The history of polling in the United States goes back to 1824, when two newspapers, the Harrisburg Pennsylvanian and the Raleigh Star, organized "show votes" to determine the political preferences of voters prior to the presidential election of that year. In 1883 the Boston Globe attempted to speed up its reporting of election returns by sending reporters to poll various precincts. By the turn of the century many newspapers were conducting polls to determine political preferences. Later polls were conducted by magazines; the first among them were the Farm Journal (1912) and the Literary Digest (1916). Those early polls were generally local or regional rather than national and were confined to obtaining election preferences rather than opinions on political issues. During World War I, however, a poll as to whether or not the United States should enter the war was conducted.
The methods used in the early polls made no claim to being scientific; polling was usually done by canvassers hired to go out and question people or by "straw ballots" in the newspapers, which readers were asked to fill out and mail in. A more scientific method of polling called sampling was developed in the mid-1930s. This method enables the polltaker to question a small percentage of the group whose opinions he wishes to ascertain and to analyze from their responses the opinions of the whole group. The superiority of this method over the old straw-ballot system was demonstrated in the 1936 presidential election when the Literary Digest poll, which based its predictions on the older technique, produced a staggeringly inaccurate forecast, while the poll of a newer group organized by George Gallup predicted the result of the election correctly. By the 1940s the polls were concerned with social and economic questions as well as with political issues. An unusual failure of polling took place in 1948 when the polling organizations predicted the defeat of Harry S. Truman, who won.
Sampling techniques have become increasingly sophisticated and include various types, which may be random, stratified, or purposive, or a combination of any of these. The information may be elicited by personal interview, telephone interview, or mail questionnaire, and the polling is completed only after the data have been tabulated and evaluated. Polling has been much used by politicians to determine the opinions of voters on significant issues. It has also been used to forecast patterns of voting.
Besides playing an increasingly important role in national and local political campaigns, the technique of modern polling has developed into one of the more important tools in the methodology of contemporary social science, particularly in sociology. Commercial polltakers claim that they not only provide valuable information in such fields as market research and advertising but that they also aid the process of democratic government by making known the views of the people. Critics of polling question the validity of the claim that it provides a true picture of public opinion, and it has been suggested that the polls themselves may influence public opinion by creating a "bandwagon effect."
Some of the pioneer commercial polling organizations were the Fortune survey (1936) conducted by Elmo Roper; the Crossley Poll (1936); and the Gallup Poll (1935). The Harris Survey, begun in 1956, together with the Gallup Poll, are the best-known polling organizations. Nonprofit polling organizations include the Princeton Office of Public Opinion Research (1940), the National Opinion Research Center (1941), and the National Council of Public Polls (1968). Many other countries have polling organizations, and a number of international societies (e.g., The European Society for Public Opinion and Market Research) facilitate exchanges of information.
See G. H. Gallup and S. F. Rae, The Pulse of Democracy (1940, repr. 1968); L. Bogart, Silent Politics (1972); C. W. Roll, Jr., and A. H. Cantril, Polls (1972); I. Crespi, Public Opinion Poll and Democracy (1989).