Over and over again this election season, you are going to see polls that try to assess which issues people think are important. Sometimes pollsters ask an open-ended question so respondents can state their priorities in their own words. Sometimes pollsters list issues and ask people how important each one is. Occasionally, pollsters will try to make respondents rank issues.
All of these exercises have their uses. But what are they not useful for? Identifying the factors that actually matter to voters at the ballot box.
That’s the conclusion of a 2020 paper, “More Important, but for What Exactly? The Insignificant Role of Subjective Issue Importance in Vote Decisions,” by political scientists Thomas Leeper and Joshua Robison.
Leeper and Robison have two key pieces of evidence. First, they look at seven national election studies in the United States between 1980 and 2008, each of which asked voters’ positions on a bunch of issues, where they perceived the presidential candidates to stand on each issue, and how personally important each issue was to them. The big question is whether the distance between voters’ positions and the two candidates – i.e., whether voters were closer to the Democratic or Republican candidate – was a stronger predictor of how they voted if they said the issue was important to them.
The answer? Not really. “We find little evidence of a consistent role played by subjective importance,” Leeper and Robison write. Here and there, they saw the expected finding: Proximity to the candidates on Issue X had a stronger relationship to how they voted if they said that Issue X was important to them. But at most the evidence was “very qualified” across a wide range of issues and elections.
Second, Leeper and Robison did an experiment where respondents chose between hypothetical candidates who varied randomly in their issue positions. Once again, the question was whether respondents were more likely to vote for the candidate closer to them on an issue when they said that the issue was important to them. And once again, Leeper and Robison didn’t find much evidence for this.
These findings fit with a long vein of research showing that people are not good at articulating the reasons for their choices. So they may say – and honestly feel! – that an issue is important. But when they say that, they are not revealing that this issue will affect how they vote. The same problem emerges even when pollsters directly ask people whether something would change their vote.
You’re going to read a lot of coverage over the next several months about what truly matters to voters in 2024. A lot of this discussion will rely on polls that ask for people’s subjective statements of their priorities (“Oh, the biggest issue for me right now is inflation/abortion rights/Gaza/whatever.”)
Just don’t assume that those are the issues that will actually affect their choices on Election Day.
