Crowdsourcing done right

Crowdsourced journalism showed its limits during the Boston bombing, but that doesn't mean it lacks value

Crowdsourcing — obtaining data, information, or ideas from a group of people — can quickly bring up vast quantities of information that might otherwise be unavailable. In the wake of the Boston Marathon bombings, crowdsourced mobile videos and photos added to the available footage. It was an iPhone photo that provided the clearest image of one of the suspects.

Yet Boston also showed the drawbacks to relying on crowdsourced information without verification: innocent men were falsely identified as suspects in the days after the bombing.

But crowdsourcing isn’t going anywhere anytime soon and people will continue to speculate and investigate when news breaks. So it is absolutely crucial that journalists figure out when and to what ends crowdsourcing is appropriate.

Good journalistic crowdsourcing takes into consideration the validity, quality, and ownership of the data journalists are accessing. When used effectively, it is a unique way to engage audiences and gather information that paints a more comprehensive picture of what’s going on in the world.

For instance, WYNC’s data team paired up with NPR’s Radiolab to build the Cicada Tracker, a project that uses DIY soil monitors to predict the re-emergence of 17-year cicada swarms. The monitors are placed eight inches into the ground, where they track changes in soil temperature. The cicadas will start to emerge from the ground when the readings remain constant at 64°F.

“What we wanted to know was, ‘Can we distribute simple sensors and little micro controllers to a bunch of people who will participate and then share the data, report back to us?’” asked John Keefe, senior editor of data news at WNYC. “And we’re showing that you can.” There are currently 125 soil monitors on WNYC’s interactive map, and Keefe’s team has sifted through more than 600 sensor readings sent in from participants along the Eastern seaboard:

“It’s just a good prototype to show what kinds of things might be possible, that we could collect data that maybe isn’t being collected, or is being collected, but not available to us,” Keefe said.

“What if it were something more crucial, like maybe it’s noise pollution, or air pollution, or some sort of chemical in the air, or radiation even?”

The New York Civil Liberties Union believes it’s tracking crucial information and has had success with its Stop and Frisk Watch app. It launched on June 6 of last year. As of mid-February, the app had more than 30,000 downloads, and it’s been downloaded since over 30,000 times.

The app gives users a way to document their stop and frisk experiences by taking short videos on their phones which are then sent automatically to the team at NYCLU. When the filming stops, users immediately receive a brief survey that asks them for additional information on the incident, but this is optional. NYCLU’s director of communications, Jennifer Carnig, said that there ways that journalists could capitalize on this widespread mobile technology in a similar way.

“When everyone has a video camera with them at all times, the potential is limitless,” she said. “But there is clearly a downside to that, because when everybody is submitting stuff, it is hard to know in realtime what is valid; there is the potential for mistruths to be out there.”

With so much data available it is important to think about what “checks and balances,” might be used to ensure that it is presented as accurately as possible, said Carnig.

At The New York Times, crowdsourcing—be it a reader comments, photos, or additional on-the-scene information—is overseen by careful editorial judgment, said social media editor Lexi Mainland.

The advantages to this approach can be seen in the NYT’s Boston bombings coverage. Mainland worked closely with reporters and editors to create an interactive graphic, based on a single freeze frame taken from surveillance footage of the finish line, that captured the moment of the first bomb explosion from the perspective of those who were there.

Times journalists and editors first tried to identify as many people in the freeze frame as possible, said Mainland. She then put out a call on the Lede Blog asking readers for their help. Four of the 19 verified accounts on the final interactive came from Mainland’s crowdsourcing efforts.

Crowdsourcing “not only gives us a window into a story that we might not have had otherwise, but it also fosters a deeper connection between us and the people who are consuming our journalism,” said Mainland.

But just because it is easy to reach out doesn’t mean it is always the answer, said Mainland. She believes that as much consideration should be given to crowdsourcing as to any other form of reporting.

“I think now more than ever, news organizations—media organizations—should be stepping in to do what they do best, which is ferret out what’s true and what’s real and to figure out what the true story is in a mass of information,” said Mainland.

Has America ever needed a media watchdog more than now? Help us by joining CJR today.

Katie Akagi and Stephanie Linning are students at the Columbia University Graduate School of Journalism.
Akagi, a former high-school teacher, holds a B.A. in feminist studies and can be contacted at
A recent drama graduate of the University of Bristol in England, Linning can be reached at or @stephlinning.
Tags: ,