Results for the Late October 2004 Voter Callback Surveys are based on telephone interviews conducted under the direction of Princeton Survey Research Associates International among a sample of 519 adults, 18 years of age or older, from October 21-25, 2004. The interviews were conducted among a population of 973 registered voters who were undecided in the presidential election, only leaned toward a candidate, or said they might change their vote choice (“Swing Voters”). The Swing Voters were originally contacted in three Pew Research Center surveys in September 2004 (Sept. 9-14, Sept. 17-21, and Sept. 22-26). For results based on the total sample, one can say with 95% confidence that the error attributable to sampling is plus or minus 5 percentage points. For results based on Swing Voters who have decided on a candidate (N=269) or those who are still uncommitted (N=250), the sampling error is plus or minus 7 percentage points.

The survey also re-interviewed 482 registered voters from the September surveys who said they were certain of their vote choice (241 for Bush, 241 for Kerry). Sampling error for the certain Bush or the certain Kerry voters is plus or minus 7 percentage points. For all 482 certain voters, sampling error would be plus or minus 5 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.