(CNN) -- To project an election, CNN and its election experts use scientific statistical procedures to make estimates of the final vote count in each race. Only after an extensive review of data from a number of sources ensures that those estimates are valid does CNN broadcast a projected winner.
CNN editorial policy strictly prohibits reporting winners or characterizing the outcome of a contest in any state before all the polls are scheduled to close in every precinct in that state.
For the 2004 Democratic primaries and caucuses, CNN will receive information from the following sources:
The Associated Press: The Associated Press will provide statewide vote totals for each race as well as vote totals in each county. The AP will be gathering numbers via stringers based in each county.
Edison/Mitofsky Research: To assist CNN in collecting and evaluating this information, CNN has employed Edison Media Research (EMR) and Mitofsky International. In previous elections, Warren Mitofsky and Joe Lenski have assisted CNN in projecting winners in state and national races. Edison/Mitofsky will conduct exit/entrance polls, which ask voters their opinion on a variety of relevant issues, determine how they voted, and ask a number of demographic questions to allow analysis of voting patterns by group.
Using exit poll results, vote results from the AP and a number of sophisticated analysis techniques, Edison/Mitofsky also project a winner for each race it covers. CNN will not use exit polls to make projections in close races. CNN will only use exit poll survey data to project a winner when that data indicates one of the candidates has a significant lead at the time that the polls close in that state.
The process of projecting races begins by creating a sample of precincts. The precincts are selected by random chance, like a lottery, and every precinct in the state has an equal chance to be in the sample. They are not bellwether precincts or "key" precincts. Each one does not mirror the vote in a state but the sample collectively does.
The first indication of the vote comes from the exit polls conducted by Edison/Mitofsky. On Primary/Caucus Day, Edison/Mitofsky interviewers stand outside of hundreds of sample precincts in a given state. For primaries, they count the people coming out after they have voted and are instructed to interview every third person or every fifth person, for example, throughout the voting day. The rate of selection depends on the number of voters expected at the polling place that day. They do this from the time the poll opens until shortly before it closes.
The interviewers give each selected voter a questionnaire, which takes only a minute or two to complete it. It asks about issues that are important, and background characteristics of the voter, and it also asks for whom they voted in the most important races. During the day, the interviewer phones the information from the questionnaires to a computer center.
Next, vote totals come in via the AP from the same sample precincts as the exit poll after the voting has finished in those precincts. These are actual votes that are available after the polls have closed and the precinct officials have counted the votes in that precinct. The results are posted so anyone at the precinct can know them.
The third set of vote returns come from the vote tallies done by county officials. The countywide figures become more complete as more precincts report vote returns. The county vote is put into statistical models, and Edison/Mitofsky make estimates and projections using those models.
CNN will decide when and how to make a projection for a race depending on how close the race is.
In races that do not appear to be very close, projections may be made at poll closing time based entirely on exit poll results, which are the only information available about how people voted when the polls close. The races projected from exit polls alone are races with comfortable margins between the top two candidates. Projections from exit polls will only be made if they are consistent with pre-election polls. The victory margins for an exit poll projection are typically eight percentage points or more. If the race is closer than that, CNN will wait for actual votes to be tabulated and reported. Edison/Mitofsky will make these projections and CNN will monitor them. If they don't meet the more rigorous criteria set up by CNN after the 2000 presidential election, CNN will withhold those projections.
The margin would typically be eights points or more for an exit poll projection, but projections depend on the margin of error and the size of the lead between the candidates. The margin of error calculation is part of the model. Edison/Mitofsky will make a projection when there is a smaller lead than CNN requires. That is one reason why Edison/Mitofsky may sometimes make a projection that CNN will not broadcast immediately.
Shortly after poll closing time, Edison/Mitofsky will try to make projections using models that combine exit polls and actual votes. This can sometimes be useful for closer races, perhaps as close as four percentage points. But after poll closings, CNN analysts will rely only on actual votes. CNN has its own samples for key races and will use them along with the Edison/Mitofsky estimates that are based only on actual votes. The projections again will be made when the margin of error is small enough for the projection to be safe. The odds of making a wrong projection will be less than one chance in 100.
For extremely close races, CNN will rely on actual votes collected at the county level. These are the races that cannot be projected when the polls close from exit polls or even from actual votes collected at the sample precincts mentioned earlier. The projection will be based on a statistical model that uses the county votes. If it is too close for this model to give us a reliable projection, CNN will wait for election officials to tally all or almost all the entire vote.
Finally, there are races that CNN will not project on primary or caucus night. CNN will not make a projection if it looks like the final winning margin is less than one percentage point. That is because the unofficial election night vote tallies in the past have on occasion had errors that are almost that large. Instead, CNN will make it clear that one candidate is ahead and that announcing a winner will depend on an official verification of the vote.
CNN Decision Desk analysts will review all projections from Edison/Mitofsky before they are accepted by CNN. CNN's team of analysts also will make its own projections. When CNN's analysts project a winner in a race, whether it is based upon data from Edison/Mitofsky or from the CNN computations, it means that when all the votes are counted, CNN projects that the candidate will win the election. The projection from the analysts is as close to a statistical certainty as seems necessary to avoid any mistakes. That does not mean that a mistake cannot happen, but every precaution has been taken to see that it does not. CNN will not "declare" someone a winner because that is up to election officials. We will make projections based on our best estimate of how CNN expects the election to turn out.
When a lot of vote returns have been tallied, a race may be referred as "too close to call" by CNN anchors and analysts. "Too close to call" means the final result will be very close and that the CNN Decision Desk may not know who won. CNN will not say a race is "too close to call" early in the night because CNN doesn't have enough vote returns to say someone has won. In that case CNN anchors and analysts will instead say something like "we don't have enough information to know how the race is going," because when the votes come in the winner may, in fact, have a comfortable margin.