TruthIsAll
04-17-2009, 02:59 PM
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x509488
The naysayer claimed that Kerry's double-digit exit poll margins in MN, PA and NH
were implausible, but he chose to ignore these inescapable, fundamental factors:
1) Allocation of undecided voters (UVA)
2) Registered voter (RV) polls vs. likely voter (LV) subsets
3) Returning 2000 voters (equivalent to LV)
4) New voters (equivalent to RV - LV)
5) Uncounted votes
Polling analysis (factors 1,2) shows that:
The unadjusted state exit polls are close to the True Vote
Returning and new voter analysis (factors 3,4,5) shows that the calculated shares
are very close to True Vote.
The icing on the cake is that the pre-election RV and LV polls (factor 2) are
confirmed by returning and new voter shares. The mathematical proof is given below.
The UVA-adjusted registered voter (RV) samples virtually matched the exits.
Since the LVs are a subset of the RVs, they do not include newly registered voters.
It's interesting that Real Clear Politics (RCP) did not include one RV poll in
their final list of 15. And it was Deja vu in 2008, only this time the spread
between Obama's RV and LV average margin was greater (6%).
Kerry led in 8 of 9 Zogby pre-election battleground state polls by 50.2-44.8%.
The base case assumption was that he would capture 75% of the undecided (UVA) vote
and win all nine states by a 53.7-45.9% margin. The projection was within 0.5% of
the exit poll. The conservative assumption was that Kerry would win a 55% UVA.
In this case, he would win 8 states by 52.7-46.8%.
Kerry won just 4 states with a recorded share of 50.1-49.4%. The margin of error
was exceeded in 7 states - a one in 4.7 billion probability.
The Election Calculator True Vote Model determined that Kerry's PA share of 55.0%
was within 0.3% of the unadjusted exit poll. The True Vote was based on plausible
estimates of 2000 voter turnout in 2004 and the adjusted shares of returning 2000
and new 2004 voters. Kerry's vote shares were incremented above his 1222am NEP
shares since his recorded 51.0% share in PA exceeded his national 48.3% share.
In combination, these factors are a powerful confirmation that the UNADJUSTED exit
polls were quite accurate. On the other hand, despite a mountain of descriptive and
inferential statistical evidence, a few die-hards still disagree.
This is an analysis of the RV/LV 2004 national pre-election poll trends.
http://www.geocities.com/electionmodel/2004PreElectionNationalPolls.htm
Real Clear Politics (RCP) is the data source of 133 national polls (102 were LV
and 31 RV). The final 15 RCP polls were all LVs. The 2004 national pre-election
RV polls were closer to the True Vote than the LV subsets. After UVA adjustments,
the LV polls matched the RVs – and the unadjusted exit polls.
Kerry led the polls in August, fell behind in September but rebounded in October.
When undecided voters were allocated, Kerry was the projected winner.
Kerry's unadjusted RV average was 45.0-45.1.
His LV average was 45.6-48.1.
He led the UVA-adjusted RV (50.1-46.8).
Kerry led the final national 18 poll average (47.3-46.9).
He led the UVA-adjusted Election Model projection (50.9-48.1)
Bush led the final 15 RCP poll average (49.0-47.0).
He barely led the UVA-adjusted average (49.8-49.3).
There are plausible reasons why Kerry did better in the RV polls.
1- RV polls include newly registered voters.
2- The majority of new voters (57-62%) were Democrats.
3- There were 125.7m total votes CAST in 2004 (3.45m were uncounted).
4- There were 110.8m total votes CAST in 2000 (5.4m uncounted)
5- Approximately 5 million died and 3 million did not vote in 2004 (97% turnout).
6- An estimated 102.7 million voters returned in 2004.
7- Therefore, there were approximately 23m (125.7- 102.7) new voters in 2004.
8- Kerry won approximately 14m (60%) of new voters; Bush 9m.
The final 15 RCP poll national average had Bush leading by 49-47%. The Gallup poll
had Bush leading by 49-47. But Gallup assigned 90% of undecideds to Kerry and had
the race tied at 49-49%
Registered voter (RV) polls include all registered new voters; likely-voter (LV)
polls are a subset and exclude many newly registered. In general, only LV polls are
posted during the final two weeks before the election.
LV polls always understate the Democratic vote. RV samples are more accurate,
especially when there is a heavy turnout of new voters – as in 2004, 2006 and 2008.
The Census reported that 88.5% of registered voters turned out on 2004. The average
projected turnout of 5 final pre-election RV/LV polls was 82.8%. A regression
analysis of Kerry’s vote share vs. registered voter turnout indicated that he had a
52.6% share (assuming a 75% UVA). There was a 0.89 correlation between Kerry’s LV
poll share and LV/RV turnout.
Assuming a 50/50 split of the undecided vote, Kerry had 51.3%. In other words, the
pre-election polls underestimated voter turnout by 6%. Newly registered Democrats
came out in force. According to the 12:22am NEP, Kerry won new voters by
approximately 57-41%.
The calculations confirm that new voters comprise the difference (RV – LV) between
registered (RV) and likely voter (LV) sample-size. The Kerry / Bush share of the
difference was 57.8-42.2%, closely matching the 57-41% NEP new voter share.
The number of new voters is a function of voter mortality and turnout. It is
estimated by the simple formula:
New voters = current election voters– (prior election recorded vote – voter
mortality) * prior voter turnout
Applying the formula based on 2004 votes cast (125.7m) and 2000 votes cast (110.8m),
assuming 4.8% voter mortality and a 97% turnout of 2000 voters in 2004:
New 2004 voters = 23.5 million = 125.7 – (110.8 – 5.4)* 0.97
= 125.74 –105.4*.97= 125.74- 102.24
There were approximately 3.8 million returning Nader voters. Kerry won 2.4 million
(64%); Bush had 0.7 million (17%). According to the 1222am National Exit Poll,
Kerry won 13.4 million (57%) new voters; Bush had 9.4 million (41%).
Of the 27.3 million new and returning Nader voters, Kerry won 15.8m and Bush 10.1m.
Since Bush won the official recorded vote by 3 million, 8.7 million more returning
Gore voters must have defected than Bush voters. But the 1222am National Exit Poll
indicated that 10% of returning Bush voters defected to Kerry and 8% to Bush.
2004
Kerry-Bush
Pre-election Exit Recorded
MN 52-44 56-43 51.1-47.6
PA 50-45 55-44 50.9-48.4
NH 47-47 57-42 49.1-49.8
Adjusted LV Polls
Allocate 75% of the undecided vote to Kerry.
MN 54.3-44.7
PA 53.0-46.0
NH 50.7-48.3 (NH is a GOP fraud favorite: see Bush/Dole '88 and Clinton/Obama '08.
Conservatively, Kerry's RV poll average was 1.0% higher than the LV.
Adjusted RV Polls
MN 55.3-43.7
PA 54.0-45.0
NH 51.7-47.3
PA Vote Share Summary
Kerry Bush
Recorded 50.9 48.4
Exit Poll
WPE/IMS 55.1 44.2 (unadjusted)
GEO 56.9 43.1 (adjusted for turnout)
Composite 54.2 45.7 (12:40am, adjusted for pre-election polls)
Elec Calc
True Vote 55.0 44.0 (est. returning 2000 voters, adj. NEP vote shares)
_____________________________________________________________________________
PA Election Calculator
2000 Recorded
Gore Bush Other Total
2.485 2.281 0.120 4.886
50.9% 46.7% 2.5%
Uncounted
0.075 0.022 0.003 0.100
75% 22% 3% 2.00%
2004 Recorded
Kerry Bush Other Total
2.938 2.793 0.034 5.765
51.0% 48.4% 0.6%
Uncounted
0.057 0.019 0.000 0.076
75% 25% 0% 1.30%
* Kerry's estimated PA vote shares are higher than national shares
Uncounted Votes Kerry PA 12:22am Final
Pct Cast Unctd Share Est* NEP NEP
2004 1.30% 5.841 0.076 DNV 59% 57% 54%
2000 2.00% 4.986 0.100 Gore 93% 91% 90%
Bush 11% 10% 9%
Nader 66% 64% 71%
2004 Share 2000 Share
Kerry 75% Gore 75% Bush
Bush 25% Bush 22% Share
Other 0% Nader 3% DNV 40% 41% 45%
Gore 7% 8% 10%
2000 Voter Mortality Bush 89% 90% 91%
Total Voters 1.22% Nader 17% 17% 21%
Gore share 50%
Other
Share
2000 Voter Turnout in 2004 DNV 1% 2% 1%
Gore 98% Gore 0% 1% 0%
Bush 98% Bush 0% 0% 0%
Nader 98% Nader 17% 19% 8%
2000 Recorded
Voted Record Unct Cast Deaths Alive
Gore 2.49 0.07 2.56 0.12 2.44
Bush 2.28 0.02 2.30 0.12 2.19
Nader 0.12 0.00 0.12 0.01 0.12
Total 4.89 0.10 4.99 0.24 4.74
2004 PA True Vote
Voted Turnout Voted Weight Kerry Bush Other
DNV - 1.19 20.4% 59% 40% 1%
Gore 98% 2.39 40.9% 92% 8% 0%
Bush 98% 2.14 36.7% 11% 89% 0%
Nader 98% 0.11 2.0% 66% 17% 17%
Total 4.65 5.84 100% 55.02% 44.44% 0.54%
5.84 3.21 2.60 0.03
Exit Poll 55.3% 44.0% 0.7%
Recorded 51.0% 48.4% 0.6%
Sensitivity Analysis 1
Kerry PA Vote Share
(for incremental shares of returning Gore and new voters)
Kerry
Share
of Kerry share of New voters (DNV in 2000)
Gore 57.0% 58.0% 59.0% 60.0% 61.0%
Kerry Vote Share
94.0% 55.4% 55.6% 55.8% 56.0% 56.2%
93.0% 55.0% 55.2% 55.4% 55.6% 55.8%
92.0% 54.6% 54.8% 55.0% 55.2% 55.4%
91.0% 54.2% 54.4% 54.6% 54.8% 55.0%
90.0% 53.8% 54.0% 54.2% 54.4% 54.6%
Kerry Margin
94.0% 0.67 0.69 0.71 0.74 0.76
93.0% 0.62 0.64 0.67 0.69 0.71
92.0% 0.57 0.59 0.62 0.64 0.67
91.0% 0.52 0.55 0.57 0.59 0.62
90.0% 0.47 0.50 0.52 0.55 0.57
Sensitivity Analysis 2
Kerry PA Vote Share
(for incremental returning Gore and Bush voter turnout)
Bush
Voter Gore Voter Turnout
Turnout 96.0% 97.0% 98.0% 99.0% 100.0%
Kerry Vote Share
96.0% 55.1% 55.2% 55.4% 55.5% 55.7%
97.0% 54.9% 55.1% 55.2% 55.3% 55.5%
98.0% 54.7% 54.9% 55.0% 55.2% 55.3%
99.0% 54.6% 54.7% 54.8% 55.0% 55.1%
100.0% 54.4% 54.5% 54.7% 54.8% 54.9%
Kerry Margin
96.0% 0.63 0.64 0.66 0.68 0.69
97.0% 0.61 0.62 0.64 0.66 0.67
98.0% 0.59 0.60 0.62 0.63 0.65
99.0% 0.56 0.58 0.60 0.61 0.63
100.0% 0.54 0.56 0.58 0.59 0.61
_____________________________________________________________________________
Battleground Pre-election and Exit Polls
Average Kerry Bush
Zogby 50.2 44.8 (final pre-election polls)
Projection 53.7 45.9 (75% undecided to Kerry)
Exit(WPE) 53.2 45.8
Zogby 75% UVA 55% UVA Recorded
State** Kerry** Bush*** Kerry** Bush** Kerry** Bush*** Kerry**Bush*** Dev Prob
CO***** 47***** 48******50.3*** 49.1*** 49.4*** 50.0*** 47.4****52.1****2.97*** 3.27%
FL***** 50***** 47******51.9*** 47.6*** 51.4*** 48.2 47.3****52.3****4.65* 0.20%
IA***** 50***** 44******54.2*** 45.4*** 53.1*** 46.5*** 49.5****50.1****4.73*** 0.17%
ME***** 50***** 39******57.6*** 41.5*** 55.6*** 43.6*** 54.1****45.1****3.54*** 1.42%
MI***** 52***** 45******53.9*** 45.6*** 53.4*** 46.1*** 51.5****48.0****2.40*** 6.83%
MN***** 52***** 44******54.5*** 44.8*** 53.8*** 45.5*** 51.5****47.9****3.08*** 2.82%
OH***** 50***** 47******52.3*** 47.8*** 51.7*** 48.4*** 48.9****51.1****3.30*** 2.03%
PA***** 50***** 45******53.8*** 46.3*** 52.8*** 47.3*** 51.3****48.7****2.48*** 6.18%
WI***** 51***** 44******54.3*** 45.1*** 53.5*** 46.0*** 49.9****49.6****4.43*** 0.30%
********************************************************************************************************
Mean*** 50.2*** 44.8****53.7*** 45.9*** 52.7*** 46.8*** 50.1****49.5****3.51*** 2.58%
The naysayer claimed that Kerry's double-digit exit poll margins in MN, PA and NH
were implausible, but he chose to ignore these inescapable, fundamental factors:
1) Allocation of undecided voters (UVA)
2) Registered voter (RV) polls vs. likely voter (LV) subsets
3) Returning 2000 voters (equivalent to LV)
4) New voters (equivalent to RV - LV)
5) Uncounted votes
Polling analysis (factors 1,2) shows that:
The unadjusted state exit polls are close to the True Vote
Returning and new voter analysis (factors 3,4,5) shows that the calculated shares
are very close to True Vote.
The icing on the cake is that the pre-election RV and LV polls (factor 2) are
confirmed by returning and new voter shares. The mathematical proof is given below.
The UVA-adjusted registered voter (RV) samples virtually matched the exits.
Since the LVs are a subset of the RVs, they do not include newly registered voters.
It's interesting that Real Clear Politics (RCP) did not include one RV poll in
their final list of 15. And it was Deja vu in 2008, only this time the spread
between Obama's RV and LV average margin was greater (6%).
Kerry led in 8 of 9 Zogby pre-election battleground state polls by 50.2-44.8%.
The base case assumption was that he would capture 75% of the undecided (UVA) vote
and win all nine states by a 53.7-45.9% margin. The projection was within 0.5% of
the exit poll. The conservative assumption was that Kerry would win a 55% UVA.
In this case, he would win 8 states by 52.7-46.8%.
Kerry won just 4 states with a recorded share of 50.1-49.4%. The margin of error
was exceeded in 7 states - a one in 4.7 billion probability.
The Election Calculator True Vote Model determined that Kerry's PA share of 55.0%
was within 0.3% of the unadjusted exit poll. The True Vote was based on plausible
estimates of 2000 voter turnout in 2004 and the adjusted shares of returning 2000
and new 2004 voters. Kerry's vote shares were incremented above his 1222am NEP
shares since his recorded 51.0% share in PA exceeded his national 48.3% share.
In combination, these factors are a powerful confirmation that the UNADJUSTED exit
polls were quite accurate. On the other hand, despite a mountain of descriptive and
inferential statistical evidence, a few die-hards still disagree.
This is an analysis of the RV/LV 2004 national pre-election poll trends.
http://www.geocities.com/electionmodel/2004PreElectionNationalPolls.htm
Real Clear Politics (RCP) is the data source of 133 national polls (102 were LV
and 31 RV). The final 15 RCP polls were all LVs. The 2004 national pre-election
RV polls were closer to the True Vote than the LV subsets. After UVA adjustments,
the LV polls matched the RVs – and the unadjusted exit polls.
Kerry led the polls in August, fell behind in September but rebounded in October.
When undecided voters were allocated, Kerry was the projected winner.
Kerry's unadjusted RV average was 45.0-45.1.
His LV average was 45.6-48.1.
He led the UVA-adjusted RV (50.1-46.8).
Kerry led the final national 18 poll average (47.3-46.9).
He led the UVA-adjusted Election Model projection (50.9-48.1)
Bush led the final 15 RCP poll average (49.0-47.0).
He barely led the UVA-adjusted average (49.8-49.3).
There are plausible reasons why Kerry did better in the RV polls.
1- RV polls include newly registered voters.
2- The majority of new voters (57-62%) were Democrats.
3- There were 125.7m total votes CAST in 2004 (3.45m were uncounted).
4- There were 110.8m total votes CAST in 2000 (5.4m uncounted)
5- Approximately 5 million died and 3 million did not vote in 2004 (97% turnout).
6- An estimated 102.7 million voters returned in 2004.
7- Therefore, there were approximately 23m (125.7- 102.7) new voters in 2004.
8- Kerry won approximately 14m (60%) of new voters; Bush 9m.
The final 15 RCP poll national average had Bush leading by 49-47%. The Gallup poll
had Bush leading by 49-47. But Gallup assigned 90% of undecideds to Kerry and had
the race tied at 49-49%
Registered voter (RV) polls include all registered new voters; likely-voter (LV)
polls are a subset and exclude many newly registered. In general, only LV polls are
posted during the final two weeks before the election.
LV polls always understate the Democratic vote. RV samples are more accurate,
especially when there is a heavy turnout of new voters – as in 2004, 2006 and 2008.
The Census reported that 88.5% of registered voters turned out on 2004. The average
projected turnout of 5 final pre-election RV/LV polls was 82.8%. A regression
analysis of Kerry’s vote share vs. registered voter turnout indicated that he had a
52.6% share (assuming a 75% UVA). There was a 0.89 correlation between Kerry’s LV
poll share and LV/RV turnout.
Assuming a 50/50 split of the undecided vote, Kerry had 51.3%. In other words, the
pre-election polls underestimated voter turnout by 6%. Newly registered Democrats
came out in force. According to the 12:22am NEP, Kerry won new voters by
approximately 57-41%.
The calculations confirm that new voters comprise the difference (RV – LV) between
registered (RV) and likely voter (LV) sample-size. The Kerry / Bush share of the
difference was 57.8-42.2%, closely matching the 57-41% NEP new voter share.
The number of new voters is a function of voter mortality and turnout. It is
estimated by the simple formula:
New voters = current election voters– (prior election recorded vote – voter
mortality) * prior voter turnout
Applying the formula based on 2004 votes cast (125.7m) and 2000 votes cast (110.8m),
assuming 4.8% voter mortality and a 97% turnout of 2000 voters in 2004:
New 2004 voters = 23.5 million = 125.7 – (110.8 – 5.4)* 0.97
= 125.74 –105.4*.97= 125.74- 102.24
There were approximately 3.8 million returning Nader voters. Kerry won 2.4 million
(64%); Bush had 0.7 million (17%). According to the 1222am National Exit Poll,
Kerry won 13.4 million (57%) new voters; Bush had 9.4 million (41%).
Of the 27.3 million new and returning Nader voters, Kerry won 15.8m and Bush 10.1m.
Since Bush won the official recorded vote by 3 million, 8.7 million more returning
Gore voters must have defected than Bush voters. But the 1222am National Exit Poll
indicated that 10% of returning Bush voters defected to Kerry and 8% to Bush.
2004
Kerry-Bush
Pre-election Exit Recorded
MN 52-44 56-43 51.1-47.6
PA 50-45 55-44 50.9-48.4
NH 47-47 57-42 49.1-49.8
Adjusted LV Polls
Allocate 75% of the undecided vote to Kerry.
MN 54.3-44.7
PA 53.0-46.0
NH 50.7-48.3 (NH is a GOP fraud favorite: see Bush/Dole '88 and Clinton/Obama '08.
Conservatively, Kerry's RV poll average was 1.0% higher than the LV.
Adjusted RV Polls
MN 55.3-43.7
PA 54.0-45.0
NH 51.7-47.3
PA Vote Share Summary
Kerry Bush
Recorded 50.9 48.4
Exit Poll
WPE/IMS 55.1 44.2 (unadjusted)
GEO 56.9 43.1 (adjusted for turnout)
Composite 54.2 45.7 (12:40am, adjusted for pre-election polls)
Elec Calc
True Vote 55.0 44.0 (est. returning 2000 voters, adj. NEP vote shares)
_____________________________________________________________________________
PA Election Calculator
2000 Recorded
Gore Bush Other Total
2.485 2.281 0.120 4.886
50.9% 46.7% 2.5%
Uncounted
0.075 0.022 0.003 0.100
75% 22% 3% 2.00%
2004 Recorded
Kerry Bush Other Total
2.938 2.793 0.034 5.765
51.0% 48.4% 0.6%
Uncounted
0.057 0.019 0.000 0.076
75% 25% 0% 1.30%
* Kerry's estimated PA vote shares are higher than national shares
Uncounted Votes Kerry PA 12:22am Final
Pct Cast Unctd Share Est* NEP NEP
2004 1.30% 5.841 0.076 DNV 59% 57% 54%
2000 2.00% 4.986 0.100 Gore 93% 91% 90%
Bush 11% 10% 9%
Nader 66% 64% 71%
2004 Share 2000 Share
Kerry 75% Gore 75% Bush
Bush 25% Bush 22% Share
Other 0% Nader 3% DNV 40% 41% 45%
Gore 7% 8% 10%
2000 Voter Mortality Bush 89% 90% 91%
Total Voters 1.22% Nader 17% 17% 21%
Gore share 50%
Other
Share
2000 Voter Turnout in 2004 DNV 1% 2% 1%
Gore 98% Gore 0% 1% 0%
Bush 98% Bush 0% 0% 0%
Nader 98% Nader 17% 19% 8%
2000 Recorded
Voted Record Unct Cast Deaths Alive
Gore 2.49 0.07 2.56 0.12 2.44
Bush 2.28 0.02 2.30 0.12 2.19
Nader 0.12 0.00 0.12 0.01 0.12
Total 4.89 0.10 4.99 0.24 4.74
2004 PA True Vote
Voted Turnout Voted Weight Kerry Bush Other
DNV - 1.19 20.4% 59% 40% 1%
Gore 98% 2.39 40.9% 92% 8% 0%
Bush 98% 2.14 36.7% 11% 89% 0%
Nader 98% 0.11 2.0% 66% 17% 17%
Total 4.65 5.84 100% 55.02% 44.44% 0.54%
5.84 3.21 2.60 0.03
Exit Poll 55.3% 44.0% 0.7%
Recorded 51.0% 48.4% 0.6%
Sensitivity Analysis 1
Kerry PA Vote Share
(for incremental shares of returning Gore and new voters)
Kerry
Share
of Kerry share of New voters (DNV in 2000)
Gore 57.0% 58.0% 59.0% 60.0% 61.0%
Kerry Vote Share
94.0% 55.4% 55.6% 55.8% 56.0% 56.2%
93.0% 55.0% 55.2% 55.4% 55.6% 55.8%
92.0% 54.6% 54.8% 55.0% 55.2% 55.4%
91.0% 54.2% 54.4% 54.6% 54.8% 55.0%
90.0% 53.8% 54.0% 54.2% 54.4% 54.6%
Kerry Margin
94.0% 0.67 0.69 0.71 0.74 0.76
93.0% 0.62 0.64 0.67 0.69 0.71
92.0% 0.57 0.59 0.62 0.64 0.67
91.0% 0.52 0.55 0.57 0.59 0.62
90.0% 0.47 0.50 0.52 0.55 0.57
Sensitivity Analysis 2
Kerry PA Vote Share
(for incremental returning Gore and Bush voter turnout)
Bush
Voter Gore Voter Turnout
Turnout 96.0% 97.0% 98.0% 99.0% 100.0%
Kerry Vote Share
96.0% 55.1% 55.2% 55.4% 55.5% 55.7%
97.0% 54.9% 55.1% 55.2% 55.3% 55.5%
98.0% 54.7% 54.9% 55.0% 55.2% 55.3%
99.0% 54.6% 54.7% 54.8% 55.0% 55.1%
100.0% 54.4% 54.5% 54.7% 54.8% 54.9%
Kerry Margin
96.0% 0.63 0.64 0.66 0.68 0.69
97.0% 0.61 0.62 0.64 0.66 0.67
98.0% 0.59 0.60 0.62 0.63 0.65
99.0% 0.56 0.58 0.60 0.61 0.63
100.0% 0.54 0.56 0.58 0.59 0.61
_____________________________________________________________________________
Battleground Pre-election and Exit Polls
Average Kerry Bush
Zogby 50.2 44.8 (final pre-election polls)
Projection 53.7 45.9 (75% undecided to Kerry)
Exit(WPE) 53.2 45.8
Zogby 75% UVA 55% UVA Recorded
State** Kerry** Bush*** Kerry** Bush** Kerry** Bush*** Kerry**Bush*** Dev Prob
CO***** 47***** 48******50.3*** 49.1*** 49.4*** 50.0*** 47.4****52.1****2.97*** 3.27%
FL***** 50***** 47******51.9*** 47.6*** 51.4*** 48.2 47.3****52.3****4.65* 0.20%
IA***** 50***** 44******54.2*** 45.4*** 53.1*** 46.5*** 49.5****50.1****4.73*** 0.17%
ME***** 50***** 39******57.6*** 41.5*** 55.6*** 43.6*** 54.1****45.1****3.54*** 1.42%
MI***** 52***** 45******53.9*** 45.6*** 53.4*** 46.1*** 51.5****48.0****2.40*** 6.83%
MN***** 52***** 44******54.5*** 44.8*** 53.8*** 45.5*** 51.5****47.9****3.08*** 2.82%
OH***** 50***** 47******52.3*** 47.8*** 51.7*** 48.4*** 48.9****51.1****3.30*** 2.03%
PA***** 50***** 45******53.8*** 46.3*** 52.8*** 47.3*** 51.3****48.7****2.48*** 6.18%
WI***** 51***** 44******54.3*** 45.1*** 53.5*** 46.0*** 49.9****49.6****4.43*** 0.30%
********************************************************************************************************
Mean*** 50.2*** 44.8****53.7*** 45.9*** 52.7*** 46.8*** 50.1****49.5****3.51*** 2.58%