TruthIsAll
01-11-2009, 09:57 AM
http://www.geocities.com/electionmodel/2008MediaCherryPickLV.htm
http://www.realclearpolitics.com/epolls/2008/president/us/general_election_mccain_vs_obama-225.html
Is the Media Cherry-picking Pre-election Polls by listing Likely (LV) but not Registered Voter (RV) Polls?
TruthIsAll
The media seems intent on pushing likely voter (LV) polls and downplaying registered voter (RV) polls. RV polls include new voters - mostly young Democrats. Since LV polls do not include new voters, they underestimate the projected Democratic share. From 1992-2004, Democrats won new voters by an average 14% margin. In 2008, Obama won new voters by 71-27%.
Real Clear Politics (RCP) is not being real clear. RCP does not list the final 5 registered-voter (RV) polls that show an average Obama lead of 52.2 - 39.6%, a 12.6% margin (the latest RV poll listed is the 10/13 Ipsos poll. Instead, RCP averaged the final 15 likely-voter (LV) polls: 52.1 - 44.5%, a 7.6% margin. The average of the final 19 LV polls was 50.84 - 42.68%, an 8.16% margin. This analysis will show that the LV and RV average discrepancy is to be expected. In fact, the 19 LV polls, when combined with Obama and McCain shares of new voters, match the RV polls - and confirm that Obama won by double his official 9.5m vote margin.
On July 28, USA Today released their polling results - and raised eyebrows: “Obama was ahead 47%-44% among registered voters, down from a 6-percentage point lead he had last month. McCain led 49%-45% among likely voters, reversing a 5-point Obama lead among that group. In both cases, the margin of error is +/-4 points”. Obama’s margin dropped 3 points in the RV poll but by a whopping 9 points in the LV poll.
According to Gallup’s Frank Newport: "Registered voters are much more important at the moment" and that McCain's lead among likely voters suggests "the possibility's there that Republicans can get energized.” Obama led McCain 48%-40% among registered voters in a separate Gallup tracking poll over the same three days. But it was not even suggested that the reason for the discrepancy between the RV and LV polls was due to the fact that the Democrats were registering millions of new voters while Republicans were deserting the GOP.
Why did RCP fail to list these 5 RV polls?
RV Poll Date Obama McCain Spread
Gallup 11/02 53 40 +13
ABC/WP 11/02 54 41 +13
Pew 11/01 50 39 +11
Pew 10/26 52 36 +16
NBC 10/20 52 42 +10
Total Average 52.2 39.6 +12.6
With 131.37m votes recorded as of today, Obama’s margin is 9.5m (69.46 - 59.94m). He has 59.2% of the 10.2m votes since Election Day, raising his total vote share to 52.87 - 45.62%. He’s won new voters by 12.1 - 4.6 million (71-27%), a 7.5m vote margin – almost equal to his total.
Neither the RV nor LV polls allocate undecided voters. Assuming the official third-party 1.5% recorded share, we can allocate the undecided vote. Typically, undecided voters break strongly for the challenger - and Obama is considered the challenger since McCain represented Bush’s third-term.
Considering that LV polls do not include new voters, why not calculate the total vote in two stages: likely (LV) and new (RV) voters?
1) Use the average final LV poll to calculate Obama/McCain shares of returning (likely) Bush and Kerry voters.
2) Use the 2008 National Exit Poll (NEP) for Obama/McCain shares of new voters.
The Final NEP is always forced to match the official vote count. According to the 2008 NEP, 13% (17.1m) of the 131.37m who voted were new voters. Let’s check that number. We know that there were 122.3m votes recorded in 2004. Applying the 1.2% annual voter mortality rate and assuming a 95% turnout of 2004 voters still living, 110.6m former Bush, Kerry and third-party voters returned to vote in 2008.
Since 131.37m votes have been recorded, there had to be approximately 20.77m new voters (15.8% of the total). Therefore, the NEP understated the number of new voters by approximately 4 million (3%) and overstated the percentage mix (4%) of returning third-party voters by the same 3% - an impossible 4 million. Although this does not have a material effect on the total Obama share (he had 71% of new voters and 66% of others) it does indicate that more weight should be given to the pre-election RV polls, which included new voters, than the LV polls, which did not.
Consider the final Nov. RV and LV polls:
- 3 RV polls had Obama leading 52.33 - 40.0%. Assuming he had 75% of the undecided vote, his projected RV share was 56.96%.
- 9 LV polls had Obama leading 51.00 - 43.6%. Assuming he had 75% of the undecided vote, his projected LV share was 53.96%
Based on his projected LV share, he won returning (LV) voters by 59.7 – 49.3m.
- The National Exit Poll indicates that Obama won new voters by 71-27% (14.7 - 5.6m).
Combining returning and new voter shares, Obama won by 74.4 - 54.9m (56.7 - 41.8%), within 0.3% of the RV 56.96% projection.
The 56.7% share was also close to the Election Calculator model (57.5%) – an independent confirmation based on a plausible returning 2004 voter mix and Final 2008 Mational Exit Poll vote shares.
Note: the final shares do not include uncounted votes, which are heavily (70-80%) Democratic. In 2004, the Census survey indicated that 125.7m votes were cast - but only 122.3m recorded. In 2000, 105.4m were recorded out of 110.8m cast.
Based on the above, we can conclude that:
1. Pre-election LV polls were correct in predicting the Likely Voter (returning) shares.
2. Pre-election RV polls were correct in predicting the Registered Voter (returning and new) shares.
3. LV polls were misleading since they failed to sample over 20m new voters, lowering Obama’s share by 3%.
4. RV polls were realistic since new, first- time voters were sampled.
5. The True Vote can be calculated as a combination of two components:
a) Final LV poll average with undecided voters allocated to determine the Obama/McCain shares of returning 2004 voters
b) National Exit Poll to determine the shares of new voters.
The 2008 Election Analysis is a comprehensive compendium of pre-and post election polls, projections and a True Vote sensitivity analysis.
LV and RV Projections of Returning 2004 Voters
Recorded Total Obama McCain Other
Vote 131.37 69.46 59.93 1.98
Share 100% 52.87% 45.62% 1.51%
Calculating the 2008 True Vote
(Assume final LV polls for returning voters and National Exit Poll for new voter shares)
Pre-election Nov. Poll shares
Poll BO JM Other Undecided
LV (9) 51.00% 43.56% 1.50% 3.94%
RV (3) 52.33% 40.00% 1.50% 6.17%
Allocate undecided voters
UVA Obama McCain Other
LV 53.96% 44.55% 1.50%
RV 56.96% 41.55% 1.50%
Calculate returning voters from prior election
2004 122.29
Died 5.87 4.8%
Alive 116.42
Turnout 110.60 95.0% (LV)
2008 131.37
New 20.77
Vote Obama McCain Other Total
NEP 71% 27% 2% 100%
New 14.75 5.61 0.42 20.77
Return 53.96% 44.55% 1.50% 100.0% (LV)
Turnout 59.67 49.27 1.66 110.6
True Vote
Total 74.42 54.87 2.07 131.37
Share 56.65% 41.77% 1.58% 100.0%
Recorded Vote
Total 69.46 59.93 1.98 131.37
Share 52.87% 45.62% 1.51% 100.0%
Final RV Polls (11/02)
75%UVA 25%UVA
Poll Poll Sample Obama McCain Spread Obama McCain Spread
3 RV Date Average 52.33 40.00 12.33 56.96 41.54 15.42
Gallup 10/31 - 11/02 2847 53 40 13 57.125 41.375 15.75
ABC/WP 10/31 - 11/02 2446 54 41 13 56.625 41.875 14.75
Pew 10/29 - 11/01 1325 50 39 11 57.125 41.375 15.75
LV and RV Poll Averages (10/03-11/03)
75%UVA 25%UVA
Average Obama McCain Spread Obama McCain Spread
10 RV 50.30 40.00 10.30 56.45 42.05 14.40
25 LV 50.84 42.60 8.24 54.64 43.87 10.77
Final 10 RV Polls (10/04-11/02)
75%UVA 25%UVA
Poll Poll Sample Obama McCain Spread Obama McCain Spread
10 RV Date Average 50.30 40.00 10.30 56.45 42.05 14.40
Gallup 10/31 - 11/02 2847 53 40 13 57.13 41.38 15.75
ABC/WP 10/31 - 11/02 2446 54 41 13 56.63 41.88 14.75
Pew 10/29 - 11/01 1325 50 39 11 57.13 41.38 15.75
Pew 10/23 - 10/26 1325 52 36 16 59.88 38.63 21.25
NBC/WSJ 10/18 - 10/20 1159 52 42 10 55.38 43.13 12.25
Ipsos 10/09 - 10/13 1036 48 39 9 56.63 41.88 14.75
Newsw 10/08 - 10/09 1035 52 41 11 56.13 42.38 13.75
FOX 10/08 - 10/09 900 46 39 7 56.13 42.38 13.75
Ipsos 10/02 - 10/06 858 47 40 7 55.63 42.88 12.75
NBC/WSJ 10/04 - 10/05 658 49 43 6 53.88 44.63 9.25
Final 25 LV Polls (10/03-11/03)
75%UVA 25%UVA
Poll Poll Sample Obama McCain Spread Obama McCain Spread
25 LV Date Size 50.84 42.60 8.24 54.64 43.87 10.77
Marist 11/03 - 11/03 804 52 43 9 54.63 43.88 10.75
Zogby 11/01 - 11/03 1201 54 43 11 55.13 43.38 11.75
FOX 11/02 - 11/03 971 50 43 7 54.13 44.38 9.75
Hotline 10/31 - 11/02 882 50 45 5 52.63 45.88 6.75
Rasmus 10/31 - 11/02 3000 51 46 5 52.13 46.38 5.75
Res2k 10/31 - 11/02 1100 51 44 7 53.63 44.88 8.75
NBC/WSJ 11/01 - 11/02 1011 51 43 8 54.38 44.13 10.25
CNN 10/30 - 11/01 1017 51 43 8 54.38 44.13 10.25
Pew 10/30 - 11/01 2587 49 42 7 54.63 43.88 10.75
CBS 10/28 - 10/31 1005 54 41 13 56.63 41.88 14.75
Battleg 10/27 - 10/30 1000 49 45 4 52.38 46.13 6.25
Marist 10/29 - 10/29 543 50 43 7 54.13 44.38 9.75
FOX 10/28 - 10/29 924 47 44 3 52.63 45.88 6.75
Ipsos 10/23 - 10/27 831 50 45 5 52.63 45.88 6.75
Newswk 10/22 - 10/23 882 53 41 12 56.38 42.13 14.25
CBS/NYT 10/19 - 10/22 771 52 39 13 57.63 40.88 16.75
FOX 10/20 - 10/21 936 49 40 9 56.13 42.38 13.75
Pew Res 10/16 - 10/19 2382 53 39 14 57.88 40.63 17.25
Ipsos 10/16 - 10/20 773 50 42 8 54.88 43.63 11.25
CNN 10/17 - 10/19 764 51 46 5 52.13 46.38 5.75
Pew Res 10/12 - 10/14 1191 50 40 10 56.38 42.13 14.25
CBS/NYT 10/10 - 10/13 699 53 39 14 57.88 40.63 17.25
LAT 10/10 - 10/13 1030 50 41 9 55.63 42.88 12.75
ABC/WP 10/09 - 10/11 766 53 43 10 54.88 43.63 11.25
CBS/NYT 10/03 - 10/05 616 48 45 3 52.13 46.38 5.75
Probability Obama Exceeds Given Vote Share for various MoE
(based on 11/2 RV Polls, 5293 Sample MoE = 1.33%)
MOE 1.00% 1.20% 1.33% 1.50% 1.75% 2.00%
2pty Full Probability
55% 54.18% 100% 100% 100% 100% 100% 100%
56% 55.16% 100% 100% 99% 99% 97% 96%
57% 56.15% 93% 89% 86% 83% 80% 77%
57.74% 56.87% 50% 50% 50% 50% 50% 50%
58% 57.13% 31% 34% 35% 37% 39% 40%
59% 58.12% 1% 2% 3% 5% 8% 11%
60% 59.10% 0% 0% 0% 0% 1% 1%
The Election Calculator Confirmed the LV poll / NEP new voter combination
(the 2004 unadjusted exit poll (Kerry by 52-47%) is assumed as the basis for calculating returning voter mix)
True Vote
Turnout 2004 Alive Voted Mix Obama McCain Other Obama McCain Other
DNV 20.77 15.8% 71% 27% 2% 14.74 5.61 0.42
95% Kerry 60.54 57.51 43.8% 89% 9% 2% 51.19 5.18 1.15
95% Bush 54.72 51.98 39.6% 17% 82% 1% 8.84 42.63 0.52
95% Other 1.16 1.11 0.8% 66% 24% 10% 0.73 0.27 0.11
Total 116.42 131.37 100.0% 57.47% 40.86% 1.67% 75.50 53.68 2.20
Recorded 131.37 52.87% 45.62% 1.51% 69.457 59.935 1.978
Diff 0.00 4.60% -4.76% 0.17% 6.04 -6.26 0.22
http://www.realclearpolitics.com/epolls/2008/president/us/general_election_mccain_vs_obama-225.html
Is the Media Cherry-picking Pre-election Polls by listing Likely (LV) but not Registered Voter (RV) Polls?
TruthIsAll
The media seems intent on pushing likely voter (LV) polls and downplaying registered voter (RV) polls. RV polls include new voters - mostly young Democrats. Since LV polls do not include new voters, they underestimate the projected Democratic share. From 1992-2004, Democrats won new voters by an average 14% margin. In 2008, Obama won new voters by 71-27%.
Real Clear Politics (RCP) is not being real clear. RCP does not list the final 5 registered-voter (RV) polls that show an average Obama lead of 52.2 - 39.6%, a 12.6% margin (the latest RV poll listed is the 10/13 Ipsos poll. Instead, RCP averaged the final 15 likely-voter (LV) polls: 52.1 - 44.5%, a 7.6% margin. The average of the final 19 LV polls was 50.84 - 42.68%, an 8.16% margin. This analysis will show that the LV and RV average discrepancy is to be expected. In fact, the 19 LV polls, when combined with Obama and McCain shares of new voters, match the RV polls - and confirm that Obama won by double his official 9.5m vote margin.
On July 28, USA Today released their polling results - and raised eyebrows: “Obama was ahead 47%-44% among registered voters, down from a 6-percentage point lead he had last month. McCain led 49%-45% among likely voters, reversing a 5-point Obama lead among that group. In both cases, the margin of error is +/-4 points”. Obama’s margin dropped 3 points in the RV poll but by a whopping 9 points in the LV poll.
According to Gallup’s Frank Newport: "Registered voters are much more important at the moment" and that McCain's lead among likely voters suggests "the possibility's there that Republicans can get energized.” Obama led McCain 48%-40% among registered voters in a separate Gallup tracking poll over the same three days. But it was not even suggested that the reason for the discrepancy between the RV and LV polls was due to the fact that the Democrats were registering millions of new voters while Republicans were deserting the GOP.
Why did RCP fail to list these 5 RV polls?
RV Poll Date Obama McCain Spread
Gallup 11/02 53 40 +13
ABC/WP 11/02 54 41 +13
Pew 11/01 50 39 +11
Pew 10/26 52 36 +16
NBC 10/20 52 42 +10
Total Average 52.2 39.6 +12.6
With 131.37m votes recorded as of today, Obama’s margin is 9.5m (69.46 - 59.94m). He has 59.2% of the 10.2m votes since Election Day, raising his total vote share to 52.87 - 45.62%. He’s won new voters by 12.1 - 4.6 million (71-27%), a 7.5m vote margin – almost equal to his total.
Neither the RV nor LV polls allocate undecided voters. Assuming the official third-party 1.5% recorded share, we can allocate the undecided vote. Typically, undecided voters break strongly for the challenger - and Obama is considered the challenger since McCain represented Bush’s third-term.
Considering that LV polls do not include new voters, why not calculate the total vote in two stages: likely (LV) and new (RV) voters?
1) Use the average final LV poll to calculate Obama/McCain shares of returning (likely) Bush and Kerry voters.
2) Use the 2008 National Exit Poll (NEP) for Obama/McCain shares of new voters.
The Final NEP is always forced to match the official vote count. According to the 2008 NEP, 13% (17.1m) of the 131.37m who voted were new voters. Let’s check that number. We know that there were 122.3m votes recorded in 2004. Applying the 1.2% annual voter mortality rate and assuming a 95% turnout of 2004 voters still living, 110.6m former Bush, Kerry and third-party voters returned to vote in 2008.
Since 131.37m votes have been recorded, there had to be approximately 20.77m new voters (15.8% of the total). Therefore, the NEP understated the number of new voters by approximately 4 million (3%) and overstated the percentage mix (4%) of returning third-party voters by the same 3% - an impossible 4 million. Although this does not have a material effect on the total Obama share (he had 71% of new voters and 66% of others) it does indicate that more weight should be given to the pre-election RV polls, which included new voters, than the LV polls, which did not.
Consider the final Nov. RV and LV polls:
- 3 RV polls had Obama leading 52.33 - 40.0%. Assuming he had 75% of the undecided vote, his projected RV share was 56.96%.
- 9 LV polls had Obama leading 51.00 - 43.6%. Assuming he had 75% of the undecided vote, his projected LV share was 53.96%
Based on his projected LV share, he won returning (LV) voters by 59.7 – 49.3m.
- The National Exit Poll indicates that Obama won new voters by 71-27% (14.7 - 5.6m).
Combining returning and new voter shares, Obama won by 74.4 - 54.9m (56.7 - 41.8%), within 0.3% of the RV 56.96% projection.
The 56.7% share was also close to the Election Calculator model (57.5%) – an independent confirmation based on a plausible returning 2004 voter mix and Final 2008 Mational Exit Poll vote shares.
Note: the final shares do not include uncounted votes, which are heavily (70-80%) Democratic. In 2004, the Census survey indicated that 125.7m votes were cast - but only 122.3m recorded. In 2000, 105.4m were recorded out of 110.8m cast.
Based on the above, we can conclude that:
1. Pre-election LV polls were correct in predicting the Likely Voter (returning) shares.
2. Pre-election RV polls were correct in predicting the Registered Voter (returning and new) shares.
3. LV polls were misleading since they failed to sample over 20m new voters, lowering Obama’s share by 3%.
4. RV polls were realistic since new, first- time voters were sampled.
5. The True Vote can be calculated as a combination of two components:
a) Final LV poll average with undecided voters allocated to determine the Obama/McCain shares of returning 2004 voters
b) National Exit Poll to determine the shares of new voters.
The 2008 Election Analysis is a comprehensive compendium of pre-and post election polls, projections and a True Vote sensitivity analysis.
LV and RV Projections of Returning 2004 Voters
Recorded Total Obama McCain Other
Vote 131.37 69.46 59.93 1.98
Share 100% 52.87% 45.62% 1.51%
Calculating the 2008 True Vote
(Assume final LV polls for returning voters and National Exit Poll for new voter shares)
Pre-election Nov. Poll shares
Poll BO JM Other Undecided
LV (9) 51.00% 43.56% 1.50% 3.94%
RV (3) 52.33% 40.00% 1.50% 6.17%
Allocate undecided voters
UVA Obama McCain Other
LV 53.96% 44.55% 1.50%
RV 56.96% 41.55% 1.50%
Calculate returning voters from prior election
2004 122.29
Died 5.87 4.8%
Alive 116.42
Turnout 110.60 95.0% (LV)
2008 131.37
New 20.77
Vote Obama McCain Other Total
NEP 71% 27% 2% 100%
New 14.75 5.61 0.42 20.77
Return 53.96% 44.55% 1.50% 100.0% (LV)
Turnout 59.67 49.27 1.66 110.6
True Vote
Total 74.42 54.87 2.07 131.37
Share 56.65% 41.77% 1.58% 100.0%
Recorded Vote
Total 69.46 59.93 1.98 131.37
Share 52.87% 45.62% 1.51% 100.0%
Final RV Polls (11/02)
75%UVA 25%UVA
Poll Poll Sample Obama McCain Spread Obama McCain Spread
3 RV Date Average 52.33 40.00 12.33 56.96 41.54 15.42
Gallup 10/31 - 11/02 2847 53 40 13 57.125 41.375 15.75
ABC/WP 10/31 - 11/02 2446 54 41 13 56.625 41.875 14.75
Pew 10/29 - 11/01 1325 50 39 11 57.125 41.375 15.75
LV and RV Poll Averages (10/03-11/03)
75%UVA 25%UVA
Average Obama McCain Spread Obama McCain Spread
10 RV 50.30 40.00 10.30 56.45 42.05 14.40
25 LV 50.84 42.60 8.24 54.64 43.87 10.77
Final 10 RV Polls (10/04-11/02)
75%UVA 25%UVA
Poll Poll Sample Obama McCain Spread Obama McCain Spread
10 RV Date Average 50.30 40.00 10.30 56.45 42.05 14.40
Gallup 10/31 - 11/02 2847 53 40 13 57.13 41.38 15.75
ABC/WP 10/31 - 11/02 2446 54 41 13 56.63 41.88 14.75
Pew 10/29 - 11/01 1325 50 39 11 57.13 41.38 15.75
Pew 10/23 - 10/26 1325 52 36 16 59.88 38.63 21.25
NBC/WSJ 10/18 - 10/20 1159 52 42 10 55.38 43.13 12.25
Ipsos 10/09 - 10/13 1036 48 39 9 56.63 41.88 14.75
Newsw 10/08 - 10/09 1035 52 41 11 56.13 42.38 13.75
FOX 10/08 - 10/09 900 46 39 7 56.13 42.38 13.75
Ipsos 10/02 - 10/06 858 47 40 7 55.63 42.88 12.75
NBC/WSJ 10/04 - 10/05 658 49 43 6 53.88 44.63 9.25
Final 25 LV Polls (10/03-11/03)
75%UVA 25%UVA
Poll Poll Sample Obama McCain Spread Obama McCain Spread
25 LV Date Size 50.84 42.60 8.24 54.64 43.87 10.77
Marist 11/03 - 11/03 804 52 43 9 54.63 43.88 10.75
Zogby 11/01 - 11/03 1201 54 43 11 55.13 43.38 11.75
FOX 11/02 - 11/03 971 50 43 7 54.13 44.38 9.75
Hotline 10/31 - 11/02 882 50 45 5 52.63 45.88 6.75
Rasmus 10/31 - 11/02 3000 51 46 5 52.13 46.38 5.75
Res2k 10/31 - 11/02 1100 51 44 7 53.63 44.88 8.75
NBC/WSJ 11/01 - 11/02 1011 51 43 8 54.38 44.13 10.25
CNN 10/30 - 11/01 1017 51 43 8 54.38 44.13 10.25
Pew 10/30 - 11/01 2587 49 42 7 54.63 43.88 10.75
CBS 10/28 - 10/31 1005 54 41 13 56.63 41.88 14.75
Battleg 10/27 - 10/30 1000 49 45 4 52.38 46.13 6.25
Marist 10/29 - 10/29 543 50 43 7 54.13 44.38 9.75
FOX 10/28 - 10/29 924 47 44 3 52.63 45.88 6.75
Ipsos 10/23 - 10/27 831 50 45 5 52.63 45.88 6.75
Newswk 10/22 - 10/23 882 53 41 12 56.38 42.13 14.25
CBS/NYT 10/19 - 10/22 771 52 39 13 57.63 40.88 16.75
FOX 10/20 - 10/21 936 49 40 9 56.13 42.38 13.75
Pew Res 10/16 - 10/19 2382 53 39 14 57.88 40.63 17.25
Ipsos 10/16 - 10/20 773 50 42 8 54.88 43.63 11.25
CNN 10/17 - 10/19 764 51 46 5 52.13 46.38 5.75
Pew Res 10/12 - 10/14 1191 50 40 10 56.38 42.13 14.25
CBS/NYT 10/10 - 10/13 699 53 39 14 57.88 40.63 17.25
LAT 10/10 - 10/13 1030 50 41 9 55.63 42.88 12.75
ABC/WP 10/09 - 10/11 766 53 43 10 54.88 43.63 11.25
CBS/NYT 10/03 - 10/05 616 48 45 3 52.13 46.38 5.75
Probability Obama Exceeds Given Vote Share for various MoE
(based on 11/2 RV Polls, 5293 Sample MoE = 1.33%)
MOE 1.00% 1.20% 1.33% 1.50% 1.75% 2.00%
2pty Full Probability
55% 54.18% 100% 100% 100% 100% 100% 100%
56% 55.16% 100% 100% 99% 99% 97% 96%
57% 56.15% 93% 89% 86% 83% 80% 77%
57.74% 56.87% 50% 50% 50% 50% 50% 50%
58% 57.13% 31% 34% 35% 37% 39% 40%
59% 58.12% 1% 2% 3% 5% 8% 11%
60% 59.10% 0% 0% 0% 0% 1% 1%
The Election Calculator Confirmed the LV poll / NEP new voter combination
(the 2004 unadjusted exit poll (Kerry by 52-47%) is assumed as the basis for calculating returning voter mix)
True Vote
Turnout 2004 Alive Voted Mix Obama McCain Other Obama McCain Other
DNV 20.77 15.8% 71% 27% 2% 14.74 5.61 0.42
95% Kerry 60.54 57.51 43.8% 89% 9% 2% 51.19 5.18 1.15
95% Bush 54.72 51.98 39.6% 17% 82% 1% 8.84 42.63 0.52
95% Other 1.16 1.11 0.8% 66% 24% 10% 0.73 0.27 0.11
Total 116.42 131.37 100.0% 57.47% 40.86% 1.67% 75.50 53.68 2.20
Recorded 131.37 52.87% 45.62% 1.51% 69.457 59.935 1.978
Diff 0.00 4.60% -4.76% 0.17% 6.04 -6.26 0.22