By CLAY WIRESTONE
Kansas Reflector
You want to talk about that new poll showing Kansas as a potential swing state? OK, let’s talk about that new poll showing Kansas as a potential swing state.
The annual Kansas Speaks survey from the Docking Institute at Fort Hays State University showed Democratic presidential candidate Kamala Harris a mere five points behind Republican candidate Donald Trump. Given that the survey covers traditionally conservative Kansas, political pundits pounced.
“BREAKING: Trump only winning by 5 In KANSAS,” wrote legal commentator Tristan Snell.
“If Harris is only down 5 to Trump in Kansas – election night is going to be very good for Democrats,” added Obama campaign and administration alumnus Tim Fullerton.
“…holy sh*t?” exclaimed journalist and lawyer Seth Abramson.
Let’s step back from excited progressive reactions for a moment and actually talk about the poll, what it means and whether Harris has a shot in Kansas. (Spoiler alert, she almost certainly does not). We found a great person to help us break things down: Brett Zollinger, director of the Docking Institute itself. He sat down with Kansas Reflector staff on Tuesday afternoon to give background on those surprising results.
First off, we should understand that the data presented comes from a survey, not a poll. Kansas Speaks primarily tracks public opinion on issues rather than candidates. It gathers a representative panel of Kansans and collects information from them via a carefully designed survey online.
“The presidential election is a very minor focus for us in Kansas Speaks,” he said. “We are far more focused on policies, issues that are going to be relevant to Kansans, likely to come up in the legislative session, that sort of thing. But every four years, we get this unique opportunity to see what our survey methodology bears out in terms of some actual voter decisions in the state in the presidential election. So in 2020, we were within 0.2% actually, in the spread between Trump and Biden with our survey panel methodology.”
For reference, in 2020 Trump finished at 56.2% in Kansas, while President Joe Biden reached 41.6%, a 14.6 percentage point gap. The 2020 Kansas Speaks survey found 52% support for Trump and 37.6% for Biden, a 14.4 percentage point gap.
That earlier accuracy was much remarked upon by online commentors. But Zollinger wanted Kansas Reflector readers to understand that the modeling and weighting decisions made for the Kansas Speaks survey are not necessarily the same ones made by political polling firms. They’re playing different games.
“If we were actually trying to measure, with a lot of predictive ability, the presidential vote turnout, we would probably engage in a fuller methodology common to the survey industry, especially the presidential polling industry,” Zollinger said. “We don’t. But we do winnow it down to registered voters who say that they plan to vote, which is not a bad methodology necessarily for doing that. And the survey panel is what it is. These are folks who have opted in to be surveyed. It’s not uncommon to use these today. Pew Research has its own big one. Gallup has one.”
Researchers looked at various demographic factors, he said, “and we did pretty much a simple proportionate weighting.” (You can dig into the crosstabs on page 96.) They did the same with their 2020 results.
“We took the proportion in the sample who fit a certain age profile, and we adjusted it to the state’s age profile of adults, same with gender, same with education. … We believe it is the most reasonable approach to trying to make up for some underrepresentation by some age groups in the panel and some underrepresentation by certain education levels,” Zollinger said.
Which brings us all down to the basic question: Do we believe the survey results?
I do, with caveats.
Kansas has changed. In the eight years since I returned to the state, I’ve watched as a motley mix of Republicans, Democrats and independents have rejected Trump and extreme ideologies. I’ve watched a constitutional amendment that would have allowed the Legislature to ban abortion fail by an 18-point margin. I’ve watched voters twice elect Democrat Laura Kelly as governor. I’ve also heard, over and and over, and read, in repeated email messages, that Kansans don’t like the current political atmosphere.
Now, that doesn’t turn us into an immediate swing state. It doesn’t guarantee a Democratic Legislature anytime soon. But it does, at least from my perspective, suggest that Republicans can’t count on a blank check from Kansas voters.
Trump will almost certainly win Kansas. But that margin will matter.
Alex Middlewood, a political scientist at Wichita State, weighed in on the survey results via a Tweet thread. She said the results showed that Trump has been losing ground in Kansas over the past eight years.
“His vote share in 2020 was lower than it was in 2016, and 2024 is very likely going to be even lower. Make no mistake, Trump is going to win Kansas,” she wrote. “However, any shift in the margins, like we’re seeing from these results, will likely have major impact on the state legislature races. Ds are working hard (and spending lots of money) to break the R supermajority. These results should give them hope.”
More Kansans speak
I’ve been keeping track of interesting presidential endorsements in Kansas news outlets over the past couple of weeks. You might not be able to read an endorsement from the Washington Post, but these plucky Kansans have a lot to say.
“Rep. Steven Howe reflects on voting for Trump after earlier criticism,” by Steven Howe, Salina Post, Oct. 26.
“My fellow conservatives, we have to split the ticket to save the US Constitution,” by David Mastio, Kansas City Star, Oct. 29.
“Oops, I voted for Hitler…again,” by Dane Hicks, Kansas Informer, Oct. 30.
“Kamala Harris has what it takes,” by Susan Lynn, The Iola Register, Oct. 25.
Clay Wirestone is Kansas Reflector opinion editor. Through its opinion section, Kansas Reflector works to amplify the voices of people who are affected by public policies or excluded from public debate. Find information, including how to submit your own commentary, here.