Exploring The Link Between Weather, Elections
ANDREA SEABROOK, host:
Now, let's move on to something that really matters to voters, the weather. We've dug in to the swing state forecasts at weather.com, and the verdict is, a few showers in Virginia and North Carolina, maybe a little snow in the rocky mountains, but, otherwise, gorgeous in the battleground states from Nevada to Missouri to Ohio to Florida. And sunny skies could mean storm clouds for Republicans. That according to a research paper titled, "The Republicans Should Pray For Rain: Weather, Turn Out, and Voting in U.S. Presidential Elections." Brad Gomez helped write that paper. He's an assistant professor of political science at Florida State University.
Dr. BRAD GOMEZ (Department of Political Science, Florida State University): For every inch of rain that a county receives above its average rainfall, turnout decreases roughly about one percentage point. And that reduction in turn out was for the benefit of the Republican party.
SEABROOK: And why is that? Did you determine that from your study?
Dr. GOMEZ: Yeah. One of the things that we know about voter turn out is that turn out comes with cost for someone who can easily take off from work and go to the polls if they have their own sources of transportation. All of these sorts of circumstances make it easier for some higher social-economic status voters to go to the polls, and those voters have a tendency to vote Republican.
Now, for some voters, you know, if you're working a job or maybe even two jobs, and your boss won't let you off to go and vote, well, you might decide that, you know, it's not worth the effort. Maybe you have to take a bus to the polls. If you've got to stand in line for the bus, and it's raining on your head, maybe you decide, well, we'll let someone else determine who the next president is going to be.
SEABROOK: OK. So, how did you even test this? In a scientific way?
Dr. GOMEZ: Sure. But what we did was we gathered voter turn out data for each of the 3,000 counties in the continental U.S.
SEABROOK: Every single county?
Dr. GOMEZ: Every county in the U.S. for every presidential election from 1948 to 2000.
(Soundbite of laughter)
Dr. GOMEZ: We then matched that data with weather reports from over 20,000 weather stations located in the U.S., and then we tested our model against standard variables that we would put in a normal turnout model, such a socio-economic variables and legal variables, such as what the electoral registration laws are. And that's how we got our results.
SEABROOK: Wow. That's a lot of data.
Dr. GOMEZ: It is a lot of data and particularly for social science.
SEABROOK: Are there any elections over that span of time that you talked about that you can point to the specifically - you think the outcome was affected by the weather?
Dr. GOMEZ: Yeah. What we found was that, in our simulation of 1960, which was an incredibly dry day, we simulated what the turnout would have been had it essentially experienced average rainfall, and our results predict that Richard Nixon would have received an additional 106 electoral college votes.
Dr. GOMEZ: That would have given him 55 more than he needed to become president. And we would have experienced a Richard Nixon presidency perhaps eight years earlier than we did. We also found that, in Florida in 2000, Florida experienced a little bit of rain in the panhandle area around Pensacola, and had it been a dry day in Florida, Al Gore would have won the state. So, our results give the Democrats another reason to complain about Florida in 2000.
SEABROOK: Professor Brad Gomez, he joined me from member station WFSU in Tallahassee, Florida, where there's a 10 percent chance of rain on Tuesday.
NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR’s programming is the audio.