Survey and Sampling
Public opinion polls are a relatively new phenomenon. In 1948, as a result of telephone surveys of likely voters, all of the major organizations Gallup, Roper, and Crossley consistently predicted, throughout the summer and into the fall, that Thomas Dewey would defeat Harry Truman in the November presidential election. By October the results seemed so clear that Fortune magazine declared, “Due to the overwhelming evidence, Fortune and Mr. Roper plan no further detailed reports on change of opinion in the forthcoming presidential campaign…”
Of course, Harry Truman went on to win the 1948 election, and the picture of Truman in the early morning after the election holding up the Chicago Tribune (printed the night before), with its headline declaring Dewey the winner, has become legend. The public’s faith in opinion polls plummeted after the election, but Elmo Roper vigorously defended the pollsters. Roper was a principal and founder of one of the first market research firms, Cherington, Wood, and Roper, and director of the Fortune Survey, which was the first national poll to use scientific sampling techniques. He argued that rather than abandoning polling, business leaders should learn what had gone wrong in the 1948 polls so that market research could be improved. His frank admission of the mistakes made in those polls helped to restore confidence in polling as a business tool.
What did go wrong?!!!
The GFK Roper Report was an example of a telephonic sample survey designed to ask questions of a small group of people in the hope of learning something about the entire population and, most important, it represents the entire population. It was a time of 1948 and not everyone had telephones only people who can actually afford it. This means samples list made from telephonic directory was biased, representing only upper class people. It means opinion poll conducted was made on the answers of only rich families but what about citizens of other class. What a blunder!