RSS

Monthly Archives: September 2011

Unadjusted 2008 State Exit Polls: Further Confirmation of the True Vote Model

Richard Charnin (TruthIsAll)

Sept. 20, 2011

It is instructive to see how the unadjusted 2008 exit polls polls compare to the recorded vote and the True Vote Model (TVM). The basic results are not surprising: Obama did better in the aggregate state exit polls (58.1%) than the vote count (52.9%). But the Democrats always do better in the polls. What is surprising is that he did 5.2% better – exactly matching the TVM. By way of comparison, Kerry did 3.7% better in the unadjusted exit polls (52%) than in the recorded vote (48.3%). He had 53.6% in the TVM.

A Triple Confirmation

In the 2008 National Exit Poll (NEP), 4178 of the 17836 responders were asked how they voted in 2004: 1815 (43.4%) said they were Kerry voters, 1614 (38.6%) Bush, 188 (4.5%) third-party and 561 (13.4%) did not vote. Applying Final 2008 NEP vote shares to the returning voter mix, Obama had a 58.1% share – exactly matching a) his 58.1% share of the aggregate unadjusted state exit polls and b) his 58.1% TVM share! The returning voter mix implied that Kerry won by 50.2-44.6%.

But all exit polls are forced to match the recorded vote. The pollsters needed an impossible 46/37% Bush/Kerry mix which implied that Bush won by 52.6-42.3%. His (bogus) recorded margin was 50.7-48.3%. Kerry won the True Vote with 53.6% (Table 6). In the Final 2008 NEP, pollsters effectively converted 269 of 1815 (15%) Kerry responders to Bush responders in order to force a match to the recorded vote.

To summarize, the unadjusted 2008 NEP exactly matched the weighted aggregate share of the unadjusted state exit polls, based on how the the exit poll responders said they voted in 2004 and 2008. It also matched the TVM which used 2004 votes cast, voter mortality, a best estimate of living 2004 voter turnout in 2008 – and the Final 2008 NEP vote shares. Obama had 58.1% in each calculation – a triple confirmation that Obama won a 23 million vote landslide, far exceeding his 9.5 million recorded vote margin.

But that’s not all. The National Exit Poll of 17836 respondents is a subset of the 80,000 sampled in the state exit polls. Obama won the unadjusted National Exit Poll by 61-37%, a landslide of historic proportions. However, the state exit polls have a smaller margin of error and are probably a better estimate of the True Vote.

This graph shows that Obama’s 58% True Vote share is confirmed by three independent statistical measures: 1) Unadjusted National Exit Poll, 2) Unadjusted state exit polls, 3) and 10 million late (paper ballot) votes.

The key result is the state exit poll aggregate vote share. The national sample size was approximately 80,000. The average state exit poll margin of error was 3.35% (including a 30% “cluster” effect). The margin of error was exceeded in 37 states; in 2004 it was exceeded in 29. Of the 50 states and DC, 45 shifted to McCain from the exit poll. The difference in margin between the exit poll and the recorded vote is the average Within Precinct Discrepancy (WPD). The WPD was 10.6 in 2008, far above the 7.4 in 2004.

The True Vote Model has closely matched the unadjusted state and national exit polls in every presidential election since 1988. In the 11 presidential elections from 1968 to 2008, the Republicans had a 49-45% recorded vote margin while the Democrats had a 49-45% True Vote margin.

In a given state, the exit poll varies from the corresponding True Vote calculation. But the total aggregate share is an exact match, illustrating the Law of Large Numbers and the Central Limit Theorem.

The National True Vote Model is based on previous election votes cast and turnout of previous election voters, current votes cast and National Exit Poll (NEP) vote shares. The State Model works the same way. It’s based on returning state voters with NEP vote shares adjusted according to the state/national vote share ratio.

It should be obvious by now that final weighting adjustments made to the exit polls are made to match the recorded vote. In 2004, in addition to the impossible return voter mix, the 12:22am preliminary national exit poll vote shares had to be adjusted in the Final NEP. The required turnout of living Bush voters was 110%. Kerry had a 52.0% aggregate share and a 53.6% TVM share. Of course, all demographic categories had to be adjusted to match the vote count: Final NEP “Party-ID”, “When Decided” and “Bush Approval” crosstab weights did not match the corresponding pre-election polls and were adjusted to force a match to the recorded vote.

 
1 Comment

Posted by on September 20, 2011 in 2008 Election

 

Tags: , , ,

Footprints of Systemic Election Fraud: 1988-2008 State Exit Poll Discrepancies

Richard Charnin (TruthIsAll)

Updated: Sept.9, 2012

This is an updated analysis of state and national exit poll discrepancies in the 1988-2008 presidential elections. The unadjusted data has made available on the Roper Center for Public Opinion (UConn) website. Now we know what the respondents actually said as to how they voted. It is fundamental information that was not previously available. But it is not the raw precinct level data that analysts would love to see and which the corporate media (the National Election Pool) will not release.

Nevertheless, the unadjusted state and national exit poll data is the mother-lode for SERIOUS exit poll analysis. The pattern is clear: the Democrats always do better in the polls than in the recorded count. There is no evidence that this one-sided result is due to anything other than vote miscounts.

Each presidential election consists of 50 state exit poll files (and Washington DC) in PDF format. In order to utilize the data for a meaningful analysis, it had to be re-organized and consolidated in a single workbook. The workbook contains individual worksheets for each election, as well as other sheets for relevant graphs and tables.

This graph summarizes the discrepancies between the 1988-2008 State Exit Polls vs. the corresponding Recorded Votes

It has long been established that Final National Exit Polls are always forced to match the recorded vote, often with impossible returning voter weights. The unadjusted data shows just how the exit pollsters had to adjust the actual responses to force the match. Furthermore, and most important, it confirms True Vote Model calculations in each election. The pattern of massive discrepancies totally confirm that the adjusted Final National Election Poll is fiction and debunks the corresponding myth that elections are fair and that the votes are counted accurately.

The original post was based on 1988-2004 data from the Edison/Mitofsky 2004 Election Evaluation Report.

1988-2008 Unadjusted Exit Polls

According to the unadjusted state and national exit polls and the True Vote Model, the Democrats won the 1988-2008 popular vote by a far bigger margin than the recorded vote indicates.

1988-2008 Average National Presidential Vote Shares
....Measure........Dem...Rep...Margin....Note
1) Recorded : 47.9-45.9% (2.0%) - Vote count
2) WPE / IMS : 50.8-43.1% (7.7%) - Edison-Mitofsky
3) State Exit : 51.8-41.6% (10.2%) - Roper
4) National Exit: 51.7-41.7% (10.0%) - Roper
5) True Vote 1 : 50.2-43.8% (6.4%) - previous recorded vote
6) True Vote 2 : 51.6-42.5% (9.1%) - previous votes cast
7) True Vote 3 : 52.5-41.5% (11.0%) - previous unadjusted exit poll
8) True Vote 4 : 53.0-41.0% (12.0%) - previous True Vote

1988-2008: 274 STATE EXIT POLLS

PROOF OF SYSTEMIC ELECTION FRAUD BEYOND ANY DOUBT

This table illustrates the one-sided red-shift from the Democrat in the state exit polls to the Republican in the recorded vote. The margin of error includes a 30% cluster effect. The MoE was exceeded in an astounding 126 of 274 state presidential exit polls from 1988-2008. The probability is ZERO. At the 95% confidence level, we would expect 14 polls to exceed the MoE. Of the 126 elections, 123 red-shifted to the GOP and just 3 to the Democrat. The probability is 5.4E-106 – ZERO.

State Exit Poll Margin of Error

......................Total..1988...1992..1996..2000..2004..2008
....................... 3.26% 3.34% 3.42% 3.07% 3.64% 3.11% 2.97%
Exit Polls:
red-shift to GOP........226 20 44 43 34 40 45
exceeding MoE...........126 11 26 16 13 23 37
exceeding MoE (GOP).....123 11 26 16 12 22 36

Probability of..........Average.1988.....1992....1996....2000....2004....2008
126 exceeding MoE.......8.0E-75 6.6E-09 2.1E-15 1.5E-09 7.5E-07 2.1E-15 2.1E-15
123 exceeding MoE (GOP).5.E-106 5.0E-11 2.4E-25 4.8E-13 8.7E-09 3.5E-20 2.4E-39
226 red-shift to GOP....3.7E-31 7.7E-04 1.6E-08 1.0E-07 7.7E-03 1.2E-05 2.1E-09

States in which the Democrats won the exit poll and lost the 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.

In 1988, Dukakis won the unadjusted National Exit Poll (11,586 respondents) by 49.8-49.2%. He won the exit polls in the battleground states by 51.6-47.3%. There were 11 million uncounted votes, an indicator that Dukakis may have won since 70-80% of uncounted votes are Democratic. But he lost by 7 million recorded votes (53.4-45.6%).

In 1992, Clinton won the unadjusted state exit polls (54,000 respondents) by 18 million votes (47.6-31.7%). He won the unadjusted National Exit Poll (15,000 respondents)by 46.3-33.4%. He had 51% in the True Vote Model (TVM). But his recorded margin was just 5.6 million (43.0-37.5%). The Final National Exit Poll (NEP) was forced to match the recorded vote. The NEP implied that there was a 119% turnout of living 1988 Bush voters. There were 10 million uncounted votes. The landslide was denied.

In 1996, Clinton won the unadjusted exit polls (70,000 respondents) by 16 million votes (52.6-37.1%). He had 53.6% in the TVM. His recorded margin was 8 million (49.2-40.8%). The Final National Exit Poll (NEP) was forced to match the recorded vote. There were 10 million uncounted votes. The landslide was denied.

In 2000, Gore won the unadjusted state exit polls (58,000 respondents) by 6 million votes (50.8-44.4%). He had 51.5% in the TVM. But he won the recorded vote by just 540,000. There were 6 million uncounted votes. The election was stolen.

In 2004, Kerry won the unadjusted state exit poll aggregate (76,000 respondents) by 51.1-47.5%. He won the unadjusted National Exit Poll (13,660 respondents) by 51.7-47.0%, a 6 million vote margin. He had 53.6% (a 10 million margin) in the True Vote Model But he lost by 3.0 million recorded votes. There were 4 million uncounted votes. The election was stolen.

In 2008, Obama won the unadjusted state exit poll aggregate (83,000 respondents) by 58.1-40.3%, a 23 million vote margin – a near-exact match to the TVM. He won the unadjusted National Exit Poll (17,836 respondents) by a whopping 61-37%. Officially, he had 52.9% and won by 9.5 million votes. The landslide was denied.

http://richardcharnin.wordpress.com/2011/11/13/1988-2008-unadjusted-state-exit-polls-statistical-reference/

___________________________________________________________________________

http://richardcharnin.com/StateExitPollDiscrepancies.htm

The full set of 2008 exit polls and 24 of the 1988 state polls are from the Roper website. The analysis is displayed in the following 12 data tables:

1988-2008
1 State Exit Poll Discrepancies

1988
2 True Vote Model vs. Final National Exit Poll
3 Battleground Exit Polls vs. Recorded Vote

2004
4 National Exit Poll
5 True Vote Model
6 Sensitivity Analysis
7 State Recorded, Exit Poll, True Vote Shares, 8 State Exit Poll Timeline

2008
9 National Exit Poll
10 True Vote Model
11 Unadjusted State exit polls vs. Recorded Vote and True Vote
12 Unadjusted National Exit Poll vs. Final

Within Precinct Error (WPE) is the difference between the unadjusted exit poll and recorded vote margins. “Error” implies that the exit polls were wrong and the election was fraud-free. But millions of votes are uncounted in every election (nearly 11 million in 1988 and 4 million in 2004). Therefore, it is more accurate to refer to Within Precinct Discrepancy (WPD). A positive WPD indicates that the vote shift favored the GOP; a negative WPD favored the Democrat. In 2004, Kerry won the state exit polls by 52-47% but lost the recorded vote by 50.7-48.3%, a WPD of 7.4%.

In the 274 state elections which were exit polled, 226 shifted from the exit poll to the Republican and 48 shifted to the Democrat. The one-sided red-shift to the Republican implies that the exit polls were incorrect or the votes were miscounted. It could not have been due to chance. Exit polls are known to be quite accurate – outside the USA.

Were the discrepancies due to Republican voter reluctance to be polled in each of the six elections? Not likely. Were they due to Democratic voters misstating how they voted to the exit pollsters in each of the six elections? Not likely. Or were they due to the millions of mostly Democratic votes that were uncounted? That is more than likely. It’s a fact. Were they due to votes that were miscounted in favor of the Republican? Quite likely.

- In 15 Democratic states, the average WPD was 6.3. The MoE was exceeded in 41 state elections. All shifted in favor of the Republicans.
– In 15 Battleground states, the average WPD was 5.0. The MoE was exceeded in 37. All shifted in favor of the Republicans
– In 21 Republican states, the average WPD was 3.7. The MoE was exceeded in 35. All but two shifted in favor of the Republicans.

Given a 95% level of confidence, approximately 14 of 274 elections would be expected to fall outside the margin of error. The probability that the MoE would be exceeded in a state is 5%. But the MoE was exceeded in 126 elections, all but threein favor of the Republicans. The probability is ZERO that this was due to chance.

1988
The 1988 CBS exit poll indicate that Dukakis did substantially better than the Edison/Mitofsky report. They show Dukakis winning the 24 battleground state aggregate by a solid 51.6-47.3%. But George H.W. Bush won the recorded vote by 53.4-45.6%. There were 68.7 million recorded votes in the battleground states (75% of the 91.6 million recorded). Seven of the 24 flipped to Bush from the exit polls – a total of 132 electoral votes: CA, MD, PA, MI, IL, VT and NM. The margin of error was exceeded in 11 of the 24 states. Dukakis may very well have won the election. According to the Census, there were at least 10.6 million net uncounted votes (i.e. net of stuffed ballots).

Dukakis won the Roper California exit poll in a landslide (57.7-40.8%), yet Bush won the recorded vote (51.1-47.7%) – an amazing 20.4% discrepancy. He won the IL exit poll by 8% but lost by 2%. In MI, Dukakis had a 3.5% exit poll margin and lost by 8%. In MD, his 12% exit poll win morphed into a 3% defeat. In PA, he won the exit poll by less than 1% and lost by 3%. In Bush’s home state of Texas, he barely edged Dukakis by 1% in the exit poll. He won the state by 13%.

1988 Battleground State Exit Polls
http://richardcharnin.com/1988RoperExit_16115_image001.gif

2004
– In 15 strong Democratic states, the average WPD was 8.9.
The MoE was exceeded in 11 states (73%) – all shifted to Bush.
– In 15 Battleground states, the average WPD was 6.9.
The MoE was exceeded in 10 states (67%) – all shifted to Bush.
– In 21 Republican states, the average WPD was 3.8.
The MoE was exceeded in 7 states (33%) – all shifted to Bush.

The margin of error was exceeded in a total of 23 states – all but one in favor of Bush. The probability is 1 in 19 trillion that the MoE would be exceeded in 16 states. Imagine what the probability is for 28 states. Assuming a 2% MoE, the probability is even lower since the MoE was exceeded in 36 states: 34 in favor of Bush, 2 in favor of Kerry.

The WPDs indicate the GOP election theft strategy:
1) Cut Dem margins in BLUE states: NY, CA, CT, NJ, MD, MA, MI
2) Steal BLUE battleground states: FL, OH, NM, CO, NV, MO, IA
3) Pad the Bush vote in big RED states: TX, MS, AL, TN, SC
4) Ignore small RED states: ND, SD, OK, MT, KY

2008
The exit poll discrepancies (10.6 WPD) were substantially greater than in other elections. The True Vote Model (TVM) exactly matched Obama’s 58% aggregate share of the unadjusted state exit polls – a 23 million vote margin. McCain’s recorded share exceeded his exit poll in 45 states. The exit poll margin of error was exceeded in 37 states, all but one in favor ofr McCain. Obama won by nearly 23 million True votes; he won officially by 9.5 million.

2008 Unadjusted State Exit Polls confirm the True Vote Model:
http://richardcharnin.com/2008ExiPollConfirmationTVM.htm

This graph tells you all you need to know about the 2008 election. Obama had a 58% True Vote share – not the official recorded 53%. This is confirmed by at least 4 independent statistical measures: 1) Unadjusted National Exit Poll, 2) Unadjusted state exit polls, 3) True Vote Model and 4)10 million late (paper ballot) votes.

http://richardcharnin.com/2008NEPUnadjustedRoper_28080_image001.gif

 
3 Comments

Posted by on September 16, 2011 in 2004 Election

 

Tags: , , ,

Monte Carlo Simulation: Election Forecasting and Exit Poll Modeling

Richard Charnin

Updated: July 8, 2012

2004 Monte Carlo Electoral Vote Simulation (pre-election and  exit polls)

The simulation model consists of 200 election trials based on pre-election state polls and post-election exit polls. It is strong circumstantial evidence that the election was stolen.

In the pre-election model, the state and national polls are adjusted for the allocation of undecided voters. The post-election model is based on unadjusted and adjusted state exit polls. Monte Carlo simulation is used to project state and aggregate vote shares and calculate the popular and electoral vote win probabilities.

The state win probability is a function of 1) the projected vote shares (after allocating undecided voters) and 2) the state poll margin of error.

The expected (theoretical) electoral vote can be calculated using a simple summation formula. It is just the product sum of the state win probabilities and corresponding electoral votes.

The purpose of the simulation is to calculate the overall probability of winning the electoral vote. As the number of election trials increase, the average (mean) electoral vote will approach the theoretical expected value.

The electoral vote win probability is the ratio of the number of winning election trials to the total number of trials.

In every presidential election, millions of voters are disenfranchised and millions of votes are uncounted. Forecasting models should have the following disclaimer:

Note: The following forecast will surely deviate from the official recorded vote. If they are nearly equal, then there must have been errors in the a) input data, b) assumptions, c) model logic and/or methodology.

Kerry led the weighted pre-election state and national polls by 1%. After allocating 75% of undecided voters to him, he was projected to win by 51.4-47.7%. Kerry had 51.1% in the unadjusted state exit poll aggregate (76,000 respondents) and 51.7% in the unadjusted National Exit Poll (13,660 respondents).

The National Election Pool, a consortium of six media giants, funds the exit polls. The published National Exit Poll is always forced to match the recorded vote. The Final 2004 NEP adjusted the actual exit poll responses to force a match to the recorded vote (Bush by 50.7-48.3%).

The large discrepancy between the exit polls and the vote count indicates that either a) the pre-election and unadjusted exit polls were faulty or b) the votes were miscounted, or c) a combination of both. Other evidence confirms that the votes were miscounted in favor of Bush.

The True Vote always differs from the official recorded vote due to uncounted, switched and stuffed ballots. Were the pollsters who forecast a Bush win correct? Or were Zogby and Harris correct in projecting that Kerry would win?

None of the pollsters mentioned the election fraud factor – the most important variable of all.

MODEL OVERVIEW

The workbook contains a full analysis of the 2004 election, based on four sets of polls:

(1) Pre-election state polls
(2) Pre-election national polls
(3) Post-election state exit polls
(4) National Exit Poll

Click the tabs at the bottom of the screen to select:
MAIN: Data input and summary analysis.
SIMULATION: Monte Carlo Simulation of state pre-election and exit polls.
NATPRE: Projections and analysis of 18 national pre-election polls.
In addition, three summary graphs are provided in separate sheets.

Calculation methods and assumptions are entered in the MAIN sheet:
1) Calculation code: 1 for pre-election polls; 2 for EXIT polls.
2) Undecided voter allocation (UVA): Kerry’s share (default 75%).
3) Exit Poll Cluster Effect: increase in margin of error (default 30%).
4) State Exit Poll Calculation Method:
1= WPD: average precinct discrepancy.
2= Best GEO: adjusted based on recorded vote geographic weightings.
3= Composite: further adjustment to include pre-election polls.
4= Unadjusted state exit polls

Note: The Composite state exit poll data set (12:40am) was downloaded from the CNN election site by Jonathan Simon. The polls were in the process of being adjusted to the incoming vote counts and weighted to include pre-election polls.The final adjustment at 1am forced a match to the final recorded votes.

2004 ELECTION MODEL

The Election Model was executed weekly from August to the election. It tracked state and national polls which were input to a 5000 trial Monte Carlo simulation. The final Nov. 1 forecast had Kerry winning 51.8% of the two-party vote and 337 electoral votes> He had a 99.8% electoral vote win probability: the percentage of trials in which he had at least 270 electoral votes.

Simulation forecast trends are displayed in the following graphs:

State aggregate poll trend
Electoral vote and win probability
Electoral and popular vote
Undecided voter allocation impact on electoral vote and win probability
National poll trend
Monte Carlo Simulation
Monte Carlo Electoral Vote Histogram

POLL SAMPLE-SIZE AND MARGIN OF ERROR

Approximately 600 were surveyed in each of the state pre-election polls (a 4% margin of error). The national aggregate has a lower MoE; approximately 30,000 were polled. In 18 national pre-election polls, the samples ranged from 800 (3.5% MoE) to 3500 (1.7% MoE).

In the exit polls, 76,000 voters were sampled. Kerry won the unadjusted state exit poll aggregate by 51.1-47.5%. He also won the unadjusted National Exit Poll (NEP) by 51.7-47.0%. The NEP is a 13,660 sample subset of the state exit polls.The NEP was adjusted to match the recorded vote -using the same 13660 respondents.

Assuming a 30% exit poll “cluster effect” (1.1% MoE), Kerry had a 98% probability of winning the popular vote. The Monte Carlo simulation indicates he had better than a 99% probability of winning the Electoral Vote.

The “cluster effect” is the percentage increase in the theoretical Exit Poll margin of error. When it is not practical to carry out a pure random sample, a common shortcut is to use an area cluster sample: primary Sampling Units (PSUs) are selected at random within the larger geographic area.

The Margin of Error (MoE) is a function of the sample size (n) and the polling percentage split: MoE = 1.96* Sqrt(P*(1-P)/n)

ELECTION FORECASTING METHODOLOGY

The Law of Large Numbers is the basis for statistical sampling. All things being equal, polling accuracy is directly related to sample size – the larger the sample, the smaller the margin of error (MoE). In an unbiased random sample, there is a 95% probability that the vote will fall within the MoE of the mean.

There are two basic methods used to forecast presidential elections:
1) Projections based on state and national polls
2) Time-series regression models

Academics and political scientists create multiple regression models to forecast election vote shares and run the models months in advance of the election. The models utilize time-series data, such as: economic growth, inflation, job growth, interest rates, foreign policy, historical election results, incumbency, approval rating, etc. Regression modeling is an interesting theoretical exercise, but it does not account for daily events which affect voter psychology.

Polling and regression models are analogous to the market value of a stock and its intrinsic (theoretical) value. The latest poll share is the equivalent of the current stock price. The intrinsic value of a stock is based on forecast cash flows. The intrinsic value is rarely equal to the market value.

The historical evidence is clear: state and national polls, adjusted for undecided voters and estimated turnout, are superior to time-series models executed months in advance.

Inherent problems exist in election models, the most important of which is never discussed: Election forecasters and media pundits never account for the probability of fraud. The implicit assumption is that the official recorded vote will accurately reflect the True Vote and that the election will be fraud-free.

MONTE CARLO SIMULATION

Monte Carlo is a random process of repeated experimental “trials” applied to a mathematical system model. The Election Simulation Model runs 200 trial “elections” to determine the expected electoral vote and win probability.

Statistical polling (state and national) ideally is an indicator of current voter preference. Pre-election poll shares are adjusted for undecided voters and state win probabilities are calculated. The probabilities are input to a Monte Carlo simulation based on random numbers. The final probability of winning the electoral vote is simply the number of winning election trials divided by the total number of trials (200 in the ESM; 5000 in the Election Model).

The only forecast assumption is the allocation of undecided/other voters. Historically, 70-80% of undecided voters break for the challenger. If the race is tied at 45-45, a 60-40% split of undecided voters results in a 51-49% projected vote share.

ELECTORAL AND POPULAR VOTE WIN PROBABILITIES

The theoretical expected electoral vote for a candidate is a simple calculation. It is just the sum of the 51 products: state electoral vote times the win probability. In the simulation, the average (mean) value will converge to the theoretical value as the number of election trial increase.

The probability of winning the popular vote is a function of the projected 2-party vote share and polling margin of error. These are input to the Excel normal distribution function. The simulation generates a electoral vote win probability that is not sensitive to minor changes in the state polls.

Prob (win) = NORMDIST (P, 0.50, MoE/1.96, True)

For each state in an election trial, a random number (RND) between 0 and 1 is generated and compared to the probability of winning the state. For example, if Kerry has a 90% probability of winning Oregon and RND is less than 0.90, Kerry wins 7 electoral votes. Otherwise, if RND is greater than 0.90, Bush wins. The procedure is repeated for all 50 states and DC. The election trial winner has at least 270 EV.

The electoral vote win probability is directly correlated to the probability of winning the national popular vote. But electoral vote win probabilities in models developed by academics and bloggers are often incompatible with the projected national vote shares.

For example, assume a 53% projected national vote share. If the corresponding EV win probability is given 88%, the model design/logic is incorrect; the 53% share and 88% win probability are incompatible. For a 53% share, the win probability is virtually 100%. This is proved using Monte Carlo simulation based on state win probabilities in which there is a 53% aggregate projected national share.

The state win probability is based on the final pre-election polls which typically sample 600 likely voters (a 4% MoE). In the 2004 Election Simulation Model, the electoral vote is calculated using 200 election trials. The average (mean) electoral vote is usually within a few votes of the median (middle value). As the number of simulation trials increase, the mean approaches the theoretical expected value. That is due to the Law of Large Numbers.

SENSITIVITY ANALYSIS

A major advantage of the Monte Carlo Simulation method is that the win probability is not sensitive to minor deviations in the state polls. It is not an all-or-nothing proposition as far as allocating the electoral vote is concerned. A projected 51% vote share has less electoral “weight” than a 52% share, etc. Electoral vote projections from media pundits and Internet bloggers use a single snapshot of the latest polls to determine a projected electoral vote split. This can be misleading when the states are competitive and often results in wild electoral vote swings.

In the Election Model, five projection scenarios over a range of undecided voter allocation assumptions display the effects on aggregate vote share, electoral vote and win probability.

Snapshot projections do not provide a robust expected electoral vote split and win probability. That’s because unlike the Monte Carlo method, they fail to consider the two bedrocks of statistical analysis: The Law of Large Numbers and the Central Limit Theorem.

For example, assume that Florida’s polls shift 1% from 46-45 to 45-46. This would have a major impact in the electoral vote split. On the other hand, in a Monte Carlo simulation, the change would have just a minimal effect on the expected (average) electoral vote and win probability. The 46-45 poll split means that the race is too close to clearly project a winner; both candidates have a nearly equal win probability.

NORMAL DISTRIBUTION

The Excel function has a very wide range of applications in statistics, including hypothesis testing.

NORMDIST (x, mean, stdev, cumulative)
X is the value for which you want the distribution.
Mean is the arithmetic mean of the distribution.
Stdev is the standard deviation of the distribution.

EXAMPLE: Calculate the probability Kerry would win Ohio based on the exit poll assuming a 95% level of confidence.
Sample Size = 1963; MoE =2.21%; Cluster effect = 20%; Adj. MoE = 2.65%
Std Dev = 1.35% = 2.65% / 1.96

Kerry win probability:
Kerry = 54.0%; Bush = 45.5%; StdDev = 1.35%
Kerry 2-party share: 54.25%
Probability = NORMDIST(.5425, .5, .0135, TRUE)= 99.92%

BINOMIAL DISTRIBUTION

BINOMDIST is used in problems with a fixed number of tests or trials, where the outcome of any trial is success or failure. The trials are independent. The probability of success is constant in each trial (heads or tails, win or lose).

EXAMPLE: Determine the probability that the state exit poll MoE is exceeded in at least n states assuming a 95% level of confidence. The one-tail probability of Bush exceeding his exit poll share by the MoE is 2.5%. Therefore the probability that at most N-1 states fall within the MoE is:
Prob = BINOMDIST (N-1, 50, P, TRUE)
N = 16 states exceeded the MoE at 12:22am in favor of Bush.
The probability that the MoE is exceeded in at least 16 exit polls for Bush:
= 1- BINOMDIST (15, 50, 0.025, TRUE)
= 5.24E-14 or 1 in 19,083,049,268,519

A SAMPLING PRIMER

The following is an edited summary from http://www.csupomona.edu/~jlkorey/POWERMUTT/Topics/data_collection.html

A random sample is one which each outcome has an equal probability of being included. It is an unbiased estimate of the characteristics of the population in which the respondents are representative of the population as a whole.

The reliability of the sample increases with the size of the sample. Ninety-five times out of a hundred, a random sample of 1,000 will be accurate to within about 3 percentage points. The sample has a margin of error of approximately plus or minus (±) 3 percent at a 95 percent confidence level. If a random sample of 1,000 voters shows that 60 percent favor candidate X, there is a 95 percent chance that the real figure in the population is in the 57 to 63 percent range.

Beyond a certain point, the size of the population makes little difference. The confidence interval is not reduced dramatically. Therefore pre-election national polls usually don’t survey more than about 1,500 respondents. Increasing this number increases the cost proportionately, but the margin of error will be reduced only a little.

Often it is not practical to carry out a pure random sample. One common shortcut is the area cluster sample. In this approach, a number of Primary Sampling Units (PSUs) are selected at random within a larger geographic area. For example, a study of the United States might begin by choosing a subset of congressional districts. Within each PSU, smaller areas may be selected in several stages down to the individual household. Within each household, an individual respondent is then chosen. Ideally, each stage of the process is carried out at random. Even when this is done, the resulting sampling error will tend to be a little higher than in a pure random sample,[7] but the cost savings may make the trade-off well worthwhile.

Somewhat similar to a cluster sample is a stratified sample. An area cluster sample is appropriate when it would be impractical to conduct a random sample over the entire population being studied. A stratified sample is appropriate when it is important to ensure inclusion in the sample of sufficient numbers of respondents within subcategories of the population.

Even in the best designed surveys, strict random sampling is a goal that can almost never be fully achieved under real world conditions, resulting in non-random (or “systematic”) error. For example, assume a survey is being conducted by phone. Not everyone has one. Not all are home when called. People may refuse to participate. The resulting sample of people who are willing and able to participate may differ in systematic ways from other potential respondents.

Apart from non-randomness of samples, there are other sources of systematic error in surveys. Slight differences in question wording may produce large differences in how questions are answered. The order in which questions are asked may influence responses. Respondents may lie.

Journalists who use polls to measure the “horse race” aspect of a political campaign face additional problems. One is trying to guess which respondents will actually turn out to vote. Pollsters have devised various methods for isolating the responses of “likely voters,” but these are basically educated guesses. Exit polls, in which voters are questioned as they leave the voting area, avoid this problem, but the widespread use of absentee voting in many states creates new problems. These issues are usually not a problem for academic survey research. Such surveys are not designed to predict future events, but to analyze existing patterns. Some are conducted after the election. The American National Election Study, for example, includes both pre and post election interviews. Post election surveys are not without their own pitfalls, however. Respondents will sometimes have a tendency to report voting for the winner, even when they did not.

The American National Election Study split its sample between face to face and telephone interviews for its 2000 pre-election survey. The response rate was 64.8 percent for the former, compared to 57.2 percent for the latter. An analysis of a number of telephone and face-to-face surveys showed that face-to-face surveys were generally more representative of the demographic characteristics of the general population. Note that many telephone surveys produce response rates far lower than that obtained by the ANES.

Another approach is the online poll, in which the “interaction” is conducted over the Internet. Like robo polls, online polls are also less expensive than traditional telephone surveys, and so larger samples are feasible. Because they require respondents to “opt in,” however, the results are not really random samples.

When samples, however obtained, differ from known characteristics of the population (for example, by comparing the sample to recent census figures), samples can be weighted to compensate for under or over representation of certain groups. There is still no way of knowing, however, whether respondents and non-respondents within these groups differ in their political attitudes and behavior.

 
1 Comment

Posted by on September 1, 2011 in 2004 Election

 

Tags: , , , ,

 
Follow

Get every new post delivered to your Inbox.

Join 768 other followers