When she headed the Huffington Post’s “OfftheBus” project, Amanda Michel led a nationwide team of citizen reporters as they looked into stories of the 2008 presidential election that weren’t covered by the media. Her work landed her a position as ProPublica’s editor of distributed reporting.
I wondered: How do you conduct investigative reporting, which seems to require special skills — not to mention some degree of secrecy — with a loose confederation of amateurs?
Michel (pronounced like nickel) has been working on this for a few months now, and she has some answers. In short, it’s not easy.
In a matter of weeks, her reporters have bird-dogged individual projects funded by the federal stimulus, gotten a big-picture view of how this money will be spent, and laid the groundwork to use this reporting to judge how transparent the Obama administration is. Last week, ProPublica launched a blog on which Michel’s reporters will share what they’ve learned. And at some point, the group will be able to use a social networking platform to share important details with each other.
Michel told me over e-mail how she is building her team of reporters, how they are approaching the daunting task of reporting on the stimulus and how this type of reporting fits into ProPublica’s investigative work. Here’s an edited version of our exchange.
Steve Myers: What’s your strategy for building a corps of citizen journalists at ProPublica?
Amanda Michel: We have three strategies for building a corps of pro-am journalists and integrating them into ProPublica’s mission. Everything will be done under the auspices of the ProPublica Reporting Network, our pro-am journalism initiative, which we launched about a month ago.
First, we’ll coordinate collaborative reporting projects. Like major metro dailies that often publish stories drawing from reporting by a half-dozen or so reporters and stringers around the country, we’ll manage assignments that draw on the insight and experiences of many people. I call them assignments because we’re not inadvertently scooping up old news — we’re intentionally organizing people to report.
Second, we’re going to make available data and documents hidden from public view and hold them up to public review. This technique was, I believe, pioneered by Talking Points Memo and is quickly becoming an industry staple. Just two weeks ago The Guardian in the United Kingdom asked its readers to go through almost half a million MP [Members of Parliament] expense reports.
Here at ProPublica we published the financial disclosures and ethics waivers filed by the Obama administration that we obtained through the Freedom of Information Act. Even in a newsroom as young as ProPublica’s, reporters are sitting atop massive stacks of documents. We won’t just cherry-pick from documents sitting around the office — we also plan to ask our readers what documents they’d like us to FOIA on their behalf.
Third, we’ll create resources that people can use to critically assess what’s happening in their towns and cities. Take a look at our recent post detailing ways to perform a background check on a company. Our suggestions aren’t exhaustive, but they are tools available to anyone who wants or needs them. Most of the time these resources will accompany an assignment, but they’ll live in perpetuity online, too.
Those are our big vision strategies. However, my role in the process is more personal and hands-on, especially during the beta phase. We’re not throwing assignments up online and waiting for something to happen. Two weeks ago I hosted a conference call for a handful of our most involved members. We brainstormed our strategy for reporting the stimulus and came up with ideas for next assignments and projects.
I’m basically knee-deep in stimulus data. Getting into the thick of things shows me firsthand what to troubleshoot and develop before we scale up to the next stage. When we get bigger, my colleagues won’t be telling folks that I’m off hiking the Appalachian Trail. I’ll still work alongside people, but numbers will change my role dramatically. We handled the beta phase for OffTheBus much the same.
How many people have signed up for the ProPublica Reporting Network? Who are they?
Michel: Our membership just soared past 1,000. That’s sizable growth for an effort just six weeks old. To be clear, though, I’m not focused on increasing membership at the moment — that comes later. What the strong sign-up rate demonstrates, I believe, is a real appetite for watchdog journalism.
As for who the volunteers are, I am looking forward to finding out! I do not pre-screen them. In general, I’ve always gotten a better sense of people by discovering their points of view and talents as the project we’re working on develops.
But I can tell you that volunteers represent almost all 50 states (no East Coast, West Coast battles here) and are diverse in their backgrounds and interests. Among our core contributors — let’s say the top 200 — you’ll find plenty of accountants and managers, policy wonks, some construction workers and IT professionals. So, analytical backgrounds predominate. They’re puzzlers, not performance artists.
What do you look for in a citizen journalist?
Michel: There is no “litmus test” for who can be a citizen journalist. I don’t operate with a Platonic ideal in mind. However, the Web enables self-selection. Many get involved because they want to improve media and they’re particularly interested in, say, the environment. When I promote projects I try to clearly communicate what skills people should have and how much time is required of them. This is a self-selecting process, and I can recruit a better network by acknowledging this.
Will citizen reporters be compensated in any way?
Michel: We’re not providing monetary compensation at this time — we’re running the program much like CNN’s iReport, HuffPost’s Eyes & Ears, the Talking Points Memo community, and others.
What tools are you using to manage this network? How do you communicate with them?
Michel: Right now we’re working with a sign-up form, e-mail, phone, AOL Instant Messenger and Skype. Recently I held several conference calls with members to discuss our strategy. Last week we launched the ProPublica Reporting Network blog. Over the next two months we’ll install more tools, including a social networking site where our members can easily publish and connect with each other. There are so many atomic bits of data associated with the stimulus that we need to invest in technology. I tend to adopt tools once I understand how people relate to information.
Your focus right now is the massive federal stimulus package. Tell me how you are using your citizen journalists to report on the stimulus.
Michel: We started off simply with a project called “Adopt a Stimulus Project.” We’re coordinating this project with public radio station WNYC and “The Takeaway,” its morning news show. Sure, it sounds “warm and cuddly” — as Sunlight Foundation Executive Director Ellen Miller jokingly tweeted — but that’s really not the case. We are asking people to monitor a local bridge or road reconstruction effort funded by the stimulus. Volunteers file on-the-scene reports. They’re charged with identifying subcontractors and helping us perform background checks on companies receiving stimulus contracts. This is slow going, as many construction projects have not yet started.
The Adopt a Stimulus Project is our gateway initiative. Several dozen people are highly dedicated to the project, and together we’ve conducted a handful of “special assignments” over the last few weeks. We have not found a story per se, but together we’ve learned a lot about the states’ varying strategies and successfully tested several operating assumptions about the stimulus.
For example, it’s been widely reported that bids for construction projects funded by the stimulus have come in below estimates, sometimes by a margin of 40 percent. This has given Vice President Joe Biden grounds for telling reporters that the government will finance more construction projects than expected.
We wanted to know if we could trust government estimates. From our volunteers we quickly learned that states were dusting off projects they’d shelved a few years back — projects they had spec’d out in, say, 2006, but planned to complete in 2012. We wanted to make sure project estimates had been re-evaluated in the last few months before projects were put out to bid.
We worked with our small team to call almost all 50 state departments of transportation, specifically asking whether estimates for their individual projects had been recently recalibrated. (The American Recovery and Reinvestment Act requires them to do this.) Almost all states checked out. You’ll be hearing about the lone state that didn’t very shortly.
States are required to obligate 50 percent of their funding for transportation projects by June 30. Once they do we are going to shift gears. I can’t say more about that now, but we’ll shift away from partnering people up with individual projects to investigating core themes.
What have you learned about the stimulus so far? How about using this type of reporting for the story?
Michel: From our volunteers we have learned a lot about how individual state departments of transportation work — which ones are better organized, prepared and accessible. State employees seem to be quite forthcoming with our members. We learned the states’ varying strategies for spending stimulus funds and what guarantees they were willing to make to their constituents. So, we’re learning just as much about citizens’ access to government information and documents as we are about the stimulus itself.
This, the kinds of information we initially have access to, is shaping our project’s direction, pushing us to focus on “accountability” stories in addition to deep-dives. The questions motivating our assignments — Is our money being spent well? Has the government put proper checks and balances in place to monitor the stimulus? Is it working? — are shared by America’s millions of taxpayers.
In the future, we hope to make sense of the Obama administration’s promise to be the most transparent administration in history. Their promise to be transparent set even higher expectations for the stimulus, telling people that they’ll be able to track every dollar spent through Recovery.gov.
One can grade transparency by evaluating technological prowess, for example giving the administration an A if it uses RSS feeds or widgets. I’d prefer to use a human test — if a reasonably well-informed person who has a few hours on her hands can find answers to basic questions about the stimulus, then the government has succeeded in being transparent. Many of our members never called their department of transportation before joining the ProPublica Reporting Network. Participating in our assignments is showing them how they can get answers to their questions. Their experiences in following the stimulus at the local level should be lead indicators for a few things, like how closely the government is monitoring projects and spending.
By proxy we’re amassing significant data and anecdotes so that we can evaluate the administration’s transparency efforts from the ground up. Karen Frillman, WNYC’s editor-at-large, is especially interested in this and we’ll be working closely with her.
How is this type of reporting different from the reporting you shepherded at “OffTheBus”? Are there any challenges unique to investigative reporting?
Michel: Investigative reporting is far more challenging. Instead of analyzing existing information you ask whether “the situation presented to us is the reality” — that’s how it’s described on Wikipedia. “OffTheBus”‘ editorial strategy focused on countering the horse-race narrative. People could more easily cover the presidential election’s impact on their communities than we can dig up corruption. At “OffTheBus” about 20 percent of our small-scale distributed projects failed to turn up a story. I suspect the ratio will be higher here.
To address this difference, we’re figuring out how to produce meaningful content from our investigations even when we don’t crack corruption. Blogging out what former Defense Secretary Donald Rumsfeld called the “known unknowns” is an obvious place to start, which we’re about to start. Our mission here at ProPublica is to invoke change through our work. Being open about what we can’t figure out is part of that process, as well as communicating to the public what the administration refuses to answer. I feel very lucky to be working at ProPublica — not just because I can learn from such great journalists, but also because ProPublica’s principles insist on a genuine engagement with the public.
There are other differences, too:
- Investigative journalism frequently deals with more sensitive material, so we have to make sure we don’t “let the cat out of the bag” and check such information before making it public.
- Investigative stories frequently take longer. That means I’ll need to develop techniques for managing a network through a sometimes awkward pairing of projects.
- And I dealt with this issue at “OfftheBus,” but it is magnified here: Frequently I didn’t tell people what we were examining because I didn’t want to bias their findings. Sometimes you can easily find what appears to be an example of something, but it alone is not proof that a trend exists. I will deal with this more frequently at ProPublica.
On the flip side, the stories we deal with have real impact on people. My colleague Abrahm Lustgarten has reported relentlessly on “hydrofracking” and T. Christian Miller reported on injured war contractors fighting AIG for medical benefits. Their stories have been the foundation for debate on the Hill and have raised consciousness of these issues. So, while networked investigative reporting can be more challenging, I’m finding that people want to participate because they genuinely care about holding those in power accountable.
What’s your plan for presenting what your citizen journalists learn?
Michel: Starting this week I will blog out a lot of our work, quoting and featuring our members’ reports. You’ll also hear a lot directly from our members. Several years ago when I was working at the Berkman Center, Craigslist founder Craig Newmark described his role as “customer service representative.” I found him — and his intuitions about managing a public good — really inspiring. Since then I’ve consciously positioned myself in the same role. These last few weeks I’ve indirectly put many of my previous ideas about crowdsourcing the stimulus to a practical test. Some we will develop. I plan to write about what we learned over the next few weeks.
The next phase will be faster-paced and will focus on collaboratively produced features. Once our social networking site launches, our members can publish their reports and analysis within a ProPublica Reporting Network site. Deep-dives will be treated no differently than ProPublica’s other stories — we’ll offer them to partner organizations for publication. Of course, our members will be credited for their work. Since all of our work is licensed through Creative Commons, we’ll also encourage our members to submit pieces they helped produce to their local newspapers.
How does this type of reporting fit in with ProPublica’s work? Have any of the journalists there used crowdsourcing in their stories, or have they used leads or reporting from your citizen journalists?
Michel: This type of reporting extends ProPublica’s reach. Other journalists here have used crowdsourcing in their stories, some for quite some time. One colleague — Paul Kiel, who helped break the U.S. attorneys firing scandal while he worked at Talking Points Memo — has employed crowdsourcing since he began reporting. Alexandra Andrews and Ben Protess recently recruited ProPublica readers who participated in the Obama administration’s loan modification program. Right now I’m focused almost exclusively on the stimulus. We have pursued several leads from the network so far. But keep in mind that ProPublica’s Reporting Network just celebrated its one-month anniversary!
What’s your long-term vision of the role citizen journalists will play at ProPublica?
Michel: Long-term, we envision our citizen journalists using the skills fine-tuned by our reporting projects to hold the powerful accountable in their towns and states.
Does a news organization need to have a person in a position like yours to cultivate citizen journalists, or can they rely on their reporters to put out calls for help on projects?
Michel: Good question. When I worked on the Kerry-Edwards Internet team as an online organizer, we joked that our positions would be extinct by 2008 or 2012. We were wrong. But I suspect we’ll largely be proven right in half a generation or so.
Organizations that enable reporters to independently solicit readers’ views or to call out for collaboration can do good work. Absolutely. But they’ll fail to optimize the network’s potential unless they have someone thinking about collaboration globally across the site and within their newsroom. I obsessively analyze metrics so I can redesign projects to be more effective and thank those who’ve contributed a lot of their time. After working with a network for several months I can match people up with projects of their interest. Given the sensitive nature of ProPublica’s reporting, we think having someone overseeing systems to promote accountability and accuracy is important.
I’ve also got a different skill set. I’m used to managing a lot of people and processes all at once. Until more reporters are taught these skills, this work is better done by someone who’s trained in online organizing. Technology is essential to managing large-scale reporting. But the reporting process isn’t always linear. There’s serendipity. There’s hunches. Automating the crowdsourcing of large data sets, like The Guardian did, would make Henry Ford proud. But there’s plenty of reporting out there that can’t be so easily dropped into Excel. Sometimes things are gonna get messy.
