3 min
Why Polls Are a Snapshot, Not a Forecast
The poll numbers will keep coming between now and Election Day. Keith Gaddie wants you to ask a different question: not who's ahead, but why. Gaddie is the Hoffman Chair of the American Ideal and a professor at Texas Christian University. He says the biggest mistake people make is treating polls like a horse race. A poll is a photo, not a prediction "A poll is just a snapshot of a sample of reality at a particular point in time, and they age out really quickly." The numbers show what one group of people thinks right now. The real work, Gaddie says, is figuring out why they think that way and how it shapes the choices they make. And those opinions are often shallow. Gaddie says that "most of the time, when polling, people haven't thought about a question until you ask them, so their opinions are immediate. It is a knee-jerk reaction to the question." Noise vs. a real trend Poll movement comes from two places. One is the sample itself. Gaddie explains that "any oscillation within 4 or 5 points, if it's happening all the time, that's just from the sample. That's just noise." A real shift looks different: numbers that keep moving in the same direction over time. Don't expect many minds to change. "There are very few true undecideds. Most people are out there actually seeking information to verify prejudice that they have or preference that they have." What does move an election is a true crisis. The 2008 financial meltdown helped lock in Barack Obama's first win, and Gaddie says crises like that tend to work against the party in power. October surprises, on the other hand, are rare. Gaddie says, "They can't be engineered or manufactured." Today's fragmented media makes it harder still. He says, "You can't create a singular story that moves in one direction." Why turnout is the real story Hardcore partisans always vote. The voters in the middle turn out when something gives them a reason, and you can't predict which party they'll back. This year, Gaddie sees a signal in Texas, where more people voted in the Democratic primary than the Republican one. He says, "Primary turnout was at a record level." And in his view, high turnout usually means one thing: "When you get high turnout, you're usually voting for change." The reason, he says, is that "anger and disappointment and desire are much more powerful motives." Presidential years pull more voters because the race is national and the choices feel simple. Midterms are harder because, as Gaddie puts it, "life gets in the way." He adds: "Once you get registered, you have to work really hard to not be able to vote, even in Texas." What to watch instead of the numbers Look at how each candidate presents themselves for the job they're running for. Then compare the reality each candidate describes with the one you see. Gaddie's reminder: "only you know what that is, and we don't all see the same reality." The election isn't over when the votes are counted Gaddie also has a warning about what comes after Election Day. He says, "This election will not be over until the last lawsuit." He believes there are "procedural concerns that everyone ought to have." He points to a line from political scientist V.O. Key that he thinks captures what's at stake: "If a democratic regime is to work successfully, it must be generally agreed that contestants for power will not shoot each other and that ballots will be counted as cast." The takeaway A poll number is where the story starts, not where it ends. Turnout, timing, trust in the process and the candidates themselves will tell you far more about how this midterm plays out. Covering the midterms? Click on Keith's icon below to connect with him. He can speak to: Polling and public opinion Midterm turnout Texas and Southern politics Voting rights and redistricting Democracy and trust in elections
