Public Opinion Poll

The first public opinion research goes back to July 24, 1824, straw vote. The Harrisburg Pennsylvanian issued a report of a straw vote done at Wilmington, Del., "without discrimination of parties." In that poll, Andrew Jackson was a winner receiving 335 votes followed by John Quincy Adams, 169; Henry Clay, 19, and William H. Crawford, 9.

Dr. Gallup laboriously collected voting record of United States since 1836 and, late in 1933, he sent out voting forms to a small, but scientifically selected, group of voters in each state. On the basis of his returns, he calculated the expected results of the 1934 congressional elections. When the official results were announced, they matched his predictions within one percent. He continued his experiment until 1935 when he founded the American Institute of Public Opinion and predicted Roosevelt's victory in 1936 presidential election.

Of all the newspaper and magazine polls, those conducted by The Literary Digest received the greatest public attention and acclaim. The Literary Digest went on to mail out ballot cards to residential telephone subscribers and to lists of automobile owners for the 1936 election. The results pointed to a Landon victory over Roosevelt, indicating that he would get some 54% of the total vote, or 57% of the major-party vote. However, Roosevelt polled 62.5% of the major-party vote and won 523 out of possible 531 electoral votes.
The disaster of The Literary Digest poll resulted from its sampling method which selected only wealthy people as a sample group.

In 1935, Gallup had predicted the outcome of the 1936 presidential election as a victory of Roosevelt over Landon with an error of 6.8 percent.
Thereafter Gallup gained fame and his scientific method became prevailent in polling.

However, Gallup made an error in predicting the 1948 presidential election. This was his own great disaster: the prediction that Thomas Dewey would defeat Harry Truman by anywhere from five to fifteen percentage points in 1948. When Truman actually won by more than four percentage points, Gallup was under attack.

In 1949, the Social Science Research Council (SSRC) conducted investigations of the possible errors in the polling process. The SSRC committee announced two major causes of error in 1948: (1) errors of sampling (quota sampling) and interviewing and (2) errors involving the pollsters' failure to assess the future behavior of undecided votes and to detect shifts near the end of the campaign.

Gallup himself identified two causes of error: the failure to conduct a last-minute poll and misinterpretation of undecided voters.
Gallup thus turned 1948 into a useful learning experience and, in effect, an argument for more extensive polling. In the 1956 election, Gallup abandoned "quota" sampling, where a representative percentage of people was surveyed, and switched to a new method, probability sampling where everyone in population being surveyed has an equal chance to be sampled (Vivian, 1997).

figure1.jpg

Today, the Gallup Poll Monthly provides a "Gallup Poll Accuracy Record" that starts with 1936 and calculates the deviation for every election.
The American Institute of Public Opinion's surveys have become a distinctive feature of American journalism and results appeared in 120 leading U.S. daily newspapers. The purpose of the Institute was to measure public opinion on current political, social, and economic issues and to report the facts for the benefit of all (AIPO, 1957).

The American Institute of Public Opinion developed rapidly into a research organization with a reputation for reliability and was conducting surveys for academic social researchers, private industry, new media, and political groups. The institute's major asset was the capacity for conducting representative national surveys with skillful interviewers and analyzing the data in a short period of time. After 1958, Gallup's diverse polling operations were grouped under the Gallup Organization Inc.

Currently there are two different organizations under the name of Gallup. Gallup Organization Inc., which was sold by the Gallup family in 1988 to the Selection Research Institute in Lincoln, Nebraska, has wholly or majority-owned subsidiaries in more than 20 countries. More than 3,000 research, consulting, and teaching professionals work together to provide clients with comparable practices, procedures, and standards across national, cultural, and linguistic boundaries. An employee-owned firm, Gallup's revenues have grown by an average of 25% annually over the past decade.

The other one, the George H. Gallup International Institute, is a public charity whose purpose is to discover, test and encourage the application of new approaches to social problems--in education, environment, health, religion and human values. When promising approaches are proposed, the institute asks the public to decide which ones to accept. This process allows the public to become a positive force for energizing and uplifting society (Long, 1994).