As Election Day for the midterms inches closer, the polling industry across the country is focused on getting accurate predictions about what will happen on Nov. 3. Pollsters have struggled to gain trust back from the American people since off-target forecasts in the 2016 presidential election.
In recent years, midterm election polling has proven to be more accurate than presidential election polling, though challenges remain. One concern is partisan nonresponse bias, which occurs when supporters of one political party participate in surveys or polls at a lower rate than supporters of another party, skewing the results.
In the traditionally purple North Carolina, polls have long been underway to determine who will take the open U.S. Senate seat and House seats, along with the state House and Senate. Most polls target between 700 and 1,200 people, with a margin of error in most pre-election polls between three and five points.
We discuss polling methodology, margin of error, how to interpret poll results and more on this episode of Charlotte Talks.
GUESTS:
Chris Cooper, director of the Haire Institute for Public Policy at Western Carolina University and author of the book “Anatomy of a Purple State: A North Carolina Politics Primer”
Jason Husser, associate professor of political science and policy studies and director of the Elon Poll
Ruth Igielnik, polling editor at the New York Times
