Is Hillary Clinton’s so-called “firewall” in Nevada slipping away? Is Bernard Sanders gaining traction in the Silver State after his dominating win in New Hampshire?
Pollsters have no idea.
With the Nevada caucus only four days away, only six public polls have been conducted in the last year parsing out who the winners or losers may be in the Silver State.
Before the first-in-the-nation caucus of Iowa, on the other hand, nine polls were conducted in January alone, and about 100 were commissioned in the year prior. In the weeks leading up to the New Hampshire primary, residents were polled nearly 50 times, records show.
The last major public survey of Nevada voters was conducted four months ago by CNN/ORC, long before the Vermont Senator challenged Mrs. Clinton for the win in Iowa and beat her by double digits in the Granite State. In that survey, Mrs. Clinton bested Mr. Sanders by 16 points.
A more recent poll commissioned by the conservative website Washington Free Beacon this month found the race had tightened considerably in Nevada, with Mr. Sanders and Mrs. Clinton tied at 45 percent apiece. The poll had a wide sample of voters, but the right-leaning website that commissioned it also littered the survey with questions about Mrs. Clinton’s trustworthiness, which may have swayed its outcome.
SEE ALSO: Bernie Sanders’ frumpy image combined with message assists his popular appeal
“A few words about all of these polls on the presidential race in Nevada: Don’t believe them,” wrote Jon Ralston, who runs an all-things Nevada politics website “Ralston Reports,” on his blog.
The tricky situation has the Clinton campaign, which just last month was touting its massive lead in the Silver State and calling it part of her post-New Hampshire primary firewall, now hedging its bets.
Campaign spokesman Brian Fallon now says Nevada is “a state that is 80 percent white voters,” suggesting that makes it a more natural Sanders stronghold than they’d first portrayed it. Mrs. Clinton also canceled an event in Florida to spend an extra day campaigning in Nevada this week before the caucus.
Mr. Ralston wrote that the problem for pollsters is trying to predict voter turnout and registration, which in Nevada can happen the day of the caucus.
For example, in 2008 about 118,000 Democrats turned out to caucus compared to 44,000 Republicans that presidential cycle. Four years later, Republicans’ numbers shrunk even further to 33,000. Democrats didn’t have a competitive caucus that year, as President Obama was running for re-election.
Those numbers are extremely low compared to the Iowa caucus, where more than 180,000 Republicans turned out this year and 171,109 Democrats.
SEE ALSO: Hillary Clinton barks like a dog in attack on Republicans
Nevada’s population, unlike New Hampshire and Iowa, is transient, making it tough for pollsters to know the people they’re calling are actually going to be in the state on caucus day. In addition, many don’t have land lines, making them tough to contact, or the polls are more expensive to conduct because they need to include cellphone contacts. Same-day registration also makes it hard for pollsters to predict who will turnout for the election.
Then there’s the dynamic of a caucus where many people who may go into their precincts with a preconceived notion of who they will be voting for, but then leave it having voted for another — swayed by their neighbor or boss that their initial choice wasn’t the right one.
“Nevada is a caucus state which is harder to do because of low turnout and the getting the county participation numbers correct,” said Cliff Zukin, a Rutgers University professor and former president of the American Association for Public Opinion Research. “It’s also a long caucus and pollsters may worry about getting it wrong and so be less inclined to try.”
Demographics also play a part — Latino voters respond less frequently to initial polling calls, pollsters have found, and the surveys need to be conducted in both English and Spanish, making them more expensive. Not properly polling the Latino vote skewed many polls taken in 2010, when most forecasted a tighter Senate race between Republican Sharron Angle and Democrat Harry Reid.
“Nevada is one of the hardest states to poll,” said Samuel Wang, a Princeton University professor and polling expert. “The reasons have to do with the difficulty of getting a representative sample of voters. According to Census data, it has one of the highest population of people who moved there recently. Many people are cellphone-only and are likely to have out-of-state area codes. In addition, Nevada has a large Hispanic population, and Hispanics are relatively hard to reach in polls.”
Mr. Wang added any Nevada poll that is published should be taken “with a large grain of salt.”
This election cycle voters of color will make up 39.4 percent of Nevada’s electorate, while Latino voters are expected to make up 21.2 of the voting population, according to the Center for American Progress Action Fund.
Because of these factors, many polling firms have decided to sit this election cycle out.
Mason-Dixon Polling and Research, which has polled the Silver State before for the Las Vegas Review-Journal, won’t be in the field this election cycle, and Brad Coker, the managing director of the polling firm, said the difficulty and high cost of polling a caucus state may be keeping the number of polls low.
“You’re dealing with a caucus, and they’re tricky because there’s such low turnout affairs,” said Mr. Coker. “You have people who say they’re going to turnout, but when Saturday comes along they’re doing something else and not showing up and that throws your results off.”
He added the number of “day sleepers” is another unique problem in Nevada, in that most blue-collar workers are employed during the night shift at casinos are tough to get a hold of and then you do contact them during the day there’s a 50-50 chance of them being upset.
• Kelly Riddell can be reached at kriddell@washingtontimes.com.
Please read our comment policy before commenting.