Questions of constitutionality

The media successfully challenge Election Day exit polling restrictions as courts reinforce earlier decisions striking down similar barriers.

From the Winter 2007 issue of The News Media & The Law, page 42.

By Rani Gupta

In 2006, one quarter of self-identified Florida Republicans voted for incumbent Democratic Sen. Bill Nelson over his GOP challenger, Katherine Harris. Seven out of 10 Nevada voters who believe the United States is safer now than before Sept. 11 voted for the Republican nominee for governor.

For decades, the media have relied on exit polls to provide a window into the public’s mindset on Election Day.

But in 2006, The Associated Press and the TV networks that collaborate to obtain the polls were forced to fight limitations on exit polling in two states and won.

It is not the first time media organizations have challenged exit poll restrictions. Since the 1980s, news organizations have successfully fought limitations in 10 lawsuits.

Invariably, judges have held in all the cases that the restrictions are an unconstitutional restraint.

The rules in question all required exit pollsters to stand a certain distance from a polling place. In the most recent lawsuits in Florida, Nevada and Ohio, the distance was 100 feet.

“It’s just pretty clear legally now that election officials can’t put undue restrictions on interviewing voters after they voted,” said Joe Lenski, executive vice president of Edison Media Research, one of two firms that AP and the networks have contracted with to conduct the polls.

With exit polls, one pollster is typically stationed outside each polling place. They approach voters after they have cast their ballot and ask if they would fill out a brief, anonymous questionnaire.

One problem with polling restrictions is that pollsters approach voters in a predetermined pattern, such as every fourth or fifth voter. The farther away a pollster is from the entrance, the more likely it is a voter will blend into a crowd or get into a car before she can be interviewed.

That disturbs the scientifically determined selection process and makes the polls less accurate. In places where interviewers stood 100 feet or more from the polling place in the 2004 election, error rates were more than double the rates of places where pollsters could stand within 25 feet of a location, according to an internal evaluation conducted by the two polling companies.

That research was cited by federal judges in the most recent cases, who said it demonstrated the real harm that an exit poll barrier would cause. By contrast, they noted that election officials in Nevada, Ohio and Florida had no real evidence that exit pollsters hindered voters.

U.S. District Judge Paul Huck in Miami, for instance, noted that none of the 5,090 voter complaints submitted by Florida voters regarded exit polls.

Problematic laws

Typically, polling difficulties arise when state legislatures pass new restrictions or secretaries of state issue new interpretations of existing laws.

In Ohio, media organizations had been allowed to conduct exit polls with no restrictions. But shortly before the 2004 election, the media learned that then-Secretary of State Kenneth Blackwell had advised local election officials that exit polling would not be allowed because of a statute that prevented loitering within 100 feet of the polls but did not specifically address exit polling.

U.S. District Judge Michael Watson barred officials from enforcing that directive during the 2004 election and in April 2005, Blackwell issued a new directive stating that the anti-loitering statute did not necessarily prevent exit-polling, but if pollsters were “loitering” or “congregating,” they could be violating the law.

In September 2006, Watson ruled Blackwell’s new order unconstitutional.

In Nevada and Florida, legislators had passed statutes that did not use the phrase “exit polling” but had language that seemed to apply. Nevada’s law prevented “speak[ing] to a voter on the subject of marking his ballot,” while Florida’s law barred “seeking or attempting to seek any vote, fact or opinion” within 100 feet of a polling place.

There are similar laws on the books in other states. But Dave Tomlin, AP’s assistant general counsel, said the media coalition — which comprises the AP, ABC, CBS, CNN, FOX and NBC — decided to focus its 2006 lawsuits on states where the barrier was at least 100 feet and where close, politically significant races were expected.

Tomlin said legislators pass such laws for a host of reasons. Some are responding to the requests of election officials who say they need the laws, he said, while others simply dislike exit polling.

That sentiment stems from the 2000 election, when networks called the state of Florida for Al Gore based on exit poll results, which some felt hurt voter turnout in the western part of the state, where polls close one hour later.

Tomlin said the networks have addressed these complaints by waiting until the polls have closed to call a state for a candidate.

Unhappiness in 1980

Susan Buckley, the New York attorney who argued the cases for the media, said it is unclear whether the legislators in these states intended to target exit polling in passing the laws.

“I happen to think — and this is speculation on my part — that when Florida and Nevada passed those statutes, they didn’t intend them to apply to exit-polling,” Buckley said.

Florida’s statute was passed as part of a sweeping election reform bill in 2005 — almost two decades after the same coalition of media groups successfully sued to overturn a Florida law that was almost identical to the 2005 law, except that the barrier was 150 feet instead of 100 feet.

That law was part of the first wave of exit poll barriers, which were passed because of anger over the 1980 election, which was called for Ronald Reagan when polls on the West Coast were still open. Many blamed the networks and their broadcasting of exit poll results, though President Jimmy Carter himself conceded the election early.

“A lot of West Coast states felt that exit polls had helped disenfranchise their voters,” Lenski said, “even though there’s no academic study that has proven that.”

Indeed, some legislators said they wanted to prevent the broadcasting of early returns when they passed barriers in the early 1980s. But those statements may have made it even easier for media organizations to get the barriers overturned because they made clear the rules were not meant to stop conduct, but speech.

“They didn’t like what the exit polls did,” Buckley said of legislators. “The problem wasn’t with the exit poller, it was with what was done with the data.”

In 1988, a federal appeals court (9th Cir.) ruled a Washington state law prohibiting exit polling within 300 feet of a polling place unconstitutional. In its opinion, the court said it appeared one motive for the statute was to prevent early returns from being broadcast, and that was an “impermissible” purpose of the statute.

The same year, federal trial court judges prohibited election officials from enforcing exit poll barriers in Minnesota, Florida, Georgia, Montana, Kentucky and Wyoming.

Since then, Buckley said, “you don’t see legislatures citing” dislike of exit polls as the reason for restricting them.

Some of the recently challenged laws do not even make oblique references to exit pollsters, such as the Ohio anti-loitering law that prohibits anyone from entering the 100-foot zone.

Tomlin suggests such laws may be in part “an attempt to draft a statute that doesn’t directly target exit polling.”

“With a generally applicable law that covers anybody within a certain distance, the feeling may have been that they could get away with it in the courts,” Tomlin said. “The court, we believe, saw that the law when applied to exit polling activities was still unconstitutional.”

Election officials who approved exit poll restrictions may have been emboldened by a U.S. Supreme Court case decided after the wave of 1980s court decisions striking down polling barriers. In 1992, the Supreme Court upheld a Tennessee law that prohibited soliciting voters and displaying campaign material within 100 feet of a polling place.

However, the judges in the recent exit poll litigation drew a distinction between electioneering and exit polling. The key difference, Lenski said, is that electioneering targets voters before they cast their ballots, while exit polling takes place afterward.

For 2008, the media groups will keep an eye on legislation and other political developments, but Tomlin said he is not aware of any states that pose a problem.

If any materialize, he said, “We will consider the possibility of action to challenge them.”

‘Part of the political debate’

But new laws and secretary of state opinions are just part of the problems facing exit polls, Lenski said.

Lenski noted that states such as Oregon have essentially changed their rules to prevent exit polling by having all their voters cast ballots by mail.

Nationwide, one quarter of voters cast their 2006 ballots by mail or in early voting, which has forced the polling companies to find other ways to survey those voters.

And as exit polls become a partisan issue in many places, Lenski said it is harder to obtain accurate results.

“Since 2000, the exit polls themselves have become part of the political debate and I think that does make it harder for us to get high response rates,” Lenski said.

Still, those involved in exit polling note that it is still rare for the media to have to resort to a lawsuit. Most times, conflicts are resolved by obtaining attorney general opinions or speaking with state and local officials.

“We try very hard to assure election officials that we are just trying to give voters a chance after they voted to express their opinion and inform the public how people voted and why they voted,” Lenski said.