Reporting the Gun Beat Q&A

Cheryl Thompson, Investigative Reporter at the Washington Post, Mark Follman, National Affairs Editor at Mother Jones, and Jim MacMillan, Founder and Editor of, in conversation with Dart's Bruce Shapiro. Below is a lightly edited version of their conversation.

Bruce: One of the hardest things as a writer is actually getting past the readers’ reluctance to dive in – that initial resistance that says I know what I need to know, I'll take a look at this but then I don't need to read. 

Jim, in your case, you need to engage people to look at your work in a consistent way over time. Cheryl, you've got this enormous series. And Mark, you too, where there's lots of data, lots of characters – that really requires the sustained attention of folks on these issues. Strategically, how did you make these packages or projects seductive as well as interesting? How did you make me engage? Cheryl?

CT: We have to think multimedia now. For me, I try to lead with human voices. And in this story, I got right to the point and the point was numbers. We try not to use a lot of those in our stories – we usually put it in a graphic – but they were so compelling: 107 of the guns were bought legally. 

I also found videos of police officers actually being shot and killed. And there was great debate among some people in the newsroom. Should we put them up? They said we shouldn't but I thought we needed to. 

And people like interactive. We had a map of the database that I built, and it was interactive so you could click on your city or state and find out who had been killed. People get bored if you're too dull, so for me it was the human voices that I think made the story. 

Jim MacMillan: The problem, at least in Philadelphia, was more one of desensitization. Because there's an enormous volume of garbage reporting, stenography, basically reading my emails: “An 18 year old was shot at the corner of this and that so many times, rushed to the hospital in whatever condition and we're never going to check again.” So you didn't have to raise the bar a whole lot. We tried everything and there was some traction around the simple visualizations, some around the videos, some around the maps. Most of the page views – this is uncomfortable – most of the page views were around the crime scene photos. So I worry about exploitation and voyeurism and sensationalism. 

But most of the engagement, most of the social sharing, came around the solutions reporting, so when we talk about what was working, that's what was sticky on social media anyway, which was where our audience was. 

There was a consensus among a lot of the Philadelphia journalists I knew – they would say, “Those people aren't on social media. Didn't you see the study that shows 40 percent of Philadelphia homes don't have Internet connections?” And it's more obvious now. It's mobile devices. Everybody's tweeting, “RIP” or “SMH or “Some guy shot my brother, Someone shot a guy from my school.” Just just about every night. 

And so it was a lot easier to go out and engage the people that were already talking about it than trying to collectively just get everybody to our site since we weren't trying to drive page views to make money, we were trying to meet people. They were already talking, guess what, you just have to listen to them.

MF: I think, yes, multimedia and yes, voices, those things are important. For me, the way I like to think about this now is how can we make this a conversation that's going to go deeper? What are the questions that I really want to have answered, that therefore my readers want to have answered? And how can I present it in that way, down to the headline: What does this really cost us? That’s a question that I think a lot of us really have wanted the answer to for a long time. You see the same story over and over and over in the daily news or on the nightly news: Kid on the corner, taken to the hospital, cop shot, whatever. Well, how much is this happening? What's the impact of it? 

Going for that deeper context and I think framing it that way becomes so important to fight against the desensitizing factor with this subject, because if you just pay attention to the drumbeat day-in and day-out, anyone's eyes are going to glaze over, right? 

CT: Can I just say that I also – for every single story, every single project I do, whether it's guns, and it's usually guns – ask why should anyone care. Every word I write I keep in mind: why would anyone want to read this?

Audience Member: What has been official reaction to the stories, in terms of policy and also just the folks who have been affected by gun violence?

MF: Well, with this project the response has been overwhelmingly positive. It's been great. We've had members of Congress respond to it. Two senators, both senators from Connecticut and Carolyn Maloney, congresswoman from New York, they all spoke out publically about this and basically said, “Now that we have data on the economic impact, it makes it that much harder to hide from this issue. We need to do something.” Now that fits with their politics. They're political leaders who want gun reform so it’s important to think about this with that context in mind. But that was a very gratifying response. 

This is a question that has been haunting the subject for a long time and now we can begin – I think what we did is just the beginning of an answer, really. We took one year of data and used a model that economists use to analyze it, and dug deep and told stories with it. But there's so much more opportunity to do that from my perspective. 

Audience Member: Is there any body that I can essentially appeal to to obtain these documents and also can I expect them to have the history of that movement when you're talking about crossing state and international lines? 

CT: Police departments? They absolutely will. Go in with a positive attitude. You will get cooperation.

Audience Member: And who's desk is it likely to be on? Within a police department hierarchy, where is it most likely to be?

CT: A detective. An investigative unit. Criminal investigation division. Homicide. 

Audience Member: But you say it's not illegal for them to pass it [trace data reports] on to you but they don't have to. Again, there's no sort of protocol?

CT: If you have source I would tap my sources first. You could trying FOILing it and see. FOIL like a case file on a homicide. FOIL the whole case file, it may be in there. 

Audience Member: That's going to depend on the state. 

CT: Yeah, it will. Where are you?

Audience Member: Indiana. 

CT: Oh, good luck with that. [laughter]

MF: Just one quick comment. I think you're expressing – there's a lot of assumption about an adversarial dynamic between reporters and police departments. And one strategy we've used covering some of the policing stories lately is to approach PIOs at departments and say to them, “Look, we're going to cover this story and we want to help the public understand it better. Help us do that.” Appeal to them that way, that you're interest and desire here is explanatory. 

CT: Also the PIO in my video, Indiana State Trooper Brian Olehy, who was explaining where the bullet went in, he's very good. All the photos and everything – he's very helpful. So I agree with Mark that I would just explain what you're doing and that it's sort of in their best interest. 

MF: And it's a negotiation, right, you've got to go back and forth some and say, “I understand your position but let me tell you a little bit more about what we're thinking” and kind of build that rapport. It can go a long way. Don't just assume they're going to shut you out.

CT: Right, no, don't ever assume they're going to shut you out. You don’t need 20 trace data reports, just 19. The art of negotiating. 

Audience Member: For those of us who work in small markets, sometimes it's harder to pitch editors on a really long project. So in the meantime what are some tips for maybe doing some really eye-opening work if you only have a lead of two or three weeks?

CT: Well, that's sort of the norm now, one, two or three weeks. But try to just think about a daily story. Here's how I got like some quick hits. I would think of a daily story I did and just expand it a little bit. 

BS: Also think about this. You can think about having a secret project. And the secret project happens to appear one byte at a time and you can go the whole project and do it without ever telling your editor you're on a project, or you can go three, four stories in and then say this is so cool, wouldn't it be great if…? So you can ratchet up your editor's interest even if she or he doesn't know at the outset that they're interested in this. 

MF: A model that we really, really favor now at Mother Jones is iterative coverage of a subject. So you have a big question about healthcare, nonviolence, whatever it is, start with a little story, then do another story, then do another one. But as you're doing them, you're thinking about them in the context of your bigger ideas, you're also kind of refining that idea as you go because you're discovering new things. This gun cost project was in many ways a cumulative result of three years of reporting. I had the question at the beginning but how we ended up doing it and the people involved in it were very much a function of iterative coverage over a long time. 

Audience Member: So kind of piggybacking off that, how would you suggest that we add more like reporting on those things that [unintelligible 1:11:56] say there's a shooting and there's not a lot of information [coughing] police department because they don't want jeopardize their investigation. How can we make that first story, or like those couple of follow up stories more impactful about being able to spend like a few months on a big investigation?

MF: Collect the data as you go. Keep a good spreadsheet with as much categorized information as you can, try to anticipate the questions you'll ask about what I have ten of these cases? And that will serve you really well. I learned that the first time I did a big project like this with mass shootings. We didn't track all of our source material, we had to go back and find all the sources we used. So just keep a running tally. Build a database. 

CT: You'll start to see patterns with the database.

JM: One of the resources I listed but didn't talk about is the Solutions Journalism Network. They've got a workbook on how to write a solutions-oriented story. Look at what's worked before, what's worked in other disciplines, what's worked in other places, and you'll get some traction around that. 

Audience Member: So this is for everybody, but specifically Mark, I mean fantastic work on all three of your parts. You're an inspiration. So when I do a story, I need the data. I want to have that foundation so that when I get attacked by whichever side of the political aisle doesn't like my story, I can at least fall back on that. 

Where did you get the data from? And I know Ted worked with you. How do all three of you, and I think Cheryl did this really well, keep it neutral so that you can say, well, there isn't an agenda on who gave me the data or analyzed the data or helped me compile the data. Talk a little bit about that and just about foundational data – is there a bias in the data?

MF: Well, I have two things to say about that. One is test your own assumptions vigorously. Go after the data that you think is valuable, examine that data, question that data, call bullshit on that data when it's needed, and don’t come out with something until you're confident that it is solid. The other thing I would say goes back to what I was saying before about process. Transparency is so valuable. Tell your audience, tell the public what you're doing. Explain who you work with, explain how you did it, and give them the data. Every single big project I've done on this subject in the last three years, we've published the data open source, except for gun costs. We haven't done it yet and I'm hoping we'll get it to it. This was such a big project we didn't have time before we had to put it out to actually build the open source version of the data. And there are some questions about how useful it might be. It's complex in a way that some of the other projects we've done aren't so much.

But like the first one I did where I'm describing this process with mass shootings – we collected the data from news reports. There are a lot of people doing this now. I don't want to say we invented that, but not a lot of people have done this on the guns beat and I was asked, I went on "All Things Considered" to talk about this when we put it out. And Melissa Block asked me, “You used a bunch of local news reports to build this database. Is that reliable?” I just explained how we studied the information; when we needed to go and corroborate it with local police departments, we did that. 

But also acknowledge what that data is, specifically where it came from, what the limitations of it are. Be prepared to explain all that, describe it and present it in a way that people can grasp. If you do that you're in good shape.

CT: Yeah, I think the key, with the gun series the data came from the FBI, it came from, I mean it very cut and dry, you know, really wasn't open for interpretation, but…

MF: Although sometimes their data's wrong too, right?

CT: Well, that's why I cross-referenced it. But I write stories questioning data. You know, with the homicide rate, I took a hit from the police chief a couple of a years ago because she was touting this 94 percent homicide closure rate. Well, when I analyzed it and broke it down, it was really more like a 60 or 58 percent closure rate. Oh yeah, she came marching in the day after that story with six of her honchos carrying guns, by the way, I might add. Legally carrying guns. 

It's really good to do a story questioning the data, because you can't just take data and go, “Oh yeah, 94 percent closure rate.” I mean every cop in this country, every police chief was like, “That's not happening.” You know, there's no way there's a 94 percent closure right. 

MF: Frame what's missing too. That comes up a lot in this subject. I mean I did that in this story too. I explained, here are the gaps in this data, here's what we don't know from it. Here's what we're only guessing at. You need to do that too. 

BS: A very helpful intellectual process for this, adopted from the Dart Center's research director and clinical psychologist Elana Newman – When she gives presentations about trauma she very often says 1) here's what we know, 2) here's what we think we know but is in dispute, 3) and here's where there's a total gap and there needs to be primary research. And if you are transparent in that way with viewers or readers, you're covering yourself very well.