Crunching the numbers of an election

Pollster Scott Rasmussen.


Kai Ryssdal: The closer elections get, the less you can get away from 'em. Mostly in a horse-race sense -- who's ahead, who's behind, where the polls are. And in a way, makes sense, because actual elections -- not campaigns, but the elections -- are about counting votes. Today, we continue our series about the people behind the machinery of politics.

Commentator and pollster Scott Rasmussen talks about crunching numbers.

Scott Rasmussen: Being a pollster in the final weeks of an election season, it's a bit crazy. Every night, loads of new data come in from around the country that needs to be processed, reviewed, reported and put into context. Each new poll release brings forth a torrent of comments. Those who like the results tout our brilliance, and those who don't complain that some other poll is better.

We answer hundreds of press inquiries during the closing days of the season. And if we tell a reporter the race is too close to call, well, some still insist on knowing who's going to win. We patiently explain that if we knew, we wouldn't be calling that race a toss-up. Other comments come from the losing candidates and their campaigns and they just don't want to believe the numbers.

In this frenzy, my focus is always on the data. We believe in our numbers, because we believe in the system we've built. The greatest danger for pollsters is to overstate what the numbers really say.

Discipline is key, because advocates try to sway what we say about the numbers. This craziness culminates on election night, which a long night for the entire staff as we root for our poll results. If we show a Republican ahead in one race by two and a Democrat ahead in another by three, we root for the Republican by two in the first race and the Democrat by three in the other. After the election, the data flow subsides and so do the questions. During this briefest of downtimes, I make it a point to get away for a trip with my wife, because she hasn't received the attention she deserves.

But then, the real work begins. We do our most important work after the election. We look at the results we projected, we compare them to what really happened and we learn from both what we did right and wrong. We establish the protocols and procedures that will guide us during the next season. These lessons build the foundation so we can have confidence in the numbers for the next pre-election frenzy.

Ryssdal: Scott Rasmussen is the co-author of the book "Mad As Hell: How the Tea Party Movement Is Fundamentally Remaking Our Two-Party System." Tomorrow, in our series, David Frum.

We'd love to hear from you about this upcoming election. Send us your comments.

Log in to post3 Comments

To the two commenters before me: I guess we'll just have to see the election results. As for the Rasmussen poll being a non-neutral poller I'd say to anyone who believes this: (a) sorry your side seems to be losing (b) look at the past performance of the Rasmussen polls-- very accurate, historically. He in fact predicted Scott Brown's win in my state of MA; one of the few to do so. Is this what you mean, prior commenter, by asserting that Rasmussen "consistently polls to the right of most other organizations"? Maybe what it actually means is that some polls are biased to the left; many certainly were in the 2010 Massachusetts Senate race, and they were wrong. The one benefit of judging pollsters versus pundits is that, for the pollsters, the proof is in the post-election results and how well they match the pollster's predictions.

The real upset in Tuesday's election may be polling practices themselves, an institution built on land-line telephones and people who answer them with the time to chat. Younger, more progressive, and more tech-savvy voters are using technology to screen calls, see no need to have land-line telephones in their homes, and as a result, I believe, are consistently under-represented in today's polling numbers. We'll see, next Wednesday morning.

All that talk and no explanation why Rasmussen consistently polls to the right of most other organizations- he is not a neutral pollster so why is he being allowed to represent himself as such, in detail, without challenge? Why is the pundit up next tomorrow David Frum, former speechwriter to Bush? Is this a balanced approach to the topics? Maybe by Fox News standards.

With Generous Support From...