Data for Good Exchange 2018: Combining Data Science and Journalism to Increase Census Response Rates

(L-R): Justin Hendrix, Executive Director, NYC Media Lab; Mara Abrams, Managing Director, Census Open Innovation Labs, U.S. Census Bureau; and Mark Hansen, Professor, Columbia Journalism School/Director, David and Helen Gurley Brown Institute for Media Innovation

The U.S. Census faces two challenges: First, most people don’t understand why the census is important. The second challenge is related to the first: Because people don’t understand the importance of the census, they don’t always fill out the survey. In fact, during both the 2000 and 2010 Census, only 74 percent of households in the U.S. filled out and mailed back their completed census questionnaire.

A team leading a workshop at Data for Good Exchange 2018, to be held Sunday, September 16, 2018, at Bloomberg’s Global Headquarters in New York City, believes that data scientists, working together with journalists, can help address both of these problems.

“This workshop will expose people to some of the challenges and opportunities around the census, and give people a way to connect with the U.S. Census Bureau’s Census Open Innovation Labs over the next year,” says Justin Hendrix, one of the workshop’s facilitators and executive director of NYC Media Lab.

Another workshop facilitator, Mark Hansen, a professor at Columbia Journalism School, where he also serves as the director of the David and Helen Gurley Brown Institute for Media Innovation, is casting a wide net in looking for individuals who can help with these problems. That makes the Data for Good Exchange nearly ideal. “We need to get people thinking about it, especially the technical piece,” Hansen says. He’s looking for journalists, data scientists, and “anyone who can brainstorm on creative solutions to countering misinformation.”

Mara Abrams, managing director of the Census Open Innovation Labs, is charged with leveraging the talents of people outside of government to improve census response rates. Sometimes that requires a data science perspective, sometimes a journalistic one, and sometimes a bit of both. She notes that, if someone searches online for information about the Decennial Census, there is very little high-ranking content that is accurate, relevant, or compelling. “That’s why now is the time for external content creators and digital strategists to put their talents to work and normalize the census for the public,” she says.

A void of good information and good data “deters specific populations from participating,” notes Abrams. This includes Millennials, children under five years of age, historically marginalized groups, and renters, among others.

Hansen has already been working with newsrooms around the country to help them tell better stories about the U.S. Census. Many readers, he says, don’t realize that initiatives ranging from Congressional redistricting to the provisioning of school lunches all depend on accurate census counts. Local newsrooms often lack the institutional memory or the capacity to explain the importance of the census, or to tell powerful stories about it.

“There should be more easily digestible content for the public to understand how the data is collected and why it’s important — and we need trusted voices to carry the message forward,” says Lorena Molina-Irizarry, director of operations of the Census Open Innovation Labs.

Even when a news organization does mention the census in a story, the response to that story may be curiously unconnected to the facts. Hansen recalls one story which noted that couples will soon be able to indicate on the census questionnaire whether they are in a same-sex marriage. The social media discussion around that story focused exclusively on the idea that the government was over-reaching and asking for too much information – even though gay rights advocates had long lobbied in favor of this particular change.

“We need social media tools so conversations about the census can be more open and traceable,” says Hansen.

This leads directly to the challenge of fake news, and the fact that journalism, as a field, has yet to figure out how to respond to fake news and to comments that aren’t based in fact.

“What does an immune system to this sort of attack look like?” Hansen asks. It may be overwhelming to think of effective responses to fake news, but, by isolating the conversation around a single topic – the upcoming 2020 Census – it might be easier to tackle.

While these are long-term problems, Hansen urges attendees of this year’s Data for Good Exchange to get involved right away. When it comes to the census, he says, “By April of next year, it’ll all be over.”