.

.

.

.

1988-2008 Unadjusted Presidential Exit Polls: a 52-41% Democratic margin vs. 48-46% recorded

.

.

Richard Charnin

.

Updated: July 15, 2013

.

.

This workbook contains unadjusted state exit polls for the 1988-2008 Presidential Elections.

.

.

131 of 274 state exit polls moved beyond the margin of error to the Republican and just 4 to the Democrat. The probability is E-116

.

.

Data Source: Roper Center UConn

.

http://www.ropercenter.uconn.edu/elections/common/state_exitpolls.html#.Tr60gD3NltP

.

.

Unadjusted exit polls are based on actual respondent totals.

.

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

.

The unadjusted exit polls are confirmed in other surveys and independent models..

.

.

http://www.electionmathematics.org/em-exitpolls/Exit_Polls_2004_Edison-Mitofsky.pdf

.

True Vote Models

.

The TVM is based on Census votes cast, mortality, prior election voter turnout and National Exit Poll vote shares.

.

The TVM closely matched the unadjusted exit polls in each election from 1988-2008. 

.

http://richardcharnin.com/TrueVoteModels.htmhttp://www.washingtonmonthly.com/archives/individual/2004_11/005178.php

.

.

Data Summary

.

The Democrats led all 1988-2008 presidential election averages....

.

- Recorded vote: 48.1 - 46.0%

.

- Unadjusted state exit poll aggregate: 52.1- 41.4%

.

- Unadjusted national exit poll: 51.9 - 41.7%

.

- True Vote Model 1: 50.2 - 43.2% (previous recorded vote)

.

- True Vote Model 2: 51.6 - 41.8% (previous votes cast)

.

- True Vote Model 3: 52.5 - 40.9% (previous exit poll)

.

- True Vote Model 4: 53.0 - 40.4% (recursive)

.

.

.

States flipping from the Democrats in the exit poll to the GOP in the recorded vote

.

.

1988: CA IL MD MI NM PA VT 

.

Dukakis had a 51-47% edge in 24 battleground state polls. He lost by 7 million votes,

.

.

1992: AK AL AZ FL IN MS NC OK TX VA 

.

Clnton had a 18 million vote margin in the state exit polls. He won the the recorded vote by just 6 million.

.

.

1996: AK AL CO GA ID IN MS MT NC ND SC SD VA 

.

Clinton had a 16 million vote margin in the state exit polls. He won by just 8 million recorded votes.

.

.

2000: AL AR AZ CO FL GA MO NC NV TN TX VA 

.

Gore needed just ONE of these states to win the election. He won the state exit polls by 6 million, matching the TVM. 

.

.

2004: CO FL IA MO NM NV OH VA

.

Kerry needed FL or OH to win. He won the national and state exit polls by 5-6 million with 51-52%. He won the TVM by 10 million with 53.6%.

.

.

2008: AL AK AZ GA MO MT NE 

.

Obama had 58% in the state exit polls (exact match to the TVM), a 23 million margin (9.5 recorded) and 61% in the unadjusted National Exit Poll.

.

.

ALL demographic category weights and/or vote shares are adjusted to force State and National Exit Polls to match the recorded vote

.

.

.

.

.

Presidential Summary

.

(in percent)

.

1988-2008

.

1988-2008Total1988199219962000200420082012

.

Votes Cast688,813102,224113,866105,017110,826125,736131,144

.

Recorded652,31591,595104,42496,275105,417122,294132,310129,132

.

.

Recorded VoteWeightedAverage

.

Democrat48.1547.8945.6443.0149.1848.3848.2852.8751.03

.

Republican46.0245.9953.4637.4640.8247.8750.7245.6047.19

.

Recorded Margin (RV)2.121.91-7.825.558.600.51-2.407.273.84

.

.

.

State Exit Polls (unadjusted)

.

Respondents374,93762,49034,01654,41069,71658,21576,19282,388

.

24 states 

.

Democrat52.1051.9651.6047.5652.6650.7951.0958.06

.

Republican41.3941.3747.3131.6836.9944.3747.5940.29

.

Exit Poll Margin (SEP)10.7110.594.2915.8815.676.423.5017.77

.

Red Shift4.294.346.065.173.542.962.955.25

.

.

Probability AnalysisTotal198819921996200020042008

.

MoE Cluster effect131 of 274 state exit polls moved beyond the margin of error to the Republican and just 4 to the Democrat. The probability is E-116

.

30%State Exit Poll MoE +Cluster (average)3.60%3.47%4.06%3.48%3.84%3.43%3.31%

.

Exit Polls red-shift to GOP232214544344246

.

.

Exit Polls exceeding MoE 135122719172337

.

Exit Polls exceeding MoE for GOP131122718162236

.

4.71E-39

.

Probability of red-shift9.11E-351.39E-042.10E-091.62E-087.67E-035.82E-072.23E-10

.

Probability of exceeding MoE 1.20E-832.38E-064.18E-193.48E-122.15E-106.66E-163.16E-30

.

Probability of exceeding MoE for GOP3.74E-1162.49E-121.09E-262.48E-154.86E-133.45E-202.37E-39

.

2.69E-123

.

.

National Exit Poll (unadjusted)Average198819921996200020042008

.

Respondents71,42614,28511,58615,236n/a13,10813,66017,836

.

Weightedpre-elect poll

.

Democrat51.8651.5849.8146.3152.2048.5151.7160.96

.

Republican41.6541.7649.1533.4737.5046.2746.9537.23

.

Exit Poll Margin (NEP)10.219.820.6012.8415.002.244.7623.73

.

Red Shift4.043.964.213.653.200.873.588.23

.

.

State Exit Poll (WPE)

.

Democrat50.9950.7549.345.750.249.451.958.0

.

Republican43.0343.1249.734.739.846.947.140.5

.

Exit Poll Margin (WPE)7.957.63-0.411.010.42.54.817.5

.

Red Shift2.922.863.712.730.901.003.605.12

.

.

True Vote Model

.

Democrat53.1853.0050.251.153.651.553.658.055

.

Republican40.9540.9848.830.436.544.745.140.443

.

True Vote Margin (TVM)12.2312.021.420.717.16.88.517.612.0

.

Red Shift5.055.064.617.584.253.155.455.17

.

.

Margin Discrepancies

.

TVM - RV10.1110.119.215.20.56.310.910.3

.

TVM - SEP1.531.43-0.24.81.40.45.0-0.2

.

SEP - RV8.588.689.310.36.95.96.010.5

.

NEP - RV8.097.928.37.36.91.77.116.5

.

.

Dem Electoral Vote

.

Recorded 287104366376260252365

.

Unadjusted Exit Polls399260501463401349420

.

.

True Vote ModelWeightedAverage198819921996200020042008

.

Based on Prior Election

.

1 Recorded Vote 50.3750.1748.648.848.448.552.354.5

.

2 Votes Cast51.7351.5750.650.350.050.053.355.2

.

3 Exit Poll 52.6852.5050.651.251.650.954.356.4

.

4 True Vote53.1853.0050.251.153.651.553.658.0

.

.

2-party Exit Poll share

.

Democrat55.9056.5952.260.058.753.451.859.0

.

Republican44.1043.4147.840.041.346.648.241.0

.

.

.

.

.

.

Nat Exit PollState ExitRecordedState Exit

.

.

VotesDemRepDemRepDemRep> MoeRed ShiftRS to GOP

.

51.5841.7651.7241.7148.3446.16232135131

.

.

2012129,132nananana51.0347.19nanana

.

2008132,31060.9637.2358.0640.2952.8745.6463736

.

2004122,29451.7146.9550.9747.7148.2850.72422322

.

2000105,41748.5146.2750.7544.7648.3847.87341716

.

199696,27552.237.552.6437.0649.1840.82441918

.

1992104,42446.3133.4747.5931.7443.0137.46452727

.

198891,59549.8149.1550.348.745.6453.46211212

.

.

.

.

Correlation: Recorded Swing vs. EP Shift

.

Swing: change in 2-party unadjusted exit poll share from previous election.

.

Red-shift: Diffrence between 2-party unadjusted exit poll and recorded vote in current election.

.

Average198819921996200020042008

.

Recorded Swing0.01-0.090.060.20-0.16-0.14

.

Unadj Exit Poll Swing0.46-0.650.100.570.620.38

.

E/M 2004 Report (WPE)0.24-0.290.150.050.48-

.

.

.

In 2004, Bush won the recorded vote by 50.7-48.3%. Kerry won the unadjusted state exit poll aggregate (76,192 respondents) by 51.1-47.5%. He won the unadjusted National Exit Poll (a subset) by 51.7-47.0% (13660 respondents). The pollsters offered the following hypothesis to explain the discrepancy between the exit poll and recorded vote: Kerry voters were more anxious to be polled; 56 Kerry voters agreed to be interviewed for every 50 Bush voters. Well, of course - because there WERE approximately 56 Kerry voters for every 50 Bush voters. The pollster canard was dubbed the reluctant Bush responder (rBr) hypothesis. The pollsters applied the hypothesis (without any evidence) by adjusting the National Exit Poll (NEP) in order to have it match the recorded vote.

.

.

The adjusted NEP indicated that 43% (52.6 million) of the 2004 electorate were returning Bush voters and 37% (45.3 million) were returning Gore voters. But Bush only had 50.5 million votes in 2000. Approximately 2.5 million died prior to the 2004 election and 1.0 million did not return to vote. Therefore, there could have been no more than 47 miliion returning Bush voters (38.4%). In other words, there had to be 5.6 million PHANTOM BUSH VOTERS, which meant that an impossible 110% of living Bush 2000 voters had to turn out in 2004.

.

.

Returning Gore and Bush 2000 voter percentage of 2004 electorate (1.25% annual mortality and 97% Gore/Bush turnout).

.

.

We apply the following methods:

.

1) Votes Cast: Gore 51.0 million recorded + 4.5 of 6 million uncounted: Gore by 50.0-47.3%.

.

Returning Gore/Bush share of 2004 electorate: 41.0 / 38.2%

.

Kerry wins by 53.2-45.4%

.

.

2) 2000 Unadjusted State Exit Poll aggregate: Gore by 50.8-44.5%.

.

Returning Gore/Bush share of 2004 vote: 41.4 / 37.7%

.

Kerry wins by 53.6-45.1%

.

.

3) 2000 Unadjusted National Exit Poll: Gore by 48.5-46.3%

.

Returning Gore/Bush share of 2004 electorate: 39.4 / 37.6%

.

Kerry wins by 52.9-45.7%

.

.

4) 2004 National Exit Poll (adjusted to matched recorded vote)

.

Returning Gore/Bush share of 2004 electorate:37 / 43% (impossible 110% Bush 2000 voter turnout)

.

Bush wins by 50.7-48.3%

.

.

.

The National Exit Poll was forced to match the recorded vote based on an impossible 110% turnout of Bush 2000 voters. Therefore, the recorded vote must also have been impossible.

.

.

Gore won the unadjusted state exit poll aggregate by 50.8-44.5% and the National Exit Poll by 48.5-46.3%. Therefore it is likely that he won by 3-6 million; a 41.4/ 37.7% Gore/Bush returning voter mix is plausible and so Kerry won the True Vote by 53.6-45.1% - a 10.7 million landslide.

.

.

.

2004 NatIonal Exit Poll (unadjusted)

.

SampleKerryBushOther

.

13,6607,0646,414182

.

51.71%46.95%1.33%

.

.

Final 2004 National Exit Poll

.

(Impossible Bush 2000 voter turnout in 2004; forced to match recorded vote)

.

.

2000Recorded (mil.)Alive2004 Turnout2004 RecordedMixKerryBushOther

.

DNV ---20,79017%54%44%2%

.

Gore51.0048,45493%45,24937%90%10%0%

.

Bush 50.4647,933110%52,58643%9%91%0%

.

Other 3.963,79897%3,6693%64%14%22%

.

.

Total105.42100,185-122,294100%48.27%50.73%1.00%

.

59,03162,0401,223

.

.

.

True Vote Model

.

(Plausible 2000 returning voter mix)

.

.

20002004 CastMixKerryBushOther

.

DNV 22,38117.80%57%41%2%

.

Gore52,05541.40%91%8%1%

.

Bush 47,40337.70%10%90%0%

.

Other 3,8983.10%64%17%19%

.

.

Total125,737100%53.57%45.07%1.36%

.

67,36256,6661,709

.

.

Bush Approval Ratings

.

For example, to adjust Kerry's 51.1 - 47.5% unadjusted exit poll margin to Bush's 50.7- 48.3% recorded margin in the Final National Exit Poll,

.

Bush's 50.3% unadjusted approval rating was increased to 53%. Corresponding vote shares were increased as well.

.

Bush had just 48% approval in the final pre-election polls.

.

With 48% approval applied to the NEP shares, Kerry had 53.7% (see the sensitivity analysis table below).

.

.

Party-ID

.

Dem/Rep Party-ID was changed from 38.5-35.1% to 37-37% to match the recorded vote.

.

There were near-perfect correlations between Bush's unadjusted state exit poll shares, approval ratings and Party-ID.

.

.

The 2006 midterms further illustrate how unadjusted exit polls are forced to match a bogus recorded vote. Democratic Party ID was reduced from 39.8% to 38%; GOP increased from 33.4% to 36%. Similar adjustments were made in 2004 and 2008.

.

.

2006 Unadjusted National Exit Poll

.

13,251 respondentsSample Democrat RepublicanOtherMargin

.

http://www.ropercenter.uconn.edu/data_access/tag/congressional_elections.html#.T8yypbA7WAgHouse1325174705476305

.

Total56.37%41.33%2.30%15.05%

.

.

Senate978357033678402

.

Total58.30%37.60%4.11%20.70%

.

.

Governor1055257514181620

.

Total54.50%39.62%5.88%14.88%

.

.

Party ID12850511042883452

.

39.77%33.37%26.86%6.40%

.

2006 Adjusted National Exit Poll ______________________________________________________________________________________________________________

.

Adjusted to match recorded voteMale49%50%47%3%

.

http://www.cnn.com/ELECTION/2006/pages/results/states/US/H/00/epolls.0.htmlFemale51%55%43%2%

.

100%52.55%44.96%2.49%7.59%

.

Party ID

.

Democrat38%93%7%0%

.

Republican36%8%91%1%

.

Other26%57%39%4%

.

100%53.04%45.56%1.40%7.48%

.

.

.

The return voter mix was changed.Voted 2004Adjusted Mix Democrat RepublicanOtherVotes

.

(6,361 respondents)Kerry 43%92%7%1%2,735

.

Bush 49%15%83%2%3,117

.

Other4%66%23%11%254

.

Did Not Vote4%66%32%2%254

.

Total100%52.19%45.88%1.93%6361

.

.

How Voted 2004Exit PollUnadjusted Mix

.

45.48%48.43%Kerry 48.43%92%7%1%3,081

.

46.49%43.57%Bush 43.57%15%83%2%2,771

.

4.13%4.00%Other4.00%66%23%11%254

.

3.90%4.00%Did Not Vote4.00%66%32%2%254

.

Total100%56.37%41.75%1.88%6361

.

.

.

.

.

2004 National Exit PollRespondentsKerryBushOther

.

Unadjusted13,6607,0646,414182

.

Share51.71%46.95%1.33%

.

.

.

2004 Unadjusted National Exit PollFinal National Exit Poll

.

Party - ID13,660 respondents(matched to recorded vote)

.

.

AggregateFinal

.

Party-IDKerryBushOtherParty-IDKerryBushOther

.

Democrat38.76%91%9%0%37%90%10%0%

.

Republican35.12%8%92%0%37%7%93%0%

.

Independent26.12%52%43%5%26%47.62%48.55%3.83%

.

.

Share100%51.66%47.03%1.31%100%48.27%50.73%1.00%

.

Votes125,73664,96059,1341,642122,29659,03462,0441,218

.

.

.

Bush Approval

.

.

ActualKerryBushOtherUnadjustedKerryBushOther

.

Approve48.00%9%90%1%50.30%9%90%1%

.

Disapprove52.00%95%4%1%49.70%95%4%1%

.

.

Share100%53.72%45.28%1.00%100%51.74%47.26%1.00%

.

Votes125,73667,54556,9331,257122,29663,27857,7951,223

.

.

.

Sensitivity Analysis

.

RepDemocratic Party-IDTotal Bush Approval

.

Party-ID37.00%38.00%38.76%Kerry Bush50.30%49.00%48.00%

.

Kerry shareAppovalKerry share

.

35.12%50.98%51.37%51.66%10%52.25%53.35%54.20%

.

36%50.52%50.91%51.21%9%51.74%52.86%53.72%

.

37%50.00%50.39%50.69%8%51.24%52.37%53.24%

.

.

.

.

Democratic Vote Shares

.

UnweightedWeighted

.

198819921996200020042008AverageAverage

.

State Unadjusted Exit Poll51.6047.5652.6650.7951.0958.0651.9652.10

.

National Unadjusted Exit Poll49.8146.3152.2048.5151.7160.9651.5851.86

.

State Exit Poll (WPE/IMS)49.3045.7050.2049.4051.9058.0050.7550.99

.

True Vote Model50.2051.1053.6051.5053.6058.0053.0053.18

.

National Recorded Vote45.6443.0149.1848.3848.2852.8747.8948.05

.

.

Votes Cast102.25113.87105.01110.83125.74131.4651.8252.04

.

Total689.15

.

.

.

.

.

Related Links

.

.

http://abcnews.go.com/images/Politics/EvaluationofEdisonMitofskyElectionSystem.pdf

.

.

US Count Votes analysis of the Ohio 2004 exit poll discrepancies

.

http://www.electionmathematics.org/em-exitpolls/OH/2004Election/Ohio-Exit-Polls-2004.pdf

.

.

Exit Poll Response

.

http://www.electionmathematics.org/em-exitpolls/USCV_exit_poll_analysis.pdf

.

http://www.richardcharnin.com/ExitPollResponseOptimization.htm

.

.

.

.

.

.

REp112616122236

.

Dem000111

.