How AP-Yahoo poll was conducted

The AP-Yahoo News Poll is a unique study that tracks the moods and opinions of a group of people throughout the presidential election campaign. The seventh wave of this study included re-interviews of 1,740 people between Sept. 5-15 who were originally interviewed in November or early December. It included interviews with 502 people who in AP-Yahoo News polls in January and April identified themselves as supporting Clinton in one or both of those months. The extensive national survey was conducted by Knowledge Networks of Menlo Park, Calif., under the direction and supervision of AP's polling unit.

The interviews were conducted online. The original sample was drawn from a panel of respondents Knowledge Networks recruited via random sampling of landline telephone households with listed and unlisted numbers. The company provides Web access to panel recruits who don't already have it. With a probability basis and coverage of people who otherwise couldn't access the Internet, the Knowledge Networks online surveys are nationally representative.

Results were weighted, or adjusted, to reflect the adult population by demographic factors such as age, sex, region, race, and education.

No more than one time in 20 should chance variations in the sample cause results to vary more than plus or minus 2.3 percentage points from the answers that would have been obtained if all adults in the U.S. were surveyed. The sampling error margin is plus or minus 3.4 percentage points for the subsample of 825 Democrats and those who lean toward the Democratic Party and 3.7 percentage points for the subsample of 715 Republicans and those who lean toward the Republican Party. The sampling error margin for the former Clinton supporters is plus or minus 4.4 percentage points.

There are other, potentially greater, sources of variability in surveys, including the wording and order of the questions.

The questions and results for this poll will be available at http://news.yahoo.com/polls and at http://surveys.ap.org.