National Election Pool

The National Election Pool (NEP) is a consortium of American news organizations formed in 2003 to provide "information on Election Night about the vote count, election analysis and election projections." Member companies consist of ABC News, the Associated Press, CBS News, CNN, FOX News and NBC News.

NEP has relied on the Associated Press to perform vote tabulations and has contracted with Edison Media Research and Mitofsky International to "make projections and provide exit poll analysis."

NEP and election protection
Exit polls are valued throughout the world as an important election verification technique. When Mexico sought legitimacy as a modernizing democracy in 1994, Carlos Salinas instituted reforms designed to ensure fair elections. A central feature of those reforms was exit polls. In the 2000 Mexican election, Mitofsky was hired to conduct these exit polls to protect against vote fraud.

The US has funded exit polls abroad because, state dept. officials testified, it is one of the few ways to expose and ascertain the extent of large-scale fraud. Indeed, discrepancies between exit polls and the official results have been used to successfully overturn election results in Serbia, Peru, the Republic of Georgia and, in November 2004, Ukraine.

In contrast to overseas exit polls, however, the US National Exit Poll (NEP) media exit poll does not report actual survey results, but rather adjusts data so as to conform with official numbers before reporting them.

History
Warren Mitofsky, the founder of Mitofsky International, introduced exit polling at CBS News as a way to help explain why voters chose one candidate over another. In the 1980 election, NBC for the first time used the exit polls to predict the outcome of the elections. The network blew away its competition -- correctly projecting races in state after state long before either ABC or CBS, who were still relying on officially reported results.

In 1990, the networks pooled resources into a joint effort, called Voter Research Service (VRS), with Mitofsky as director, but answerable to the network representatives. In 1994, the networks also consolidated their vote counting operation (National Election Service -- or NES) with their exit poll operation. This new joint venture was renamed Voter News Service (VNS).

In 2000, VNS data led all the networks to, first, project Gore the winner, then retract the projection, then project Bush the winner -- and thereby President, before retracting that projection as well. Based on the projection of Bush as winner, Gore called Bush to concede. When the error was observed, minutes before Gore was to deliver a public concession, Gore retracted his concession, and a long recount process entailed.

In 2002, VNS contracted with the Battelle Memorial Institute of Ohio to provide a software upgrade, but Battelle failed to produce. On election night that year, the networks announced that there would be no exit poll data, either for analysis of the vote or for projections, because of glitches in the new software. VNS was disbanded in 2003 and NEP was formed.

2004 Election
Leaked exit polling data throughout the day and even after the polls closed indicated a Kerry victory. News organizations operated most of the day on an expectation of a Kerry victory.

NEP co-director, Edison's Joe Lenski responded to questions about the discrepancy between these numbers and official results by claiming that it was only "early" exit polls that showed Kerry leading and by denouncing bloggers who leaked "early" data indicating Kerry had won. Said Lenski, "I'm not designing polls for some blogger who doesn't even understand how to read the data."

Refusal to release data
NEP came under pressure for refusal to release exit poll data from a variety of sources including a congresional forum called by John Conyers, Jr., Ranking member of the House Judiciary Committee.

NEP did evnetually release both data in archival form for statistical study, but these data did not include variables identifying precinct so that they were not useful in ascertaining the veracity of the US election. NEP also released a report on Jan 19, 2005 (Inauguration eve) including state, but not precinct level, data that could be used to pinpoint large disparties that could be investigated to determine whether, in fact, there was massive polling error or the official numbers were incorrect.

The 2005 Inauguration eve report
On January 19, 2005, the eve of Bush’s inauguration, Mitofsky and Lenski released their explanation of why the exit polls diverged so greatly from thee official count in 2004. However, their report, “Evaluation of Edison/Mitofsky Election System 2004”. The report generated headlines such as MSNBC's "Exit Polls Prove That Bush Won.” But the report did not even attempt any such proof. Rather, it restates the thesis that the pollsters had previously intimated—that the discrepancy was “most likely due to Kerry voters participating in the exit polls at a higher rate than Bush voters.”

Critics claim, however, that the report's own data contradict this thesis, noting that participation rates were lower where Kerry voters predominated; and that, on the other hand, polling error cannot explain greater disparities in battleground states, states with Republican governors, and states there were more reports of Election Day voting problems. Or that in contrast to the seven-percentage-point disparity in precincts where the votes were counted by machines, no disparity is found in those few remaining precincts where votes were cast on paper and counted manually.

2006 Election
Problems struck yet again during the 2006 congressional elections. Early polling data on the Senate races across the country showed a higher percentage of Democrats having been surveyed than were being shown in raw vote tallies. Most members and subscribers decided to ignore the polling on the night of the election. 

Exit polls and election verification - moving forward
In 2004, we knew of the seven percentage point national discrepancy between how people said they voted and official numbers (11 percentage points in Ohio) only because of a technical glitch that prevented NEP from "correcting" results on election night. Such leaks and glitches are unlikely to happen again. In a presentation to the American Statistical Association, Warren Mitofsky that as an "improvement" in future exit polls,those with access to the data are to be quarantined without electronic or phone communication and subsequently sworn to secrecy.

Election protection groups, Election Integrity, Election Defense Alliance and The Warren Poll are now conducting donor-supported transparent Election Verification Exit Polls (EVEPs) with methods and data fully disclosed to the public.

Related SourceWatch articles

 * Exit poll
 * Election verification exit poll

External resources
Books
 * Steven F. Freeman and Joel Bleifuss, Was the 2004 Presidential Election Stolen? Exit Polls, Election Fraud, and the Official Count(New York: Seven Stories Press, 2006) Chapters 4 "Biased polls or biased count?" and 5 "The Inauguration Eve Exit-Poll Report"
 * In Defense of Public Opinion Polling by Ken Warren (Cambridge, Mass:Westview Press, 2003).

Websites:
 * Election Integrity: http://www.electionintegrity.org/
 * Edison Media Research and Mitofsky International National Elections Pool: http://www.exit-poll.net/faq.html
 * Edison Media Research exit poll page
 * National Election Archive Project: http://uscountvotes.org/

Articles:
 * Corn, David. 2004. A stolen election? The Nation 279, no. 18 (November 29): 5-7.
 * Steven F. Freeman, Who Really Won – and Lost – the 2004 US Presidential Election? Presentation to the American Association for Public Opinion Research, Montreal, May 19, 2006*Traugott, Michael, Benjamin Highton, and Henry E. Brady. 2005. A review of recent controversies concerning the 2004 presidential election exit polls. The National Research Commission on Elections and Voting.
 * US Count Votes. 2005 Analysis of the 2004 presidential election exit poll discrepancies. National Election Data Archive Project.
 * US Count Votes. 2005. The 2004 presidential election: Exit poll error or vote miscount? National Election Data Archive.
 * US Count Votes. 2005. History of the debate surrounding the 2004 presidential election.