Forecasting Election Results

Old-time reporters relied on their instincts and “seat-of-the-pants journalism” to help them identify the likely winners. They were assisted at times by their newspapers, which early in the nineteenth century began taking what came to be known as a straw vote before elections. The straw vote or poll began simply, with reporters asking train or steamship passengers about their [Page 213]candidate preferences. By the 1930s some straw votes had become elaborate affairs, involving extensive mailings of returnable ballots.

At about the same time an Agriculture Department analyst named Jerzy Heyman developed the systematic sampling techniques that are at the heart of modern polling. The Survey Research Center of the Inter-university Consortium for Political and Social Research at the University of Michigan is an outgrowth of the agriculture program. (See national election studies.)

Pioneer pollsters George Gallup, Elmo Roper, and Archibald Crossley soon adopted the sampling methodology and used it successfully to predict Franklin D. Roosevelt's landslide victory over Alfred M. Landon in 1936. Twelve years later the still-young polling industry received a black eye with its almost unanimous prediction that Thomas E. Dewey would unseat President Harry S. Truman in the 1948 election.

The setback was only temporary; pollsters learned from their mistakes, and the activity grew into a respected and valued profession. It would be another two decades, however, before the news media began taking full advantage of the professional pollsters' generally successful methods of forecasting election results. Washington Post political reporter David S. Broder recalls in his 1987 book Behind the Front Page that in 1960 he camped out for a week in Beckley, West Virginia, to try to sense public opinion in the crucial presidential primary between Democrats John F. Kennedy and Hubert H. Humphrey. Although Humphrey was expected to defeat Kennedy in the state, which had only a 5 percent Catholic population, Broder found considerable support for Kennedy, and he reported it in the paper he was then working for, the Washington Star.

...