Exit Poll stats via NYTimes
The Election Day polls were based on questionnaires completed by voters as they left voting stations throughout the country on Tuesday, supplemented by telephone interviews with absentee and early voters. The polls were conducted by Edison Research of Somerville, N.J., for the National Election Pool, a consortium of ABC News, Associated Press, CBS News, CNN, Fox News and NBC News. The national results are based on voters in 350 randomly chosen precincts across the United States, and include absentee voters and early voters interviewed by telephone.
The state results are based on voters in 11 to 50 randomly selected precincts across each of 18 states analyzed by The Times. In certain states, some interviews were also conducted by telephone with absentee voters and early voters. In Colorado all interviews were by telephone and in Arizona the majority were. In theory, in 19 cases out of 20, the results from such polls should differ by no more than plus or minus 4 percentage points nationally, and 4 to 5 points in each state, from what would have been obtained by seeking to interview all voters who cast ballots in each of these elections.
Results based on smaller subgroups, like demographic groupings, have a larger potential sampling error. In addition to sampling error, the practical difficulties of conducting any survey of voter opinion on Election Day, such as the reluctance of some voters to take time to fill out the questionnaire, may introduce other sources of error into the poll.
The Times was assisted in its polling analysis by Ana Maria Arumi of Studio Arumi, Barry M. Feinberg of BMF Research & Consulting, Geoffrey D. Feinberg of Yale University, David R. Jones of Baruch College-CUNY, Michael R. Kagay of Princeton, N.J., Jeffrey W. Ladewig of the University of Connecticut, Helmut Norpoth of SUNY-Stony Brook, Annie L. Siegel of New York and Janet L. Streicher of Citibank.