Bernie Sanders' unexpected win in the Michigan Democratic primary this week has made pollsters take a second look at the electorate they are supposed to be measuring. And with big states like Ohio and Illinois voting next week, surveyors are re-examining their data, says Michael Ramlet, co-founder and CEO of Morning Consult, public opinion polling firm based in Washington, D.C.
According to Ramlet, the response rates and sample sizes for election polling have plummeted in the past 20 years as fewer voters (especially young ones) use landline telephones. And while it is possible to conduct polls via cell phones, it is much more expensive.
Since the "new" primary and caucus system started in the 1970s, news outlets have analyzed each moment in every debate and election. Demand for polls increased with the demand for coverage, and even when conscientious newsrooms take polls as a general measure of a race, they risk a lot if they contradict the trend reported in numbers.
Michael Traugott, a professor at the Center for Political Studies at the University of Michigan, says the result is "horse race journalism," which emphasizes the dynamics of the campaign rather than covering the issues or how voters are actually responding to candidates' positions.
What you'll learn from this segment:
- What pollsters often get wrong, and why some communities are left out.
- How polls shape public opinion about elections and campaigns.
- The role that the media plays in validating incorrect information.
