« « Data Delver: David Donald, Center for Public Integrity

Visual confections are more than mere presentation » »

Data Delver: Mo Tamman, Wall Street Journal

Posted by on Feb 13, 2010 in Blog, CAR, data delvers | 2 Comments

I think in the modern era, CAR merits consideration across all beats. Perhaps that’s part of why I didn’t discover it for so long, it never occurred to me to search for a specialty that looked at data, because I just assumed that was the way to do journalism. But if I were pressed to think of beats that are most naturally paired with CAR, I’d name politics first, and business second. And if you say business news, I say Wall Street Journal. Does their process differ from that of other newsrooms? And how do WSJ reporters respond to a CAR specialist? What are the organization’s priorities? I found the answers to these questions in more during an interview with WSJ News Editor Mo Tamman, who has been tasked with “beefing up the news organization’s computer-assisted reporting muscle,” as he writes on his LinkedIn profile.

This profile of Tamman is a part of my continuing series I’m calling “Data Delvers,” where I pass on summaries, quotes and audio clips from conversations with journalists using technology to find, analyze and convey data-driven stories and/or projects to the modern audience.


Using CAR to provide essential evidence

Tamman was brought on to the Journal in Sept. 2008 to more thoroughly integrate computer-assisted reporting into the organization’s workflow across beats.  Tamman reads the daily news budget, and jumps in where he thinks he can help — he said it’s a relatively informal process.

He said he receives data from many different sources.  Sometimes, it’s leaked to him, other times a reporter brings it to him, other times he requests it through the Freedom of Information Act.  The longer he’s been around, the more he finds reporters come to him. Their requests include everything from analysis to printing a spreadsheet as a PDF.

Most reporters are excited to collaborate  “Most of them are overwhelmingly happy to work with me because I will give them stuff that no one else on the planet has,” said Tamman.  “It brings an exclusivity to their content and writing that they wouldn’t have otherwise.”

Sometimes he finds that reporters and editors may come to him with preconceived notions that they wish to have supported by the data, and there can be a conflict when the data doesn’t support that theory.  But Tamman said this is nothing new, and it is an editorial conflict that has been around since Gutenberg.

For analysis, he spends much of his time in SQL, but also uses “middle-level Web programming” to obtain information from within Web sites.


Learning about the computer’s role in journalism

Tamman has held many different journalistic roles in his life, and first discovered CAR while working as a night cops reporter in Camden, New Jersey.  He was the only one who knew how to work a computer, and learned more and more about it.  His rationale was that it might keep his job secure, and maybe even get him a pay raise — something he has found to be true on both counts.

Prior to arriving at the Wall Street Journal, Tamman also did database work at both the Atlanta Journal-Constitution and the Sarasota Herald-Tribune.  He served as a reporter and Web editor at various papers as well.


Media must provide context

Above all, Tamman said he sees CAR as providing the evidence behind the story, the material for the nut graf. For him, journalism is about helping people make sense of our universe.  This is why the notion of uncontextualized data centers is “rubbish” to Tamman. He said he understands the need for page views, having served as a Web editor.  But, he said, the page views don’t last, people don’t stay on the site beyond searching that data.

Audio: Why data centers aren’t fulfilling journalism’s mission.Data centers: “Let other people do it.”

“I firmly believe that if you’re going to put up a large data set, there better be an editorial reason for doing so,” said Tamman.  He encourages posting data and pairing it with a project, or contextualizing it through a data-driven application such as Politifact.

Throughout all of the jobs he’s held, Tamman has closely valued the importance of helping the public make sense of the world.  When he compares what he’s doing now to other job he’s held, he admits he misses the writing.  “At this point it is what it is.  There will be other opportunities for me to do that in the future.  I’m not dissatisfied.”


CAR specialists are a rare breed

Over time, Tamman has been surprised that the role of a CAR specialist still exists. “When I got into this business 20 years ago, I had assumed that the skills I was working on in the newsroom would be as common as people using a telephone or word processor.  But that hasn’t come to pass.”  He said he is “flabbergasted” at how little even younger reporters know about spreadsheets.

Tamman thought this might be because journalism relies so much on people who know how to communicate, and people who have an affinity for that type of work don’t gravitate to working with structured data.

Or as he put it: “I just don’t think most reporters and writers have the facility to really fully embrace computer-assisted reporting.  There is always going to be a need for a pretty rare individual who has both the ability to understand the empirical and manipulate it, and also report and write, or at least understand the need of those who report and write, so they can both produce something that’s meaningful and communicate with the public in a more generic way.”



Extended transcript

More details in Tamman’s own words follow.



Why were you brought on to the Wall Street Journal?

Before I came here, I did everything. I wrote, I reported, I did data analysis, I did web development. You name it, I did it all. Here, my primary responsibility is to build what I call the empirical spine of stories. I’ll do some reporting, but mostly what I’m doing is the analysis which effectively becomes the nut graf of some type of narrative.



Are you working equally with all beats across the paper?

Yeah, right now I’m staring at banking data. Earlier I was staring at FDA and UNBA data. I don’t really have a beat, although banking has been something I have focused on over the last year or so, only because of the necessity



Do you enjoy it more or less than other positions you’ve held over the years?

I miss writing, but at this point it is what it is. There will be other opportunities for me to do that in the future. I’m not dissatisfied



Do you think there is a large role for CAR in the future of journalism?

If there isn’t, there is no journalism. When I got into this business 20 years ago, I had assumed that the skills I was working on in the newsroom would be as common as people using a telephone or word processor. But that hasn’t come to pass. I still have people coming up to me with a spreadsheet saying, “Can you convert this into a PDF so I can print it out?” Some are older guys, but there are also young reporters who have no idea how to do even the most rudimentary things in a spreadsheet, which I find discouraging and flabbergasting all at the same time.



Why do you think that is?

Journalism has always been a trade that relies on people’s abilities to communicate, wherever it’s verbally or via a narrative or whatever – that’s what it is at the very core. The graphics department can bellyache about it, the photo department can bellyache about it, but the truth is it’s a word-driven industry. People who have certain skills when it comes to piecing together narratives and communicating information for the written word — even if it is ultimately read out loud – very rarely have the same ability to comprehend and manage numbers. I think it’s a left brain/right brain thing. While I would have thought 20 years ago that it would’ve been a natural progression for most people, I just don’t think most reporters and writers have the facility to really fully embrace computer-assisted reporting. There is always going to be a need for a pretty rare individual who has both the ability to understand the empirical and manipulate it, and also report and write, or at least understand the need of those who report and write, so they can produce something that’s meaningful and are able to communicate with the public in a more generic way.



Do you find most reporters eager to work with you?

It depends. If a reporter is honest with himself about what the purpose of the story is, he’ll realize it’s to communicate something that has some fundamental truth to it. Most of them are overwhelmingly happy to work with me because I will give them stuff that no one else on the planet has. It brings an exclusivity to their content and writing that they wouldn’t have otherwise. Of course, many reporters and editors have preconceived notions about what’s happening in the universe. They will say, “I want to do a story about this, and this is what we expect to find. Do you have some data that we could use to support this?” Well, you do your analysis, and their information is based solely on anecdotal information, or maybe even on someone who did analysis to support a particular type of point of view they are trying to enhance. If the facts don’t support it, under those circumstances it can be a little difficult. That’s a legitimate editorial argument that reporters and editors working together have had since the Gutenberg press. It’s just the way it is.



What do you think of the emerging data centers?

It’s rubbish. It’s some MBO, some pinhead web editor or executive editor who has to push page views. It serves no useful purpose whatsoever. I used to be a Web editor, so I am familiar with some of the thinking that goes on behind it. The truth of the matter is those are worthless page views. The people who go to those pages don’t click through to anything else other than to search information. I firmly believe that if you’re going to put up a large data set, there better be an editorial reason for doing so. Maybe it’s tied to a project, but there’s got to be some context for it. Just putting crap up there is a waste of time and effort, and it’s a cheap way of somebody getting an MBO bonus without necessarily serving the greater good of the paper itself or the public as a whole.



What about when it’s posted as part of the project or used in a data driven application, such as Politifact?

Then, there’s a context for it. Most of those data centers you go to, it’s pet names and salaries of public employees that are just thrown up there. Let other people do it, they are not news organizations. We’re news organizations. The world is full of clutter and annoyance and noise and crap and confusing pieces of information. Our job is to reduce the clutter. More so today than at any other time in the history of media, our job is to reduce the clutter. There is so much noise out there, and all the data centers are doing, because they don’t have the context, is to add to the noise. That’s not what we should be doing, we should be helping people to understand the universe, not get more confused by it. What does it mean if someone made $100,000 last year? Is it unusual for a person in that position? Maybe they’re being paid less than they should be. If you don’t have context, that database is just deceptive, and it serves no useful purpose.



How do you get obtain various data sets? Are you simply assigned to various projects?

We have a pretty informal system here. I read the budgets every day. I visit the bureaus, I talk to people, I hang out at the bars. When I hear something that I think I can contribute to, I just jump in. Now I’ve been here 18 months, and a lot of people ask me, “Can you help me with this?” Believe me, there is plenty of work to be had that way. As far as how I get the data, it depends. Some people leak it to me, sometimes I FOIA it, sometimes I negotiate prior to filing a lawsuit, sometimes I create it myself. In this day and age so much of it is available for download online — there’s almost an obscenity of wealth.



What skills do you find yourself using most often?

SQL. I spent my life learning SQL. I do a fair amount of what might be considered middle-level web programming, especially for importing data and cleaning data that may have some odd structures. I do a fair amount of programming for scraping and hacking into Web sites, not black hat hacking, but hacking simply to get in. I spend an awful lot of time analyzing the data using Excel to examine what’s happening



How did you first get interested in CAR?

I was a night cops reporter in Camden, New Jersey, and I was pretty much the only person who knew how to work a PC. I quickly realized that there was an opportunity and I kind of enjoyed doing it, so I started teaching myself how to work it. I figured it was job security at worst, and perhaps could pay me a little extra at best. I think that’s proven to be the case on both counts.

« « Data Delver: David Donald, Center for Public Integrity

Visual confections are more than mere presentation » »
  • Kathleen Sullivan

    Michelle — Really enjoying your Data Delvers series! Looking forward to reading the next one.

    [Reply]

  • http://www.linkedin.com/pub/john-campbell/7/742/297 John Campbell

    “Tamman … first discovered CAR while working as a night cops reporter in Camden, New Jersey. He was the only one who knew how to work a computer”

    Only reporter, that is ;)

    Mo, you done us proud.

    [Reply]