1988-2008 Unadjusted Presidential Exit Polls: A 51.8-41.6% Average Democratic Margin

1988-2008 Unadjusted Presidential Exit Polls: A 51.8-41.6% Average Democratic Margin

Richard Charnin
Nov. 13, 2011

This workbook contains a comparative analysis of unadjusted state exit polls and recorded votes for the 1988-2008 Presidential Elections.
https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdFIzSTJtMTJZekNBWUdtbWp3bHlpWGc#gid=0

The data source is the Roper site:
http://www.ropercenter.uconn.edu/elections/common/state_exitpolls.html#.Tr60gD3NltP

Unadjusted exit polls are based on actual respondent totals. The unadjusted numbers were confirmed in other surveys.

But the Final Exit Polls are always forced to match the recorded vote.

In 2004 Kerry led the unadjusted state exit poll aggregate by 51.1-47.5%. The Final National Exit Poll was forced to match the recorded vote (Bush by 50.7-48.3%). All demographic crosstabs had to be changed, either the weighting percentage mix or the vote shares – or both.

The pollsters had to inflate Bush’s pre-election rating by a full 5% in order to force a match to the recorded vote – and perpetuate the fraud. Bush had 48.5% approval based on the average of 11 final pre-election polls. He had 50.3% approval in the unadjusted state exit poll aggregate. In order to force a match to the recorded vote, the exit pollsters had to increase Bush approval from 50.3% to 53% in the Final National Exit Poll. Subtracting the 2.7% differential from Bush’s 48.5% pre-election approval, he had a 45.8% share – exactly matching the True Vote model. The unadjusted exit polls understated Kerry’s True Vote share by 2%. There was a near-perfect 0.99 correlation between Bush’s unadjusted state exit poll shares and approval ratings.

Similarly, to match the recorded vote, the pollsters had to increase Republican Party-ID from 35.1% to 37%. The Democratic/Republican Party ID split was 38.8/35.1% based on the unadjusted state exit poll aggregate. The pollsters had to force the Final National Exit Poll to match the recorded vote by inventing a bogus 37/37% split. There was a near-perfect 0.93 correlation between the Bush unadjusted state exit polls and Republican Party ID.

This chart displays Bush’s unadjusted state exit polls, approval ratings and Republican Party-ID.
https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdFIzSTJtMTJZekNBWUdtbWp3bHlpWGc#gid=21

The True Vote Model (TVM) is based on Census votes cast, mortality, prior election voter turnout and National Exit Poll vote shares. The TVM closely matched the exit polls in each election. In 2008, it was within 0.1% of Obama’s 58.0% unadjusted exit poll share.

http://richardcharnin.wordpress.com/2011/03/17/the-1988-2008-state-and-national-true-vote-model-t/

The Democrats led the 1988-2008 vote shares measured by…
1) recorded vote: 47.9-45.9%
2) unadjusted exit poll: 51.8-41.6%
3) True Vote Model (methods 2-3): 51.6-42.9%
4) True Vote Model (method 4): 53.0-41.0%
5) Exit Pollster (WPE/IMS) method: 50.8-43.1%

The Democrats won the exit poll and lost the recorded vote in the following states:
1988: CA IL MD MI NM PA VT (Dukakis won the unadjusted Nat Exit Poll 50-49%)
1992: AK AL AZ FL IN MS NC OK TX VA
1996: AK AL CO GA ID IN MS MT NC ND SC SD VA
2000: AL AR AZ CO FL GA MO NC TN TX VA (Gore needed just ONE state to win)
2004: CO FL IA MO NM NV OH VA (Kerry would have won if he carried FL or OH)
2008: AL AK AZ GA MO MT NE

This barchart displays the trend in unadjusted exit poll, True Vote and recorded vote shares from 1988-2008:
https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdFIzSTJtMTJZekNBWUdtbWp3bHlpWGc#gid=22

Advertisement