frequently asked questions
What is the National Election Pool (NEP)?
The National Election Pool is a consortium of news networks that was formed in 2003 in order to provide information on election night about the vote count, election analysis, and election projections. NEP contracts with Edison Research to make projections and provide exit poll analysis and a tabulation of the vote throughout the country. The National Election Pool is currently comprised of ABC News, CBS News, CNN, and NBC News.
What does Edison Research provide to the NEP?
Edison Research conducts statewide exit polls throughout the country as well as a national survey. In addition, Edison collects the vote count in a sample of precincts in each state. Edison provides exit poll analysis and projections for President, U.S. Senate, Governor as well as selected U.S. House, state referenda and initiatives.
What is an exit poll?
Exit polls are interviews with voters after they have cast their votes at their polling places. A sample of precincts is scientifically selected to collectively represent a state or, for the national exit poll, the nation. An interviewer gives every nth voter exiting the polling place a questionnaire to complete. There are questions about demographics such as gender, age, race and issues related to the person’s vote choice in different contests. Participation is voluntary and anonymous. The interviewing starts when the polls open and continues throughout the day until about an hour before the polls close.
Who decides what questions to ask on the exit poll?
All questions asked on state and national questionnaires are prepared by the four members of the National Election Pool.
How many polling places will be in your sample?
Sample sizes vary from state to state.
How do you select sample precincts?
Precincts are selected as a stratified probability sample of each state. The purpose of stratification is to group together precincts with similar vote characteristics. A recent past election is used to identify all precincts as they existed for that election. The total vote in each precinct and the partisan division of the vote from this past race are used for the stratification. In addition, counties are used for stratifying the precincts. The total vote also is used to determine the probability of selection. Each voter in a state has approximately the same chance of being selected in the sample.
Who makes projections of the winning candidates?
Projections are made by Edison and transmitted to each of the NEP members and subscribing news organizations. Each of the members has its own analysts who review the Exit Poll results and the tabulated data as it is collected. Each news organization makes its own decision about what to report to the public. All decisions are made after careful review by the analysts and are not automatic decisions made by a computer.
How are projections of winners made?
Projections of a winning candidate are based on models that use votes from three different sources:
- Exit poll interviews with voters
- Vote returns as reported by election officials from the sample precincts
- Tabulations of votes by county
The models make estimates from all these vote reports. The models also indicate the likely error in the estimates. The best model estimate may be used to make a projection if it passes a series of tests.
When will projections be made?
Projections of a winning candidate are only made after all the polls in a state are closed and when the best model estimates show a clear winner. There will be no projections before the last polls in a state are closed.
How do absentee votes affect projections?
In a number of states, significant numbers of voters cast their ballots before Election Day. Either they cast an absentee ballot or they vote at voting locations set up for voters in the weeks leading up to the election. A sample of absentee/early voters is interviewed by telephone shortly before the election. Edison combines the results of these absentee/early voters’ interviews with the Election Day interviews in the survey results reported by the NEP.
What is the margin of error for an exit poll?
Exit polls are surveys. As in all surveys, there is a margin of error due to sampling. The margin of error for a 95% confidence interval is about +/- 3% for a typical characteristic from the national exit poll and +/-4% for a typical state exit poll. Characteristics that are more concentrated in a few polling places, such as race, have larger sampling errors. Other non-sampling factors may increase the total error.
How many people does it take to conduct the Exit Polls and collect the vote at the Edison Research sample precincts?
For a national election, almost 3,000 people are required to do the job at Edison. This includes exit poll interviewers, telephone operators to take their calls at election headquarters, reporters at the sample precincts to get the vote, developers to program the computers, systems specialists, election researchers, technical support at all sites, analysts reviewing the computations, support staff to manage the exit polls and management of the project.
How can I have confidence in a race that you call?
Since Edison Research began conducting exit polls for the National Election Pool in 2003, the news organizations using the Edison data have not made a single mistake in deciding a winner in any race. We use a high standard of care in making sure that the information and analysis that we provide news organizations is accurate. All surveys, including exit polls have a sampling error and Edison Research makes certain to take that into account before advising news organizations of the winner in any race.
