The Associated Press-GfK poll on undecided and persuadable voters was conducted by GfK Roper Public Affairs & Corporate Communications from Oct. 19-23. It was based on landline telephone and cellphone interviews with a nationally representative random sample of 1,186 adults, including 1,041 registered voters and 839 likely voters. Interviews were conducted with 713 respondents on landline telephones and 473 on cellular phones.
One hundred thirty-one likely voters were persuadable: They did not initially choose a candidate when asked or did initially choose a candidate but said they might change their minds.
Digits in the phone numbers dialed were generated randomly to reach households with unlisted and listed landline and cellphone numbers.
Interviews were conducted in both English and Spanish.
As is done routinely in surveys, results were weighted, or adjusted, to ensure that responses accurately reflected the population's makeup by factors such as age, sex, education and race. In addition, the weighting took into account patterns of phone use — landline only, cell only and both types — by region.
No more than one time in 20 should chance variations in the sample cause the results to vary by more than plus or minus 3.5 percentage points from the answers that would be obtained if all adults in the U.S. were polled. The margin of sampling error was plus or minus 3.8 percentage points for registered voters, plus or minus 4.2 percentage points for likely voters and plus or minus 10.6 percentage points for persuadable voters.
There are other sources of potential error in polls, including the wording and order of questions.
The questions and results are available at http://www.ap-gfkpoll.com .