In the busiest month of the election cycle for pollsters, several in the business say they won’t be trusting the numbers coming out of crucial swing states like North Carolina or Georgia amid the devastation from Hurricane Helene.
More than 100 people have died, about 2 million people still don’t have power and many have lost their homes. If any polls are currently underway, experts fear the results won’t accurately represent the current political climate.
“If anyone is doing phone polling now, I would have to question [the results],” said David McLennan, director of the Meredith College poll. “If I had been in the field while the hurricane was going on or immediately after, the results would be shaped by what’s happening.”
The Center for Survey Research at East Carolina University, which has polled heavily across North Carolina in the last month, had plans to expand into Georgia this week but has canceled that poll for the foreseeable future. Raleigh, North Carolina-based Meredith College, which published poll results days before the hurricane began, will likely add questions about the disaster’s effects to its planned poll for late October. And Siena College, which partners with The New York Times for its polling, told NOTUS that it has no plans to conduct polls in North Carolina until voters have recovered from the disaster. It’s just not at all clear when that will be.
The result: Crucial questions about how those in affected areas are engaging with one of the most consequential elections in history will be unanswerable until after Election Day.
Most polling averages show North Carolina as an even tie between Kamala Harris and Donald Trump. States like North Carolina and Georgia have proven critical to building Democrats’ path to victory.
It will also be difficult to gauge voters’ feelings about disaster response. Hurricane Helene completely destroyed towns across the western part of North Carolina, including the Democratic stronghold of Asheville, as well as parts of South Carolina and Georgia.
“There’s not enough academic research out there to really kind of say this is how it’s going to move one way or another,” McLennan said. He is currently mulling some sort of question for his late October poll that asks whether Helene has shaped voter opinions on the candidates and the issues, but he’s not sure exactly what that might look like.
“I just don’t have anything solid to draw on. I’ve been doing this here at Meredith for 10 years, and I’ve never had this kind of thing happen. It’s just so unprecedented,” he said.
At East Carolina University, director of polling research Peter Francia was still processing how to respond in real time during a conversation with NOTUS. While Francia first said that he had no plans to add a question about the hurricane to future polls, he seemed open to the idea by the end of the conversation.
The only thing Francia was certain of? His ambitious polling schedule for the next few weeks would have to be heavily cut down. “At least part of what we do, some of it does involve calling people on landlines, the ways that we get some of the older folks to do the polls,” he said. “If people are unable to take part in a poll that way, that’s problematic.”
Francia and McLennan are concerned that the effects from the hurricane (both in terms of how people think and in terms of difficulty reaching a good sample) will increase the likelihood of error in the polls. While pollsters speculate that Hurricane Katrina and Superstorm Sandy caused shifts in voter preferences (away from then-President George W. Bush, in the Katrina instance, and toward then-President Barack Obama, in the case of Sandy), those shifts were hard to parse until after the election.
“Polling is not as easy as it once was, for a whole host of reasons, and this adds a whole new layer of complexity for those of us still trying to do it,” McLennan said.
—
Anna Kramer is a reporter at NOTUS.