WHY THE 1936 LITERARY DIGEST POLL FAILED

The Literary Digest poll of 1936 holds an infamous place in the history of survey research. Despite its importance, no empirical research has been conducted to determine why the poll failed. Using data from a 1937 Gallup survey which asked about participation in the Literary Digest poll I conclude that the maga- zine's sample and the response were both biased and jointly pro- duced the wildly incorrect estimate of the vote. But, if all of those who were polled had responded, the magazine would have, at least, correctly predicted Roosevelt the winner. The current rele- vance of these findings is discussed. The 1936 campaign concluded with the Literary Digest (1936a) publish- ing survey results forecasting a landslide victory for the Republican presidential candidate, Alf Landon. The actual election was, of course, won by the incumbent, Franklin Roosevelt, by a large margin. Thus the Literary Digest poll gained an infamous place in the history of survey research. Almost every book on presidential elections or survey methodology contains some scathing reference to the poll and gives reasons why it failed to forecast the correct results. Some claim the error resulted from a biased sample. A few assert that the sample was acceptable but that the low response rate produced the incorrect forecast. Many others state that a combination of these problems was responsible. Surprisingly, these claims are mere speculation; no analysis has been conducted to determine why the Literary Digest poll was wrong. Con- sequently, we have some ideas-really competing hypotheses-as to why the poll failed but no empirical research by which to determine the source of the error.