The Chicago Sun-Times proudly boasts of the “38-year record of accuracy in predicting the outcome of Illinois elections” which its Straw Poll has maintained. In this article the methodology and the record of the 1968 and 1970 Sun-Times Straw Poll are examined in detail. For both years, a set of 80 predictions for Cook County, Illinois is analyzed, and an attempt is made to partition the the error into components, using an additive linear model and least squares. Straw poll results are compared with those produced by the method of persistence forecasting, and suggestions are made for improvement of the Sun-Times Straw Poll accuracy.
No takes yet. Share an insight, caveat, or question.
Stephen E. Fienberg (1972) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: