Since starting the Marquette Law School Poll in 2012, Charles Franklin has predicted every major election result. Is he lucky or good?
How do you manage to keep your personal opinions out of poll analysis?
Just as a journalist can’t claim to be completely above any personal feelings about issues, likewise a pollster can’t claim to be a pure, uninterested observer. We are never going to ask questions that everyone universally agrees are completely neutral. You aim for a minimum of adjectives that imply an opinion.
Ever worry that some political operations will manipulate the findings?
It’s entirely possible that they do, but there is another campaign that has exactly the same information. One of the things that we have focused on is transparency. We post the entire questionnaire, the results to every question, the demographic breakdowns for every question, the methodology, analysis of what we did, and how we did it.
What is the biggest mistake you’ve made as a pollster?
Iowa in September 2008. I was co-director of the Big 10 Poll. We showed John McCain just slightly ahead when every other poll showed Obama winning Iowa handily. That was in the days when we had not yet switched to including cell phones in everything. Young people with landlines in Iowa are more Republican than young people with cell phones in Iowa. In our October data, we correctly forecast the Iowa outcome.
What makes a credible, useful opinion poll?
That is very hard to measure. In every one of our polls, we have had some issues where the balance favors the Democratic side of the issues and others where the results favor the Republican. I think it is valuable because it covers a range of issues, and we see a range of partisan and political responses across those issues rather than results that always favor one side and never favor the other side.
You’re often criticized by whomever is down in a poll you’ve taken. Have you developed a thick skin?
You cannot do public polling and have a thin skin about it because you are guaranteed to have irritated at least one side with every poll, and if you are particularly lucky, you can manage to irritate both sides with a single poll.
Do you vote?
I do not. My career has been about understanding public opinion and voting; it has never been working for candidates or campaigns. I am somewhat comforted that none of the elections I might have voted in were decided by one vote.
How did you get into polling?
It’s one of those strange childhood things. I grew up in a little town in Alabama. In the 10th grade, you could build your volcano for the science fair or you could do something else. So I decided to do a survey of second-, fourth- and sixth-graders on Saturday morning cartoon shows. That actually turned out to be a fairly creative thing to do. It got a pretty positive reaction from the judge. By the time I was into college, I was hooked on political science, and this seemed like just the thing to do.