Marist Poll Connects Students to National Discussion of How Public Opinion Data Shapes Journalism
Photo by Jakub Zerdzicki via Unsplash
On April 14, the Roper Center for Public Opinion Research at Cornell University hosted its semi-annual virtual event, the Roper Roundtable, to discuss how public opinion data shapes reporting, storytelling and analysis.
The event was hosted by Ariel Edwards-Levy, a polling and analytics editor for CNN.
Edwards-Levy was joined by Philip Bump and Nathaniel Rakich. Bump is a senior data editor and columnist at CT Insider, while Rakich is the managing editor at VoteBets and is a former senior editor/elections analyst at FiveThirtyEight.
The conversation centered around polling, with the main theme being that polling is a great tool, though it is not a total substitute for reporting.
“Polls are there to help us understand what people are thinking,” said Edwards-Levy. “They’re not there to tell us what the story is.”
Due to a series of recent political polls turning out to be incorrect, most notably during the 2016 U.S. Presidential Election, public trust in polls has significantly decreased. Many people now question the validity of polls and how accurately they measure public opinion.
“The biggest problem isn’t necessarily that polls are wrong, it’s that people don’t actually understand what they’re showing,” said Bump.
Bump attributed this to key factors such as margin of error, misunderstanding of what the poll is actually measuring, and small shifts being treated like major swings.
In modern journalism, there is pressure on news organizations to continuously produce news 24 hours a day. This is what often leads journalists or social media influencers to jump to conclusions and dramatize minor shifts in polls.
The public misconceptions about polls can oftentimes put journalists in a precarious position of wanting to use the polls as evidence, but also wanting people to trust their writing and respect their knowledge.
All of the panelists agreed that the responsibility falls on the journalist to make sure that people can understand what the polls are showing and how credible it can be when measuring public opinion.
“Data is most useful when it helps explain why something's happening,” said Rakich. “It’s not just about reporting numbers, it’s about giving them meaning.”
The strongest journalism stems from the use of data to support facts.
Doing this allows journalists to provide context and give their work credibility. With the media landscape continuously evolving and there being more data and polls in the world than ever, clearly communicating the meaning of data has never been more important.
“There’s a lot of that sort of meta service journalism, in which you're helping people understand how to be better at understanding journalism itself,” said Bump. “We see a lot more of these sorts of analyses of what polls are actually showing and what you can learn from polls.”