emerged as the most pressing professional and methodological problem currently facing survey researchers as the result of the proliferation of telephone surveys, telemarketing, and solicitations that interrupt and impose upon people’s time (Groves and Couper 1998).

Problems of survey costs and response rates have led surveyors to turn to better technology. The use of computerized RDD and computer-assisted telephone interviewing (CATI) was a major development in improving surveys (making complex surveys, including experimental designs, easier to administer, and eliminating interviewer and data processing errors) and expanding their use. The use of computer-assisted personal interviewing (CAPI) followed, including the use computers for self-administered surveys allowing for greater assurance of confidentiality for sensitive questions. Recently, there have been increasing efforts to expand sampling and polling through the Internet. While the Internet provides a way of increasing response rates through ease of contact without disruption, for potential respondents whose electronic addresses have been obtained, and through ease in administering surveys, the samples obtained may not be sufficiently representative of the target population, including those without Internet access, or who refuse to make their electronic addresses available for polling (thus echoing the limitation of the straw polls).

6. Sources Of Error

Beyond statistical sampling error and nonresponse bias, there are other sources of error in surveys that are not easily quantified. The responses to survey questions and measurement of opinions and behavior can be affected by how questions are worded, what response categories are offered (e.g., whether a middle category or a don’t know or no opinion response is offered or allowed), and whether questions should have fixed or forced choices, or whether questions should be asked and responses recorded as open-ended questions in which respondents can offer any answer they choose. Further, there may be ‘context effects’ that are produced by the order in which questions are asked, so that the way respondents answer questions may be influenced by what they had thought about and responded to in previous questions (Schuman and Presser 1981, Asher 1998).

A major source of error can occur depending on how research problems are formulated or specified, as when researchers assume respondents have familiarity with the object about which they are asked to offer an opinion or response. Care needs to be exercised in drawing inferences about actual behavior from survey measures of opinion, and self-reports of future—or even past—behavior. Reported voting in an election after the fact may not match actual behavior based upon data on ‘validated’ votes. Difficulties may also arise in how poll results are reported, such as when journalists or researchers themselves do not report enough information about their survey data and methods to allow their audiences or readers to evaluate their results (Cantril 1991, Traugott and Lavrakis 2000)

7. Contributions To Knowledge

The uses of surveys, how to improve polling methods, and how survey researchers should practice and report about their craft are matters that have been the concern of the professional associations and journals established by those actively involved in the field of survey research. These include the American Association for Public Opinion Research and its journal, Public Opinion Quarterly; the World Association for Public Opinion Research and its journal, International Journal of Public Opinion Research; the Council of American Survey Research Organizations; and the National Council on Public Polls. In addition, political science, sociology, and other social science journals also publish survey studies widely.

The early voting and public opinion studies by Lazarsfeld and others were highly sensitive to methodological issues in survey research, and this led to the development and use of panel surveys (the interviewing of the same respondents over two or more points in time) and complex research designs. In addition to contributing to a large body of knowledge about the influences on voting, survey research contributed from its inception to the understanding of other forms of social and political attitudes and behavior, ranging from Samuel Stouffer’s studies of The American Soldier (Stouffer et al. 1965 [1949]) and attitudes toward communism (Stouffer 1955), to studies of long-term trends in public opinion which utilize all the available polling data since the 1930s (Page and Shapiro 1992, Mayer 1993).

While academic researchers have made substantial use of data collected by academic survey organizations (especially the NORC General Social Surveys conducted since 1972, and the American National Election Studies covering every presidential and congressional election since 1952), increasingly they have used data made available by the public and mass media pollsters. The major archives for these data are the Roper Center for Public Opinion Research (University of Connecticut), the Inter-university Consortium for Political and Social Research (ISR, University of Michigan), and the Louis Harris survey and state poll archive at the Institute for Research in Social Science (University of North Carolina). These archives and the many polling organizations nationwide have also established and expanded access to their survey data through websites on the Internet.

Bibliography:

1. Asher H 1998 Polling and the Public: What Every Citizen Should Know, 4th edn. Congressional Quarterly, Washington, DC
2. Berelson B R, Lazarsfeld P F, McPhee W N 1954 Voting: A Study of Opinion Formation in a Presidential Campaign. University of Chicago Press, Chicago, IL
3. Campbell A, Converse P E, Miller W E, Stokes D E 1960 The American Voter. University of Chicago Press, Chicago, IL
4. Campbell A, Gurin G, Miller W E 1954 The Voter Decides. Row, Peterson, White Plains, NY
5. Cantril A H 1991 The Opinion Connection: Polling, Politics, and the Press. Congressional Quarterly, Washington, DC
6. Converse J M 1987 Survey Research in the United States: Roots and Emergence, 1890–1960. University of California Press, Berkeley, CA
7. Dillman D 2000 Mail and Internet Surveys: The Tailored Design Method, 2nd edn. Wiley, New York
8. Gallup G, Rae S F 1940 The Pulse of Democracy. Simon & Schuster, New York
9. Gollin A 1980 Exploiting the liaison between polling and the press. Public Opinion Quarterly 44: 445–61
10. Groves R M, Couper M P 1998 Nonresponse in Household Interview Surveys. Wiley, New York
11. Herbst S 1993 Numbered Voices: How Opinion Polling has Shaped American Politics. University of Chicago Press, Chicago, IL
12. Jacobs L R, Shapiro R Y 1995 The rise of presidential polling: The Nixon White House in historical perspective. Public Opinion Quarterly 59: 163–95
13. Lazarsfeld P F, Berelson B, Gaudet H 1944 The People’s Choice: How the Voter Makes Up His Mind in a Presidential Campaign. Columbia University Press, New York
14. Mann T, Orren G (eds.) 1992 Media Polls in American Politics. Brookings Institution, Washington, DC
15. Mayer W G 1993 The Changing American Mind: How and Why American Public Opinion Changed Between 1960 and 1988. University of Michigan Press, Ann Arbor, MI
16. Miller W E, Shanks J M 1996 The New American Voter. Harvard University Press, Cambridge, MA
17. Mitofsky W J 1998 Was 1996 a worse year for polls than 1948? Public Opinion Quarterly 62: 230–49
18. Moore D W 1992 The Superpollsters: How They Measure and Manipulate Public Opinion in America. Four Walls Eight Windows, New York
19. Mosteller F, Hyman H, McCarthy P J, Marks, E S, Truman D B 1949 The Pre-election Polls of 1948. Social Science Research Council, New York
20. Natchez P B 1985 Images of Voting: Visions of Democracy. Basic Books, New York
21. Nie N H, Verba S, Petrocik J R 1976/1979 The Changing American Voter (enlarged edn.). Harvard University Press, Cambridge, MA
22. Page B I, Shapiro R Y 1992 The Rational Public: Fifty Years of Trends in Americans’ Policy Preferences. University of Chicago Press, Chicago, IL
23. Rossi P H 1959 Four landmarks in voting research. In: Burdick E, Brodbeck A J (eds.) American Voting Beha ior. The Free Press, Glencoe, IL
24. Schuman H, Presser S 1981 Questions and Answers in Attitude Surveys: Experiments in Question Form, Wording, and Context. Academic Press, New York
25. Sheatsley P B, Mitofsky W J (eds.) 1992 A Meeting Place: The History of The American Association for Public Opinion Research. American Association for Public Opinion Research, Ann Arbor, MI
26. Smith T W 1990 The first straw: A study of the origins of election polls. Public Opinion Quarterly 54: 21–36
27. Squire P 1988 Why the 1936 Literary Digest poll failed. Public Opinion Quarterly 52: 125–33
28. Stouffer S A 1955 Communism, Conformity, and Civil Liberties: A Cross-section of the Nation Speaks its Mind. Doubleday, Garden City, NY
29. Stouffer S A, Suchman E A, DeVinney L C, Star S A, Williams R M Jr 1965 [1949] The American Soldier: Adjustments During Army Life. Wiley, New York, Vol. 1
30. Traugott M W, Lavrakas P J 2000 The Voter’s Guide to Election Polls, 2nd edn. Chatham House, New York