Holcombe Predicts Roosevelt Easy Victor in Democratic Landslide--Doubts Validity of Digest or Any Other Poll

If the election were held tomorrow, the Democratic candidate for president would win by an overwhelming majority," said A. N. Holcombe '06, professor of Government, in an interview yesterday. "But many things may happen in the course of a month to change the outcome. Certainly if there should be a marked revival of business there would be a considerable change of opinion among the voters.

"The current figures in the Literary Digest poll show a very large majority for Roosevelt. This is even more significant in view of the fact that most of the votes so far returned are from those who could ordinarily be expected to vote Republican. They have come more from the business and professional class than from wage earners, inasmuch as ballots are mailed only to those who have automobiles or telephones registered under their names, Consequently a Democratic landslide is in prospect.

"The accuracy of the Digest's straw vote is not beyond question, however. While the poll is conducted with scrupulous honesty and care, there are several factors which provide a possibility of error. Most of the votes are cast a month before the election, and many people are apt to change their minds. A candidate may do something foolish, or an unexpected event may occur, or Trend be evidenced, which will change the whole course of public opinion. The possibility of error is not large enough to be of importance except in a close election, however, and the forthcoming election does not appear to be close.

"Of the three types of straw votes used to forecast presidential elections in the past, the most accurate results were obtained from the interview method used by the Hearst papers in 1928, which presaged the results almost exactly. This method, however, offers possibilities of grave abuses and has in the past been distrusted by the public. The common type of newspaper poll, in which ballots are printed in the paper, is absolutely unreliable. That used by the Literary Digest, which mails about twenty million ballots and receives answers from about 23 per cent, is the most comprehensive and attracts the most interest, but has made some very serious errors in the past. Most notable of these was the prophecy that Davis and LaFollette would run about even in 1924, whereas Davis received about 29 per cent of the votes cast in the election and LaFollette only 16 per cent