I carefully set up these exit polls to compare the official vote count by machine type. The only legitimate concern regarding the meaning of these results is a biased sample. Not everybody tells the truth. Some people delight in giving false answers to surveys. How are you going to account for that? It’s a fair concern.
While I cannot prove that didn’t happen (at least, not without access to the ballots, which isn’t permitted), this is part of the normal error I expect. It always helps to state assumptions explicitly.
INTROVERTS, LIARS, AND IDIOTS ASSUMPTION : THESE TRAITS ARE RANDOMLY DISTRIBUTED AMONG ALL CANDIDATES AND POLITICAL PARTIES. I am assuming that that people who were less likely to participate (introverts) or more likely to fudge their answer (liars) or make mistakes (idiots) in filling out the survey did not differ in their response to our exit poll.
I received the following email that sums up this concern nicely and also suggests a couple of ways to check that hypothesis.
Hi Beth, The observed discrepancies between official results and your poll results very clearly show that Clinton (D) voters were more strongly represented in those polled than in the official vote count; Trump (R) voters were less well represented. There are many possible explanations for this discrepancy. One hypothesis is that a certain percentage of voters “held their nose and voted for X” and would never have participated in the poll. If these voters tended to be more of one party than the other, than that party would be less represented in the polls. Fortunately, your data provide a means to test this hypothesis about the “missing minority”, for it leads to this prediction: If a “missing minority” was biased towards X, then sites at which X had a greater percentage of the votes would be least affected by vote disparities. A corollary prediction: sites having the highest response rate would be least affected by vote disparities. Have at it! Annie
The main reason I find this hypothesis implausible is that the discrepancies for the Supreme Court judges were twice as large and followed the same pattern as the Pres. race discrepancies. There’s no reason to think more people ‘held their nose’ for judges than president!
Regarding those two predictions:
In short, we do not see the other data relationships we would expect if the introverts liars and idiots assumption were false. There is no reason to assume these individuals were more likely to vote for one candidate than another resulting in the bias in our data.
]]>The exit poll results from all five polling locations in Southeast Kansas show strong evidence of election fraud in both the patterns and size of the errors.
I had major concerns with the accuracy of our voting machines based on my previous analyses, which is why these exit polls were run. The results confirm those suspicions.
I designed this exit poll to check whether or not our voting machines are giving us accurate counts. I looked into our local election statistics in the past and found concerning indications of fraud in the data. There is no public official reconciliation of the paper records with the official vote counts provided by machine nor are citizens allowed access to do it. I have the credentials to do this; I have a Ph.D. in statistics and have been certified by the ASQ as a quality engineer since 1987. I was able to recruit enough concerned voters to man the exit polls from open to close on election day.
Voters were asked how they voted – by machine, a scanned paper ballot, or an uncounted provisional ballot. Results from the polling location give us the breakdown by machine votes and scanned ballots, which can be directly compared. The electronic voting machines used in all three Kansas counties were ES&S Ivotronic. The paper ballot scanning equipment varied, but was all from the same manufacturer: ES&S.
The results from these exit polls tell a consistent, albeit unpleasant, story: Our electronic voting machines should not be trusted. Scanned paper ballots have been impacted as well, but due to some technical issues regarding the data, results on that type of counting machinery are less compelling. Scanned paper ballot results often continued the pattern of the voting machine results, which does add to the weight of evidence against the accuracy of the official results.
I have posted the data from our exit poll and the corresponding official vote counts at Exit Poll Data
These exit poll results clearly point to manipulation of the machine counts of our votes. These are not random errors. There is no other reasonable explanation for large and consistent errors in favor (or against) particular candidates in this situation.
Presidential race results show votes shifted from Clinton to Trump in four of the five locations – all except Sumner County.
The analysis details are posted at Analysis of 2016 Citizens Exit Poll in Southeast Kansas
There is one ray of sunshine in these results – while the size of the shifts are cause for grave concern about the accuracy of the vote count, they are not sufficient to have altered the outcome in any of the races mentioned above. Kansas was Trump territory. The Judges all retained their positions. No Libertarians won.
This ‘ray of sunshine’ is limited to these results. Races polled at only one or two polling locations look even worse. There was a more than 10% shift in votes from Norton to O’Donnell in the Sedgwick County Commissioner third district race, easily sufficient to alter the winner*. The data from these local races may only affect a portion of the voters at the polling site. For that reason, the data from those races is not as solid. The lower quantity and quality of data in those races reduces confidence in any conclusions regarding the results.
Who’s doing this and How? I don’t know. My analyses shows which candidates lost votes or benefited, but that’s not justification for assuming they are knowledgeable regarding the vote theft. There’s only one conclusion about the perpetrators I can come to.
Multiple Agents – The profile of errors from Sumner County is so different from the other sites, I can conclude that more than one agent successfully altered voting machine counts in S.E. Kansas polling stations.
]]>
In the absence of election fraud, the difference in vote share between the official count and an exit poll (called the error) will be randomly distributed (both positive and negative) and relatively small. If voting machine counts have been altered, we will see telltale patterns in these error measurements. We can determine if our machine votes are being counted honestly or if some candidates benefit and others are victimized by election fraud. The exit poll results from all five polling locations show strong evidence of election fraud in both the patterns and size of the errors.
EXAMPLE: The graph above shows the results for the presidential race from SE Wichita. According to the machine totals, Hillary Clinton received 435 votes out of 983 cast on the voting machines there. That’s a 44.25% vote share. Our exit poll data showed Hillary Clinton received 306 votes out of 645 survey responses to this question from voters who cast their votes on those same machines at that polling location. That’s a 47.44% vote share. The difference between those two values, -3.19%, is the error, illustrated in the graph below. This error measurement is computed for each candidate, race, type of voting equipment and polling location.
There were some problems with some of the data. I have included data from all five sites for their electronic voting machine counts. The link above gives the raw data for both voting machines and scanned paper ballots for all five sites, but only three of the five sites had sufficiently high quality data to be included in this analysis. This post discusses what data was left out and why.
Presidential race results show votes shifted from Clinton to Trump in four of the five locations. The errors for the presidential candidates by site and voting equipment are shown in the table below.
These values are also shown in the chart below. Johnson and Stein errors look random and reasonable. Clinton and Trump errors are much larger and roughly match on the DRE machines with votes shifting from Clinton to Trump in four of the five polling locations.
To statistically analyze the size of errors, use the hypergeometric distribution. This computation is available in EXCEL as HYPGEOM.DIST. It takes into account both the size of the population (total voters in the official count) and the sample size (total exit poll responses) in computing the probability of getting an error as large or larger than our exit poll had. See this post for the technical details about how this computation is done.
The p-values for two-sided tests are given in tables below. Yellow indicates a statistical flag, a probability of less than 5% occurring if there was no election fraud. Bold red numbers indicate probabilities of less than 1 in 1,000.
The p-values clearly confirm the initial impression generated by the graph above: voting machine election fraud occurred in four of the five polling locations shifting votes from Clinton to Trump.
One interesting detail – Jill Stein actually received more scanned paper ballot votes in our exit poll in SW Wichita that they recorded at that site. Since that can’t actually happen without errors or dishonesty, that probability is an absolute zero. I wrote out ‘Zero’ to distinguish this situation from 0.0000 which indicates a probability that is below 0.00005 but still above zero.
The Senate and 4^{th} district Rep races were skewed toward the Libertarians.
The only pattern in these two races was that the Libertarians ALWAYS benefitted from the errors, with higher machine counts than exit poll percentages. Both Democrat and Republican candidates lost votes, in some cases by suspiciously large amounts approximating the size of the error of another candidate.
Polling locations differed considerably. Sumner county looks as if votes were taken from Moran (R ) in the Senate, but even more from Giroux (D) in the 4^{th} Cong. Dist and undervotes for both races were increased. Independent candidate Miranda Allen for the 4^{th} district benefited by an unusual amount in the machine vote counts in all three Sedgwick County polling locations. These errors look like fraud.
Below are tables and graphs of the errors between the official results and the exit poll results for the Kansas Senate and 4th Congressional Districts and tables of the p-values for those errors.
The data from the Supreme Court Judges show the most clarity. The pattern that fits across all five judges cannot be denied. In addition, the magnitude of the errors also exceeds that found in the other three races.
The four Supreme Court judges actively opposed by Gov. Brownback had Yes votes stolen in same four locations that favored Trump. The only positive error is a tiny one for Nuss in the SE Wichita location, with the remaining errors for those four sites all showing negative for all five judges. Sumner, different once again, showed only positive errors (More Yes votes) for all five judges
Stegall, Brownback’s only appointee up for retention, has results identical in direction to the other four, but smaller in magnitude. He has only one slightly improbable dearth of yes votes in the scanned paper ballots in the SW Wichita location. For Stegall, only the fact that his pattern matches the others is a sign of fraud against him. For the other judges, both the size and pattern of the errors testify to the rigging of the official counts by the machines.
Below are tables and graphs of the errors between the official results and the exit poll results for the Kansas Supreme Court Judges Retention Votes and tables of the p-values for those errors. Multiple graphs of the judges are shown, grouping by judge (as previous graphs) and grouping by location. That latter makes it undeniable that all sites show signs of corruption, although not in agreement on the preferred direction. Finally, a graph showing the judges next to a graph of the presidential candidates on the same scale.
This last comparison, putting the errors for the presidential race on the same scale as the judges, actually startled me when I first graphed it and I was expecting it. The average size of the errors should be approximately the same for all the races since they are all drawing from a near identical sample of voters. To a statistician, this increase in the magnitude of the error for the judges is another flashing red light saying that these machine results have been rigged. Rigged in different ways in different places, but all of the sites with exit polls show the telltale signs of the corruption.
]]>
I am not including the scanned paper ballot survey data from the Urban Wichita and Sumner County sites. The reason is that when I look at the response rate for the different methods of voting, I see signs of potential problems at those two locations.
I am concerned with the relatively low response rate for provisional ballots and relatively high rate for the scanned paper ballots. I suspect that in those two sites, some provisional voters mistakenly indicated the scanned paper ballot. This was a relatively easy mistake for voters to make; they might not be tuned into the difference between them.
Since one hypothesis I’ll be testing later is whether or not the provisional voters’ choices were different from the counted ballots, this concern renders the data from those sites as inconclusive. Under this circumstance, voter error is a reasonable alternative explanation to election fraud for any statistically significant differences. Because my main concern is with the voting machine results, I decided not to include the results for the scanned paper ballots from those two sites in my analysis. I may revisit this assumption later if the provisional votes are not found to be significantly different from the votes that were counted.
On the other hand, since the leakage appears to shift voters from provisional to scanned paper ballots, the data from provisional voters can be considered representative. However, only the Urban Wichita provisional data will be analyzed because the Sumner County provisional sample had only 13 surveys; not large enough to draw any conclusions from.
I am keeping the provisional surveys from the SW and SE locations even though had much higher rates of provisional voters, with the SW location claiming 101% response rate, with one more provisional survey than the official count. Clearly we had at least one confused survey taker. But since the response rates for the machine and scanned paper ballots are similar in those cases, my assessment is that people who voted provisionally were simply more likely to complete one of surveys. They were worried their official vote wouldn’t be counted. One such voters complained bitterly to us about it while filling out our survey. Suddenly, their name had simply vanished from the registration books despite having voted there regularly in the past! One of our volunteers had the same experience. This excess of provisional voters does not seem likely due to contamination from the scanned paper ballot voters, so the data from provisional voters can be considered representative.
Blanks – They’re a technical issue for surveys. There are two sorts of blanks with respect to survey responses. A survey taker might not indicate an answer on one or more questions. These were coded NR (No response) and were not included in any further analysis. Valid responses to other questions were retained. Overvotes on any question were treated the same.
We included ‘Write in or left blank’ as an option to the candidates as much as possible. Space on the survey form was at a premium, and some site managers deciding against including it on all questions. I insisted on it for the questions asked on all surveys. I felt it of particular importance for the presidential election given the candidate selection. For those three questions, people who didn’t answer that question are removed from the sample, but we can compare the rates for write-in or left blank with the number of write-ins and undervotes on the official results. For other questions, including the judges, undervotes and ‘write-in or left blank’ are taken out of the sample and all subsequent computations unless specifically stated otherwise.
These choices do affect the p-value computations of the hypergeometric distribution given in my tables.
]]>
The Hypergeometric distribution is used to determine the probability (p-value) of getting a random sample, drawn without replacement, as extreme or more so given the population that the sample is drawn from. If that wording sounds unnecessarily complex, I sympathize. Unfortunately, precision is often complicated to articulate. This definition is hard to parse and you need a working knowledge of what the statistical terms and phrases mean.
“With” versus “without” replacement is an important descriptor of a random sample. A situation with replacement is akin to selecting a card from a deck, then returning it back to the deck before drawing another card. Without replacement is selecting a second card from the 51 remaining in the deck.
This nuance in the drawing of the sample affects the basic assumptions statisticians build equations from. Different statistical distributions have been developed to handle the two situations. Because voters were only asked once to fill out our survey, the exit poll sample is ‘without replacement’ and the Hypergeometric distribution is the most appropriate choice for testing the size of the errors.
Another important and relevant statistical concept: One-sided and Two-sided tests.
Most distributions, including the Hypergeometric distribution, have the majority of data crowded around the average and the data gets sparser the farther away from the average. This type of distribution has ‘tails’. There are the two directions of tails relative to the average value: upper and lower.
When we perform a statistical test, we are looking at the deviations from what is expected given the underlying distribution of the data. In some cases, we may only be interested in deviations in a particular direction – high or low. In those cases, we can increase the precision of our test by only looking at one end of the distribution. This is called a one-tailed test.
In other situations, we are interested in differences in either direction, so we are examining both the upper and lower tails of the distribution. In our exit polling data, we are looking at the deviations both positive and negative, to determine if either are unusually large. Therefore, it is a two-sided test.
The EXCEL HYPERGEOM.DIST function computes only the lower tail p-value. This result can be manipulated to find the upper tail probability. Both need to be computed. This EXCEL function requires five inputs:
EXAMPLE: Hillary Clinton received 435 votes out of 983 cast on the voting machines at the SE site. That’s a 44.25% vote share. Our exit poll data showed Hillary Clinton received 306 votes out of 645 survey responses to this question from voters who cast their votes on those same machines at that polling location. That’s a 47.44% vote share. The difference between those two values, -3.19%, is the error. This error measurement is computed for each candidate, race, type of voting equipment and polling location.
Use EXCEL function HYPERGEOM.DIST with the following inputs:
Lower Tail P-value for Clinton, Machine Votes, SE Wichita
= HYPERGEOM.DIST(306,645,435,983,1) = 0.9979
Whoa!! I thought you said Hillary got cheated? This result is a near certainty.
That’s because our exit poll sample had a larger percentage of Hillary voters than the official results did. Her exit poll results lie in the upper tail of the distribution, well above the official average. We just computed the p-value for the lower tail i.e. the probability of randomly getting as many Hillary votes as we did (306) or LESS.
Next we need to compute the probability of randomly getting as many exit poll votes as she did or MORE. The upper tail of the distribution. Through the magic of math, we can find the upper tail probability with a modification to this function.
Subtract 1 from our sample size and compute the lower tail probability for that sample size. Then subtract that lower tail probability from 1 to get the correct upper tail p-value.
Upper Tail p-value for Clinton, Machine Votes, SE Wichita
= 1.0 – HYPERGEOM.DIST(305,645,435,983,1) = .00325
Finally, because we did not specify in advance what direction we expected to see, this is a two-tailed test. For two-tailed tests of this nature, the p-value is computed as double the minimum one-tailed p-value, capped at 1.0.
Two-tailed p-value for Clinton, Machine Votes, SE Wichita = 2*0.00325 = 0.0065.
Finally, putting it all together in one cell, nesting the needed functions:
Two-tailed p-value for Clinton, Machine Votes, SE Wichita
=2*MIN(+HYPGEOM.DIST(306,645,435,983,1), 1-HYPGEOM.DIST(305,645,435,983,1), 0.5)
Picture from http://www.ats.ucla.edu/stat/mult_pkg/faq/pvalue1.gif
]]>There were five sites sponsored by the Show Me The Votes Foundation. Each site was independently run by voters from that area. I provided guidance and instructions as well as training for volunteers. You can learn more about how these exit polls were set up and run here How to run an exit poll. Polls were manned the entire time the polling station was open. Voters were asked personally to participate; responses were kept anonymous.
The site managers were all amazing. We got excellent response rates. This study and these results would not have been possible without their help and the help of the dozens of other volunteers who were willing to take a few hours out of the day on Nov 8th to man my exit polls. Thank you all.
I ran the Southeast Wichita site. The questionnaire I used is the basic form that the others were built off of. All questions on that form were common to all five sites. There were no questions on races specific to my polling location.
Southeast Wichita Exit Poll Results
The Sumner County site was run by Glen Burdue in Wellington KS. He modified his survey to add some additional questions specific to his polling location. We had relatively few volunteers in the Wellington location, but he made up for it with dedication to the project. Glen spent considerable time notifying and clarifying his exit poll with the county officials not to mention collecting surveys all day long on Nov 8th. Then he counted all the surveys himself, no small task by itself.
Sumner County Exit Poll Survey Form
Sumner County Exit Poll Results
The Cowley County site was run by Pam Moreno of Winfield KS. She is an amazing organizer and and is a leader in the Women for Kansas – Cowley County chapter. We ran a volunteer training session one October evening. Her group also added a few questions to our basic survey that were specific to Winfield. She also got volunteers to help do the counting the next day and professional help with the data input into my spreadsheet for analysis.
Cowley County Exit Poll Survey Form
Cowley County Exit Poll Results
The Urban Wichita site was run by Lori Lawrence at the Health Department on 9th Street. She did a phenomenal job, doing everything from baking cookies to attract voters to respond to getting buttons printed up for our volunteers to wear. In our planning meetings, she made many excellent suggestions for ways to improve our exits polls. She ran the same basic questionnaire that I did at the Southeast Wichita site. She also did the complete first count on all her surveys. With 883 useable surveys, that was a huge chore!
Urban Wichita Exit Poll Results
The Southwest Wichita site had two co-managers, Lisa Stoller and Leah Dannah-Garcia, two excellent ladies who were devoted to accomplishing this. They collected 1,435 usable surveys, a response rate to eligible voters of 80%. Fantastic. Counting all those surveys was a daunting task. I dare say that aspect of it was as difficult for those extroverts, who were so excellent at running the exit poll, as the task of walking up to strangers and asking them to fill out a survey form was for introverts like me. They needed help. I ended up parceling them out to other volunteers as it was an overwhelming task, even for a veteran survey counter like myself.
]]>Exit polls – taken as people are leaving the polling place – are extremely accurate at capturing the vote share of candidates. Staffed by volunteers from the polls open to their close, we achieved outstanding response rates. These are shown in the table below.
eta: after final counting, the SW Wichita votes for Trump changed from 609 to 611.
Polling Station and Exit Poll Results
A rule of thumb is that 2% or larger difference in vote share between the official results and the exit poll is evidence of election fraud worth investigating. We had such excellent response rates at some of our sites that differences significantly smaller than 2% are considered suspect in some races.
In addition, we can take into account the overall composition of differences between the official results and the exit poll statistics for each site. For example, at the Wellington polling site in Sumner County, all Republican candidates had lower vote share in the machine counts than in the exit polling results. This consistency is suspicious even though the differences may be small.
The exit poll results indicate that our machine generated counts are being manipulated. Polling sites in Sedgwick and Cowley counties were manipulated for the benefit of some candidates, most notably Trump at all four of those sites. Results in Sumner County appear to be manipulated to the detriment of Republican candidates – but not necessarily to the benefit of Democrat Candidates. Libertarians performed better in the machine counts for both the Senate and the 4^{th} District races than exit polls indicated for all five sites. These differences are not sizable enough to alter the outcome of most races, but they are consistent and larger than expected by chance alone. I’ll post more about those results as I do a more detailed analysis for each polling location.
Presidential Race Analysis:
Votes for Hillary Clinton were shifted to Donald Trump in four of the five polling locations we surveyed, Sumner being the exception. This chart shows approximately 2% to 3% of the machine votes were shifted from Clinton to Trump at those sites, adding 4% to 6% of the vote share to the difference between them to benefit Trump. The other candidates show only normal error rates.
Figure 1 – This graph shows the difference between the machine vote share and the exit poll vote share for each candidate at each site. Positive values show that the machine count benefited that candidate. Negative numbers indicate a loss compared with the Exit Poll results.
Sites in Sedgwick and Cowley Counties show a distinct bias with the machine counts siphoning votes from Clinton and benefiting Trump. Sumner County exit poll results for the Clinton and Trump were not statistically significantly different from the machine counts for Sumner County.
Since Trump won Kansas with 54% of the vote to Hillary’s 36%, even assuming this shift held across Kansas (it didn’t), it was well below Trump’s margin of victory, so this manipulation of votes did not alter the outcome. Still, it is disturbing evidence that the machine vote counts are being altered. In other states, which use similar equipment, manipulation at this level could have changed who won the Presidency.
Statistical Details
I computed the exact probability of each candidate getting the vote share they received in our exit poll given the official counts for that polling location. This was computed using the Hypergeometric probability distribution, which takes into account both the size of our exit poll sample and the number of people who cast votes at that polling location on Election Day.
This probability – or p-value – is the exact computation of the probability of getting our exit poll results assuming no election fraud occurred. The p-values for the different presidential candidates at each of the five exit poll sites are given in the table below.
The p-value represents the level of concern about the official results given our exit poll results with 1.0 indicating everything’s normal, nothing of interest here and zero indicating Red Alert Danger Will Robinson! Danger! The computed p-values always fall somewhere in-between.
The probabilities for Johnson and Stein are all quite reasonable and raise no serious alarms regarding the accuracy of their vote counts. The probabilities for Clinton and Trump, on the other hand, are low enough to sound alarms for four of the five polling locations.
These exit poll results more than justify a call to audit the voting records and a profound skepticism in the results of machine counted votes.
The Cumulative Vote Share (CVS) Model is Validated as a Sign of Election Fraud
The math underlying this model dictates that this trend should level off horizontally, not start moving in the opposite direction. It means the trend is not random chance, but due to a specific cause correlated with precinct size. It is such unanticipated trends as revealed by this type of graph that motivated me to look more closely into our vote-counting process, eventually leading to conducting the exit polls in this past election.
This is CVS graph for Sedgwick County. It shows Trump getting an increase of ~2% of the total vote share and Clinton losing that same amount from their respective inflection points at around 93,000 cumulative votes. These exit poll results vindicate the use of the cumulative vote share model in assessing probability of election fraud.
]]>
The most interesting thing that happened was regarding a young black woman who had declined to participate in our exit poll. Later that evening, near to the polls closing, she was back. Apparently her mama had voted there early that day and filled out our survey. She had insisted her daughter return to do so. I was delighted! That young woman has a mama who cares about her and about making her voice heard by voting. I wanted to thank her and tell her she’s an awesome mom! But I was too busy handing out survey forms to other voters. We had excellent participation rates!
I took both Tuesday and Wednesday off work, and was able to spend all day counting ballots Wednesday. I’m also an experienced survey counter, so I managed to get the 925 surveys organized and counted by the end of the day.
My volunteers are working together to get the other sites counted, but they have other things to do too. Altogether, we have collected thousands of exit poll surveys. Counts are continuing even as I write this. That I want the surveys from each site to have two independent counts doesn’t make the task any easier.
I also need to verify the results we wrote down at the polling stations, but was informed today that they won’t be available until after the canvassing is complete, approximately a week to ten days from now. I’m glad I asked my site managers to get the totals that night from their polling stations. I can go ahead and work on my analysis, updating it with any corrections needed when the data is available. But I’ll hold off publishing the results until I can verify the numbers.
I will share some general stuff from my initial numbers for my site with the caveat that these results are considered preliminary until the data has been verified.
The scanned paper ballot official counts and our exit poll results are close, with nothing falling outside reasonable statistical bounds on any of the races.
The voting machine counts and our exit poll results are not as close, with a couple of results that bear looking into, but I had three results flagged earlier. One has already turned out to be a data input error on my part. (That’s why I want to verify the numbers before publishing).
I will present the exit poll provisional ballot results though. I’m reasonably confident there are no large errors in my counts and small changes won’t alter these results.
In general, people who had voted provisionally were more likely to have time to take our survey. We had 79 out of 92 provisional voters fill out our survey for a response rate of over 85%, nearly 20% higher than the 66% response rate of voters whose ballots were counted that day.
In our exit poll survey, I found that provisional voters were approximately 10% less likely to vote for republican candidates. I think it is reasonable to take this along with the additional data from the other exit poll locations, as a measure of the effectiveness of Kris Kobach’s efforts to disenfranchise non-republican voters. I have already contacted the league of women voters to see if this data will be of help with their lawsuit regarding his practices.
]]>
Polling stations allow voters access to the results after the polls are closed when the votes have been tallied. In Sedgwick County, they will have separate reports printed for the electronic voting machines and the scanned paper ballots.
We will also get a count of the number of provisional ballots collected from the polling location. Theseballots will not be opened until the voter’s registration is verified and there will never be an official tally of the provision votes for polling location. But we can look at the results we have for voters who submitted provisional ballots and compare them with the votes that were counted at the polling location. If there are significant differences, this is evidence of the voter suppression effect of Kris Kobachs voter registration rules.
I have created a general data collection and analysis EXCEL spreadsheet. Multiple precincts vote at each polling location and the results are reported for each precinct, not the polling location, so I’ve set up a spreadsheet to sum the numbers up and compute the appropriate probabilities.
I will be customizing this EXCEL file for each exit poll location in Kansas, but I am happy to share a general version of this worksheet with anyone who is interested in running an exit poll for their own area. All you will have to do is input the official results and your exit poll results. This is an example of of the output.
Example Data Analysis | |||
Presidential race | Chi-Squared Result: NA | ||
Candidates | Exit Poll | Official Results | Binomial Probability |
Clinton (D) | 52 | 60 | 0.0638 |
Trump (R) | 38 | 30 | 0.0530 |
Johnson (L) | 6 | 6 | 0.5593 |
Stein (G) | 2 | 2 | 0.5967 |
Other | 2 | 2 | 0.5967 |
Total | 100 | 100 |
There are two different analyses than can be used in this situation. The chi-square test will give an exact probability that the actual results differed from what would be expected under the assumption of random chance. EXCEL has this test as a built in function: CHISQ.TEST. But the chi-squared test has minimum data requirements which were not met in this example, hence “NA” or Not Applicable as the result of this test.
Since the chi-squared test will not work for every set of possible data, I also show the individual binomial probabilities for each of the candidate. The minimum probability from this set of five computations is a reasonable approximation to the exact computation using the extension of the binomial distribution and can be easily computed using built-in Excel formula BINOM.DIST.
How to interpret this:
We judge the probability of machine manipulation of the vote by evaluating the probability of our results assuming no manipulation of votes is occurring. This is referred to as the “null hypothesis”. All probabilities shown are made under this assumption. If this probability is above 0.05 (5%), we can reasonably conclude that the differences between the machine vote share and the exit poll vote share are typical of random variation due to the normal errors in the process.
If this value lies between 0.05 and 0.001, raise an eyebrow and give the numbers for that race a little extra scrutiny and consider it in concert with the other exit polling results.
If this values lies below 0.001, that is evidence of fraud. Personally, I would like to see a recount of any race with results that fall this far from normal. But only a candidate can request a recount in Kansas.
In this example, I have contrived to show Trump with a questionably low # of votes in the official count compared to the exit poll results. Hillary has a slightly elevated value. But these results are not unexpected as the minimum probability of results this far off is above 0.05.
But if the other sites have similar values and they are all benefitting the same candidate, it would be concerning. If 2 or 3 sites out of 5 show the same beneficiary of the differences, that’s reasonable. But if 5 out of 5 sites show the same beneficiary, it’s evidence of rigging.
If we see multiple races with low odds and the same slate of candidates are benefiting, we have solid evidence of machine manipulation of our official votes. If we see only the normal expected errors, then we have solid evidence it is NOT being manipulated.
While a single location and a single race might show evidence of manipulation, savvy cheaters will try to avoid this method of detection by establishing a maximum shift that falls beneath the 0.05 probability results. But looking at multiple races and sites, we can establish whether even small shifts show evidence of cheating.
We can define a slate of candidates by party and check the probability of getting the results we got using a similar binomial analysis. Under the null hypothesis of no manipulation, the probability of an error that benefits a candidate is 50%. There are three races with candidates and five judges we are asking about, for a total of 8 results for each polling location. Governor Brownback would like to see 4 of the 5 judges lose their jobs and replace them. We can also presume he supports the Republican Party candidates for President, Senate and Representative.
We will have data from 5 different locations for a total of 40 random samples with approximately 50% probability. (For example, let X be the number of errors that were the opposite of the Brownback administration preferred candidates. If we have 40 random samples as defined above, the probability of getting errors in the opposite direction of his preferred result is computed with the following excel formula: BINOMDIST(X, 40, .5, 1)
If this value is extremely low (less than .001), we conclude that the Republican Party has unduly benefited and further investigation would be appropriate.
How to interpret the Provisional Ballot Data:
We cannot know the final count of the provisional ballots collected at a polling location. They are polled at the county level and only those that are shown to be registered voters are opened and counted. What we can do is compare the results of the provisional ballots with the other responses to our exit poll. If there is a major difference between those asked to fill out provisional ballots with the automatically counted votes, we have a measure of the effect of the voter ID laws and if it made a difference to the outcome.
For each race, we can use the chi-square test if we have sufficient data. Otherwise, we can use the binomial approximation similar to the one used to compare the official count to the exit poll survey results.
electioneering and instructs them regarding what they can and cannot do. While permission is not required to run an exit poll, we do need permission from the property owner to set up a booth to collect our ballots and provide chairs and shade for our volunteers. Mainly, we want everyone to know what we are doing to avoid any issues arising on on election day.
How to Run an Exit Poll Part 1
]]>The exit poll survey ballot is important, but not complicated. The only question of interest, other than their ballot choices, is the method of voting. Data will be available at the end of the day with separate totals for the machine cast votes and the scanned paper votes. There will be no official count of provisional votes at this station, so we can only compare those votes to overall total for the polling station. But that comparison allows us to evaluate whether the giving people provisional votes amounts to a voter suppression tactic.
Since space on our survey form is at a premium, and because including that information makes their response less anonymous, I do not recommend including questions about age, race or gender. Generally speaking, you want to keep the words to minimum. (Not an easy task for me.)
Here is an example survey I have developed for exit polls in Sedgwick Co. I included a short paragraph at the top because I feel it’s important to let people know why you want this information and reassure them that results are anonymous, just like their vote.
The first question is really too long, but I wanted to be as clear as I can about this question. In Sedgwick County Kansas there are three possible options: A vote cast via electronic voting machine, a paper ballot that the voter feeds into a scanner for on-site electronic counting or a provisional ballot – a paper ballot that is sealed into an envelop to be counted later (maybe).
Asking about the specific races is straightforward. State the office and then list the candidates. Circling answers reduces the need for a blank or box to check. It saves space on the page.
Staggering the answers for questions with more than one line of answers (ex: Pres) makes it easier to discern the voters intent. When they are stacked one above another, the answer may easily become ambiguous.
Since a single polling location will have multiple precincts voting there, it’s a problem asking about races where different precincts will be voting for different candidates. Generally, I want to confine the questions to races that will appear on every ballot at the polling location. On the other hand, my site managers for the SW Wichita location are very interested in the county commissioner races. We arrived at the following:
Who did you vote for your County Commissioner Race? (Select one for District 2 OR District 3) – sw-wichita-nov-8-exit-poll-ballot
I have hopes that we won’t get too many voters identifying their choices for both district 2 and district 3, but I expect we will get some. OTOH, it’s the only question that would be spoiled and I’m reasonably comfortable in assuming that such mishaps are equally likely to occur regardless of which candidate they support. I think we will get good data from this exit poll.
How to Run an Exit Poll Part 1
How to Run an Exit Poll Part 2
]]>