POWERFUL SOFTWARE 'LIKE A TIPSTER' FOR JOURNALISTS

Date: Saturday, June 4, 2005

Since I have such a good time going to conventions, it's always a treat when one I'd like to attend happens to be in Denver. This weekend's treat is being provided by Investigative Reporters and Editors Inc., started in 1975 and based at the School of Journalism at the University of Missouri at Columbia.

IRE is dedicated to providing journalists the tools and training they need to tackle the big-picture stories, with special emphasis on keeping up with technology -- computer skills, spreadsheets, database analysis and the like.

Don't laugh. The people who started a career in journalism in 1980, say, probably never touched a computer in college, and they're now only halfway though their working lives. The tools readily available to anyone today enable a kind of reporting that no one could have done until a few years ago. An Excel spreadsheet of campaign contributors or city contracts is useful in ways that the same information on paper is not. Who gave money to the George Bush or John Kerry campaigns, and who else did those people give money to? The answers are mostly predictable, but once in a while you look at the results and think, hmmm, that's odd . . . and anomalies often turn into news.

That's why Thursday morning I picked the geekiest session on the program, on social network analysis -- that is, who in a social network is directly connected to whom, and who has power, control or influence as a result.

This is not new. Brant Houston, executive director of IRE and one of the panelists, said sociologists and anthropologists, not to mention intelligence agencies, have been using it for decades. But it's only now that the tools for doing it are sufficiently accessible that nonspecialists like journalists can use them. If you're up for some serious browsing, go to www.ire.org/sna for links to books, software and news stories and journal articles using these tools.

Panelist Paul Walmsley, a computer consultant from Boulder who has worked extensively with IRE, describes social network analysis as a cross between graph theory and mathematical sociology - a daunting prospect for most journalists. But he also pointed out, quite correctly, that the software is a tool, and you don't have to know how a tool works to use it effectively.

Plug your data into the software (even better, import it from an existing database rather than doing it by hand) and the visual display helps you understand what is going on. "It's like a tipster," Walmsley said. "It tells you where to look."

Panelist Jaimi Dowdell, now at the St. Louis Post-Dispatch, illustrated with a graph of interlocking corporate directorships for 13 big companies in Chicago. Turns out McDonald's is by far the best-connected of them (I'm trying to avoid using any of the technical vocabulary here because then I'd have to explain it). That's not implausible, once you know it, but you might not have predicted, either, that McDonald's would be more significant in the corporate pecking order than Boeing.

Dowdell, as a graduate student at Missouri, studied the network connections of IARA, an Islamic charity based in Columbia (the name, in various permutations, stands for Islamic American Relief Agency). With her colleague Aaron Kessler, she documented ties among IARA, Osama bin Laden, al-Qaida and other terrorist organizations, including Hamas.

On Oct. 13, the U.S. Treasury Department announced it was freezing the charity's assets and released a four-page fact sheet, summarizing the IARA's activities, to justify its action. IARA spokesmen dispute the allegations.

In followup stories, Dowdell, Kessler and a Kansas City Star reporter named Mark Morris laid out the details of the connections as they'd been revealed by the social network analysis analysis. "Where did you guys get all this?" U.S. officials asked them. But it was all public; you just had to know where to look.

The software "can clear out the noise," Houston said.

If you want to see what these graphs look like, Houston and Dowdell showed some of them from a paper by Valdis Krebs on the Sept. 11 hijackers, "Uncloaking Terrorist Networks" (link is on the IRE Web site). Krebs explains what it means to say Mohamed Atta was the key figure, and how the network traded off secrecy for efficiency as the climactic day approached.

On a lighter note, you can check out "Chains of affection," a study of romantic and sexual liaisons among students in an (unidentified) Midwestern high school as they evolved over a period of 18 months.

It's presented as a "flip book," a series of snapshots that show the network gradually developing a visible central ring of students each of whom is or has been connected to everyone else in the ring -- sexually transmitted diseases, anyone? -- plus all the kids who at some point hooked up with someone, but weren't in the central ring.

You could even call it revealing.