More questions continue to bubble up, adding to the feeling that all those pre-election polls in Kansas might not have been wrong after all. Consider first that Beth Clarkson, the Wichita State University mathematician, has not abandoned her pursuit of “suspicious patterns,” as she calls them, in election data; nor has she given up her court battle to obtain certain records of the 2014 statewide elections in Kansas.
And recently, poll workers in Shawnee County have come forward to recount strange electronic happenings on election night – an odd surge here, a blip there, frustrating the count in two narrow races – one for governor, the other for the U.S. Senate. More doubt is piled on mounting skepticism that electronic voting is hardly technology’s new wave for democracy, but a sinister undercurrent threatening to drown what’s left of the people’s voice.
The developing scenario is that something quite peculiar happened late on election night. What, exactly, isn’t clear. What is clear, is that Secretary of State Kris Kobach, the state’s chief election officer, is working feverishly to keep researchers from finding out. Brief background:
By the eve of the November 4 elections, prominent state and national polls were predicting, at worst, a dead heat in the U.S. Senate race, Greg Orman’s strong challenge to incumbent Pat Roberts. In the contest for governor, the polls were predicting a narrow victory for Democrat Paul Davis over incumbent Gov. Sam Brownback. No regional or national survey predicted outright victory for either Roberts or Brownback. The narrowest margins in polling showed the odds to be slim, if any, that the incumbents would keep their seats. The consensus was that the incumbents might lose. Early in the voting, the each of the challengers built narrow leads that held well into election night. Confidence began to build among election workers at both the Davis and Orman campaigns.
Then came the “surge.” Suddenly in select urban precincts, a blizzard of votes drifted in, overcoming the margins held by both challengers. They lost at the finish. The initial shock was deep and widespread. How could this have happened? How could all those polls have been so wrong? The media went to work offering slim, even weak, explanations: When it came time to mark the ballot or pull the lever, too many Republicans just could not bring themselves to vote for a Democrat. Or, last-minute campaign advertising blitzes worked better for the incumbents than anyone could imagine; people just couldn’t put a “socialist” in the Senate, or an “Obama Democrat” in the governor’s chair. Two problems with this: First, Kansans had been electing Democrats governor more often than Republicans. In nine elections since 1978, Democrats were elected governor five times and Republicans four. Moreover, Republicans in heavily-Republican regions were marking a lot of ballots in 2014 for Democrats, especially paper ballots. (In Lindsborg, for example, Davis defeated Brownback handily in all four wards, and swept the city by an overall margin of 57-43 percent.)
ON JULY 25, the Topeka Capital-Journal published a letter from Vern McFalls, who had been a volunteer worker at a Topeka polling station. McFalls’ letter was to support a July 21 letter from Sam Hargis, who questioned why “two extremely close races suddenly swung Republican. These surges were just large enough to make the Republican a winner.” A sudden surge of Republican votes, at 8 p.m. and after the polls had closed, had swung the election to the incumbents. “In the gubernatorial race, we experienced unusual issues with voting machines,” McFalls wrote. “(Hargis’s) recent letter mentioned the voting machines showing a swing to Gov. Sam Brownback during the last 30 minutes of the election. We experienced the same. In fact, I believe Paul Davis was showing a small lead throughout until the polls closed. As we pulled tapes after closing our site to check for irregularities, we saw Brownback had pulled ahead in the last 10 to 20 minutes. Keep in mind the voting machines were acting strangely throughout the evening. After the close, at least one decided to not give up its tape output without “thinking about it,” is how I would put it. When I left more than two hours after the polls closed, it still was printing tape sporadically. Our poll manager and another person from the opposing party stayed, as is necessary. “Verizon wireless 3G sending units were in use for the first time to send data to centralized data collection points,” McFalls continued. “When I mentioned this to a friend who is an electrical engineer and builds boards for communications/ broadcast equipment, he rolled his eyes and said: ‘ What a simple but reliable way to throw an election.’“That evening was the strangest I’ve ever had working polls.
“Do I think the elections were hacked? You bet I do. It wasn’t just me either. Others as well voiced suspicion. Our poll manger was getting calls all evening from other poll managers who were having all sorts of problems. I leave it to you, but everything I heard and saw that evening pointed to hacked votes.”
WSU’s Clarkson is a certified quality engineer with a Ph.D. in statistics. She has said for months that in Sedgwick County her calculations from the November election revealed strange patterns in the voting data – odd enough to suspect that “some voting systems were being sabotaged, but that doesn’t mean that no other explanations are possible for these patterns.” Clarkson has been trying for months to ascertain what happened in Sedgwick County and across Kansas. She published initial findings in June in StatsLife, the magazine for the Royal Statistical Society and the American Statistical Association. But for complete analysis, she needs the voting records. Clarkson has argued that election results are a matter of public record. Sedgwick County officials have refused to release them, saying the voting records (numbers) are ballots and must remain confidential, although they contain no names. Remarkably, a judge has agreed. The voting machines in Sedgwick County are unique for Kansas because they leave a paper record of votes, known as Real Time Voting Machine Paper Tapes. Clarkson believes the records may hold clues to the election mystery.
Most Direct Recording Electronic voting machines (DREs) in Kansas and around the country do not leave a paper record; many counties and states, citing the added cost, have omittedthe paper record option. This is dangerous, says Pamela Smith, president of Verified Voting Foundation, a nonprofit agency created to safeguard elections in the digital age. “There is a cost for not knowing the results are right in each election,” Smith told the Lawrence Journal-World last month. “In our view, it becomes kind of corrosive of voter confidence because over time you can never be sure.”
KOBACH CONTINUES to fight Clarkson’s quest for the records, preferring to keep a tight lid on the matter. Clarkson believes an audit might answer questions whether and perhaps how, the 2014 state election may have been compromised. Meanwhile, the trouble with Kansas elections stays with us. It lies in a tenuous system of electronic voting that, at most polling stations, leaves no paper record and thus eliminates any possibility for a recount.
In addition, voting machine software is proprietary, exclusive to the owners and manufacturers; not even election officials can examine it for possible malfunctions. Thus, postelection audits of machines are not possible. This leaves Kansas elections available for tampering and open to corruption. The issue demands an open, comprehensive investigation. Or, do we prefer to look the other way and leave the matter to Kobach, the great defender of free and fair elections?
– JOHN MARSHALL