About 20 Wisconsin volunteers got coached in the ways of citizen exit polling in a telephone conference call last week led by Sally Castleman, chairperson of the Election Defense Alliance, a nonpartisan national group that monitors elections.
"Plan for the weather. Bring snacks, water."
"Dress professionally, or what I call 'casual spiffy.'"
"Bring your cell phone and make sure it's charged."
"Don't block the flow between the doorway and the way people walk back to their cars."
Some 200 volunteers from Madison and across the state helped conduct citizen exit polls in the recall elections Tuesday that sought to unseat six Republican senators, according to Adam Porton, an organizer with Wisconsin Wave. The grassroots group, a project of the Madison-based Liberty Tree Foundation, helped recruit volunteers for the task along with the Election Defense Alliance. This was the first time that citizen exit polling was used in a major election in Wisconsin.
Volunteers will be out again Aug. 16 when two Democratic senators face recall.
The stakes going into Tuesday's election were high: Democrats needed a net of three seats to take back control of the state Senate and break the Republican stranglehold on state government, and they got just two. The results also put into question whether Democrats will proceed with a recall effort against Walker next year.
The purpose of citizen exit polling, say proponents, is to monitor the "integrity" of elections that rely on computerized machines to tally votes. Concerns about the validity of electronic voting machines in the 2004 elections, as well as the mishandling of some 14,000 ballots by the Waukesha county clerk in the recent Supreme Court race, have brought the issue to the fore in Wisconsin and elsewhere.
The ultimate goal, says Jonathan Simon, cofounder of the Election Defense Alliance, is to "draw attention to, expose and change our vote-counting system, which is at the very least vulnerable to manipulation."
Simon discussed the plans for citizen polling with the Government Accountability Board, which issued a July 18 letter verifying that exit polling is legal in Wisconsin as long as it does not interfere with the "orderly conduct of the election and nothing distracts voters from exercising their right to vote at the polls on Election Day." A dry run during the July 12 recall primaries landed citizen pollsters in a bit of hot water for handing out questionnaires that looked like ballots, says Reid Magney, spokesman for the GAB. That's a no-no in Wisconsin.
In his letter, GAB director Kevin Kennedy recommends that exit polling - not to be confused with electioneering - be done outside the building. The distribution of election-related materials is prohibited within 100 feet of a polling station.
Citizen exit polls were conducted Tuesday at select polling locations in all six Senate districts where Republican lawmakers were facing recall. The volunteers asked voters as they left a polling site to fill out an anonymous questionnaire identifying which candidate they voted for. Voters were also asked to sign a form indicating that their questionnaire accurately reflects their vote on the ballot. Workers signed forms testifying that they did not tamper with the questionnaires.
Some problems were reported Tuesday. Magney says that a few voters called the GAB, upset that they had been handed questionnaires asking who they voted for. Village of Butler officials also moved exit poll workers to the sidewalk because there was no room at the polling station entrance, he says. This made it more difficult for volunteers to contact people, though the work of the poll workers was not shut down, he adds.
The questionnaires, which were hand-counted immediately after the polls closed, will be used to detect disparities with the official vote count, say organizers.
Castleman of the Election Defense Alliance says citizen polling is a necessary check on the system because there is no way to verify the accuracy of electronic machines, which she says are vulnerable to mistakes, malfunction and manipulation.
"It is one of the few tools available to us as citizens," she told volunteers during her conference call.
Magney counters that his agency conducts random audits of electronic machines to check for accuracy. He also notes that there was very little difference between the recount tally for the Supreme Court race and the official machine count. "That was a huge, real-world audit," he says.
Exit polling dates to the late 1960s and was largely the creation of Warren Mitofsky, a former U.S. Census Bureau statistician who was then polling director for CBS News, says UW-Madison political science professor Charles Franklin.
The idea of exit polling is that you will get a sample of actual voters as they leave the polls; you never know if those "likely" voters polled pre-election will actually show up on Election Day. You can also record people's reasons for voting immediately after they cast their ballots.
Other networks quickly followed Mitofsky's lead, using the exit polls not only to explain why people voted the way they did, but, increasingly, to call elections, says Franklin. The mounting costs of this polling ultimately led to the creation in the 1990s of a consortium that now conducts exit polling for major media outlets.
Franklin says the accuracy of exit polls depends on the random selection of polling locations and voters. He says he's not familiar with the work of the Election Defense Alliance, which since its founding in 2006 has conducted citizen exit polling at 60 sites in eight states. But he says relying on exit polls as a check on the system "places enormous faith in the ability of exit polls to properly sample voters as they come out of the system."
Organizers from the Election Defense Alliance and Wisconsin Wave readily acknowledge they are not able to poll a representational sample of voters or control for response bias. "It's a very weak tool as far as bringing any validation to the election," says Simon. But they say the polling can flag potential discrepancies in the official vote count.
"We seek to confirm the official election results that will be reported on election night," says Mary Magnuson, state coordinator for the Election Defense Alliance. "Any data that we collect is used to double-check that our votes were recorded, counted and published accurately."
If discrepancies do materialize, Simon says volunteers would take it to the next step by launching a door-to-door canvass of voters.
The larger goal of citizen exit polling is to build awareness about the vulnerability of electronic voting machines and the vote-counting process. Ultimately, activists would like to see a return to a system of counting paper ballots by hand.
For those interested in transparent government and election integrity, voting is no longer enough, says Simon.
"If you really want to earn your democracy, not assume it, it calls for a bit more participation than every other year going down to vote."
Madison resident Maggie Thomas led a team of volunteers Tuesday at a polling station in Pardeeville, where Republican incumbent Luther Olsen successfully fought off a challenge from Democratic Rep. Fred Clark. Although much of the energy for the recall elections stems from reaction to Walker's GOP agenda, which included eliminating collective bargaining rights for most public employees, Thomas stresses that the citizen polling efforts are nonpartisan.
"It doesn't matter if you're Democratic or Republican if your vote doesn't count," she says. "It's not about right and left. It's about right and wrong."
Thomas says she expects efforts to educate citizens and reform the voting process will continue after this month's recall elections: "It feels like the birth of a movement."