Can Registration-Based Sampling Improve the Accuracy of Midterm Election Forecasts? |
| |
| Authors: | Green, Donald P. Gerber, Alan S. |
| |
| Affiliation: | DONALD P. GREEN and ALAN S. GERBER are professors of political science at Yale University. Earlier versions of this report were presented at the 2002 annual meeting of the American Association for Public Opinion Research, Nashville, TN, and the Gallup Conference on Improving the Accuracy of Polling, May 2–4, 2002, Washington, DC. |
| |
| Abstract: | We compare the predictive accuracy of preelection polls usingtwo types of sampling frames, random digit dialing (RDD) andregistration-based sampling (RBS). The latter involves stratifiedrandom sampling from voter registration lists. In order to assessthe accuracy with which RDD and RBS predict election outcomes,we collaborated with the Washington Post, Quinnipiac, and CBSNews polls, which conducted parallel RDD and RBS surveys inMaryland, New York, Pennsylvania, and South Dakota prior tothe November 5, 2002, elections. The results suggest that inthe gubernatorial and congressional elections studied, RBS performedas well, if not better, than RDD, both in terms of forecastingaccuracy and cost. |
| |
| Keywords: | |
| 本文献已被 Oxford 等数据库收录！ |
|
