Results for this survey are based on telephone interviews conducted under the direction of Princeton Survey Research Associates International among a nationwide sample of 2,006 adults, 18 years of age or older, from October 17-22, 2006. For results based on the total sample, one can say with 95% confidence that the error attributable to sampling is plus or minus 2.5 percentage points. For results based on registered voters (N=1,552), the sampling error is plus or minus 3 percentage points. For results based on Form 1 (N=1,003) and Form 2 (N=1,003) the sampling error is plus or minus 3.5 percentage points. For results based on likely voters (N=1,118) the sampling error is plus or minus 4 percentage points.

In addition to sampling error, one should bear in mind that question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.

This survey includes an oversample of 515 respondents in congressional districts with competitive U.S. House races in the 2006 election for a total of 682 respondents who reside in such districts. Competitive districts were identified using rankings by Congressional Quarterly, The Cook Political Report, The Rothenberg Political Report, The New York Times, and Larry Sabato’s Crystal Ball in early to mid-October. This yielded a list of forty competitive districts: Arizona-05, Arizona-08, California-11, Colorado-04, Colorado-07, Connecticut-02, Connecticut-04, Connecticut-05, Florida-13, Florida-22, Georgia-08, Georgia-12, Iowa-01, Iowa-03, Illinois-06, Illinois-08, Indiana-02, Indiana-08, Indiana-09, Kentucky-04, Minnesota-06, North Carolina-11, New Jersey-07, New Mexico-01, Nevada-02, New York-20, New York-24, New York-29, Ohio-01, Ohio-15, Ohio-18, Pennsylvania-06, Pennsylvania-07, Pennsylvania-08, Pennsylvania-10, Texas-17, Virginia-02, Vermont at-large, Washington-08, and Wisconsin-08.

In addition, nine states with competitive Senate races (N=490) and twelve states with competitive governor’s races (N=371) were identified using The Cook Political Report and The Rothenberg Political Report rankings. Senate races in Connecticut, Missouri, Montana, New Jersey, Ohio, Pennsylvania, Rhode Island, Tennessee, and Virginia and governor’s races in Arkansas, Colorado, Iowa, Maine, Maryland, Massachusetts, Michigan, Minnesota, Nevada, Oregon, Rhode Island, and Wisconsin are considered competitive.

This survey was conducted in association with the Brookings Institution and the Cato Institute. Michael McDonald of the Brookings Institution and George Mason University, and John Samples of the Cato Institute provided consultation on the study design. Professor Gary Jacobson of the University of California, San Diego generously provided data on voting patterns in congressional districts.