When election polls are released by Franklin & Marshall College, they often make the national news.

But Berwood Yost, director of the Franklin & Marshall Center for Opinion Research, stays behind the scenes.

It's Terry Madonna, director of the Center for Politics and Public Affairs at the college, who is the public face of the team.

"I am a researcher and methodologist," explains Yost, 47. "My job is to develop the questions and approaches and do the analysis. Terry goes out and talks about our findings."

Yost, who lives in Wrightsville and is a married father of two college-age kids, has been interested in polling since he was a student at Penn State.

"I started as an undergraduate and I never stopped," he says. "I was always interested in trying to understand why people behave the way they do."

He joined Madonna at Millersville University in 1994, and in 2003, the team moved to F&M.

Is this your favorite time, in the middle of a presidential election?

It used to be. I think the perpetual campaigning and the non-stop media attention have made it less interesting. And so many people are doing polling. Some do very good work, and some don't, but they all get treated the same, which makes it less fun. There are way too many polls. What Terry and I did 20 years ago used to be unique and is now commonplace.

The consensus seems to be that Mitt Romney won the first debate. Has that affected the polls?

It looks like there was a bounce in some overnight polls, but we'll have to see if it's sustained. These campaigns are so dynamic, I suspect the job numbers (which were released Friday) may have a big effect.

But you know, the national numbers don't matter. It's all about the Electoral College. You have to be careful: National polls are all well and good, but that doesn't tell you who is going to win. In the Electoral College, Obama continues to have an advantage.

Do the media lean too heavily on polls when covering the election?

Polling is a tool, and like any tool it can be misused. You can really understand a race and a campaign through polling. But most people don't use it for its power, they use it just to say who's ahead and who's behind. Some try to get into the underlying dynamics of the race through the polls, to clarify what is going on, but too often, it just gets boiled down.

Computers have helped improve the accuracy of the polling process, but on the other hand, they've made it easy for people who don't have a lot of skill to do this kind of work. The best pollsters understand the science behind it, the statistical adjustments, how to write questions, the topics they are studying.

Some critics have complained that pollsters have been biased against Romney. What would you say to them?

I hear it every election cycle; it's nothing new. We did our first statewide polling in the 1990s, and every poll we've done makes someone happy and someone unhappy.

We get complaints from both parties, and sometimes the complaints change in the course of the same campaign.

Is there potential for bias in a poll? Of course. But historically, the tracking is pretty good.

For me, the bottom line is reputable polling organizations stake their reputation on what they produce, so you do everything in your power not to have any bias. What motivation would there be for the entire industry to get it wrong?

Who's going to win Pennsylvania? Barack Obama or Romney?

It looks like Barack Obama is going to win Pennsylvania. We have been doing polling for over a year and Romney has never led. Obama's job performance numbers aren't great, but they are OK. He's well-positioned to win the state. Neither campaign is spending a lot of money in the state, but if they start doing that, I reserve the right to rethink my position.

jholahan@lnpnews.com