Computer-Assisted Reporting Challenges
Traditional Newsgathering Safeguards
But ethics must remain in the front seat while
journalists cruise the information superhighway
By Bob Steele and Wendell Cochran

---

Bob Steele is director of journalism ethics programs at The Poynter Institute. Wendell Cochran is assistant professor of journalism in the School of Communication at American University in Washington, D.C. They prepared this report for the ASNE Ethics Committee, and it appeared in the January 1995 edition of the ASNE Bulletin

---

"Scientist makes up data for prize-winning research."

"Politician's e-mail opened, read by opponent."

"Agency keeps secret database of private information on citizens.

If these imaginary headlines were in your paper tomorrow, you probably would say: "Great stories." But you might be more troubled if the same headlines read:

"Journalist makes up data for prize-winning story."

"Politician's e-mail opened, read by reporters."

"Newspaper keeps secret database of private information on citizens."

Luckily, few of those headlines have been written so far.

But if you think they aren't possibilities, you haven't looked lately at the way computers are changing the news-gathering process. Turning to online sources for information or using computers to analyze information has become almost as commonplace as dropping in on city hall. And, with the increased use of different newsgathering methodology, editors might face new ethical challenges, or at least, new twists on old problems.

"I would hope that we don't let technology change principle," says Ron Meador, assistant managing editor for special projects at the Minneapolis Star Tribune.

In other words, using the computer does not alter traditional concerns about accuracy, fairness, privacy, plagiarism or other ethical issues. But using the computer to gather and analyze information can put reporters and editors on unfamiliar ground. That means everyone in the newsroom should understand the nature of computer-assisted journalism techniques and be aware of the kinds of difficulties that can arise.

So far, there have been relatively few reported instances of ethical problems associated with computer-assisted journalism. Perhaps the most publicized was last winter's Olympics episode, in which a group of sports reporters snooped around in skater Tonya Harding's electronic mail account.

There have been a few cases in which journalists have used quotes taken from an electronic message board only to discover that the person quoted did not write the message.

Others have taken raw data from government files and learned, after publication, that the information was inaccurate.

One of the most common forms of technology misuse has been in the arena of photojournalism. Several publications have been caught in computer photo manipulations, including the infamous Time magazine cover art of O. J. Simpson.

There are potential legal problems as well. At least one journalist has been sued, unsuccessfully, for libel, based on his supposed misuse of materials he took from the Internet. But it's probably just a matter of time before something really serious happens to call into question the integrity of our business and give the public yet another reason to doubt our credibility.

Editors say the way to avoid problems in stories developed electronically is to follow the same kinds of procedures newspapers have always used to make certain stories are accurate and fair and to ensure that news-gathering practices can pass ethical scrutiny.

William Woo, editor of the St. Louis Post-Dispatch, says, "I don't see journalists as having a special tool kit (for computer-assisted reporting) any more than they would have a special tool kit to cover the city council or the O. J. Simpson case."

Accuracy

Accuracy, of course, remains the paramount issue for journalists, whether they're writing with quill pens or Powerbooks.

"We're on firm ground when we treat online information or bulletin board information the same way we treat any information we gather as reporters, " says Michael J. Connor, executive editor of the Syracuse (NY) Post-Standard. Information gathered electronically needs to be verified the same way any other material is checked out, he says.

Says Jeff Browne, special projects editor at the Milwaukee Journal: "The same rules apply as if information comes over the police scanner, or comes to us by way of police reports, or anybody calling us on the telephone. We check it out and make sure we're confident about the information before we publish it."

But Connor recognizes the potential for problems. "When the gates open and we have access to new approaches, there can be a temptation to stray from our basic values of fact-checking."

One fact of online life is that it's possible to disguise one's identity. Before making too much of this issue, don't forget that it's probably just as easy to be taken by a source who calls on the telephone. And it's not unheard of for people to try to fool reporters about their identity, even in face-to-face meetings. Nevertheless, reporters have been burned by quoting from messages that were sent under one name, but actually written by someone else. Tracking down the person who actually wrote the message can be tricky in some cases. But, says Connor, "We must verify authorship of what is said."

The Norfolk (Va.) Virginian-Pilot has instituted a newsroom policy that addresses this and other online reporting issues. The policy says, in part: "In quoting from electronic communications, we will make certain the communication is genuine, as it is easy to fake Internet return addresses or log on as someone else. The net is not controlled like a wire service; hoaxes can come from anywhere."

The accuracy questions relating to computer-assisted journalism go far beyond merely correct identification of sources. For example, Woo is concerned that journalists find authoritative sources, not just ones who are convenient. And he worries that reporters might selectively quote from documents or other material they find electronically.

"It's very important that we be faithful to the context, that we read entire documents in our searching so that the word or phrase or idea that we wish to develop is faithful to the author's original context," he says.

The concern for context takes Ron Meador one step further.

"To me the risk is not that we will go and cherry-pick facts to support our story and deliberately or carelessly skew reality. The risk to me is if we rely too much on cybertext rather than personal human contact and connection, and we diminish subtext, which is important to understanding the main text."

Meador says journalists should make certain that their Internet account isn't their only connection to their sources. "You get so much more information in an interview than by plain text."

In personal interviews, of course, reporters can see the person, look at his or her office, see the pictures on the wall, notice the color of the paint in the kitchen and collect scores of other clues about the person. All this helps understand the message. On the phone, one can hear laughter, pick up on sarcasm and get some idea of when a source is trying to tell you something between the lines. It's very difficult to put any of that in an e-mail message.

The Virginian Pilot's online policy deals explicitly with this problem. It states: "If we use e-mail in our reporting, we are aware that without voice inflections, it is more difficult to interpret someone's tone, so clarification may be necessary."

Who checks our data?

Computer-assisted reporting techniques complicate the usual information verification methods in other ways. Journalists who use sophisticated data analysis tools to study a problem often become more knowledgeable about a subject than any other source. And journalists can create information that didn't exist before.

For example, in the late 1980s, journalists at Gannett News Service and USA TODAY designed new ways to measure the health of financial institutions. There were no other sources for this information, no Federal Reserve Board or banking industry studies to cite. Outside sources gave advice about how to construct the formulas, but in the end, it was journalists who wrote the formulas, decided which variables to include, analyzed the results and drew the conclusions.

To Mike Connor and other editors, that approach waves red flags. "Nobody else has access to test our data and to challenge us. We must be willing to explain our methodology and how we arrived at our facts," he warns.

Connor also is concerned about the "aura of authoritativeness" imparted by computerized data.

That is to say, in a computer, everything looks equally correct, though almost everyone with a credit card knows how wrong computers can be at times. In general, data is dirty; it contains typos and outright mistakes that can throw off analytical results. That's true whether the analysis is being done by a journalist or someone else. Many stories journalists publish that rely on government records quite likely are grounded, at least to a certain extent, on databases that contain mistakes.

The Federal Election Commission individual contribution records are full of typographical errors, as just one example. This should not be surprising. After all, a large database might contain several million characters; the simple laws of probability tell you there are likely to be some errors.

Among the things you might do to recognize these inherent problems: Consider rounding off most numbers; create groupings that emphasize similarities, rather than differences, between individual cases; make certain that ranked lists don't exaggerate small differences.

University of North Carolina journalism professor Phil Meyer, who is the father of computer-assisted journalism, says there are three rules for avoiding trouble with statistics: Explain what you've done; apply your formulas consistently, and don't claim too much for the results.

Privacy

Journalists who use computers also soon confront privacy issues. For example, what should reporters do if they get access to hundreds of thousands of Social Security numbers?

Privacy concerns have led some government agencies to deny news organizations access to records, including those dealing with student loan defaults. In the 1994 crime bill, Congress made driver's license records off limits to the press and the public. States can set up special rules permitting access to driving records, but an important source of computer-assisted stories now may be much harder to get.

"The ability of journalists to tap into databases and rummage through the lives and records of citizens is certainly enhanced by the ability to go into the computer," says William Woo. He believes that "a great wrong would take place if reporters went on fishing expeditions into the lives of people with no greater purpose than to see what we can find about them."

Woo says news organizations should "re-emphasize and specialize their privacy guidelines to deal with the enormous capacity that online database searching permits."

Ron Meador doesn't discount concerns for the intrusion factor in computer-assisted reporting, but he looks at them differently. He fears journalists might inhibit their newsgathering ability if they worry too much about privacy.

"Privacy isn't a matter of what we dig up. It's a matter of what we print. I'm not disturbed by the notion that computers will allow us to know things that allow us to invade privacy. I'm concerned with what we do with that knowledge," he explains.

Original reporting?

Meador thinks reporters who rely heavily on online research might start to rely too much on others for story ideas and even for the approaches they follow in writing the story.

"I think it will erode original thinking, especially with all reporters and writers under great pressure to produce. This is going to be a big shortcut--not to the text of your story, but to the point and idea of it." Another issue to consider is the possibility that using online services to look at old stories will also mean that more stories will rely on the same old sources.

Ethical safeguards

So how might journalists avoid the pitfalls of the new technology? One approach would be to lay out guidelines such as those used by the Norolk Virginian-Pilot. Another is to make certain that the newsroom culture encourages consideration of ethical decision-making protocol.

Says Woo, "You need both culture and structure, or you only have water-cooler bull sessions, which are not useful in guiding behavior."

Syracuse's Connor emphasizes training and the need to involve reporters and editors in the decision-making process. He addresses ethical issues in his weekly staff critiques. And he urges Post-Standard journalists to make computer-assisted reporting part of their everyday life.

"We want reporters to use the tools, not make it sound like dynamite that could blow up in their faces," he says. "We must make sense of that glut of information out there. We have no value if we can't explain the 'why' of information."

Meador of Milwaukee argues that avoiding computer-assisted journalism would be ethically problematic.

He says that the power of the technology "gives us enormous responsibility on the green light of journalism ethics because it gives us a greatly expanded ability to do important investigative and enterprise stories oriented to social change. To me, there is an ethic saying 'Themore and better you can do, the more and better you ought to do.'"