A Downward Trend in Use of Anonymous Sources

Tom Rosenstiel, who directs the Project for Excellence in Journalism, spoke about some findings on the use of anonymous sources that are in the 2005 State of the Media report issued by his project.

We examined 16 newspapers of varying sizes and, in that sample, seven percent of all the stories contained at least one anonymous source. That is down from 29 percent that we found in a similar sample a year ago. With front-page stories, it’s 13 percent, also down from last year. The bigger the paper, not surprisingly, the more prevalent anonymous sourcing is. If we examined just Washington stories, my hypothesis is that the number would be even higher. In looking at other media, something interesting emerges. In network television, more than half RELATED ARTICLES
“‘The Seduction of Secrecy: Toward Better Access to Government Information on the Record’”
“The White House: Can It Control the Press?”
“Reporting in an Era of Heightened Concern About Anonymous Sources”
“Offering Anonymity Too Easily to Sources”of the stories contain anonymous sources, and when we looked at packages on TV—eliminating the stories that the anchor reads, which are very brief—the number goes up to 68 percent. Is that because those stories are more investigative and include more secret material? I’m not sure that’s the cause. We’ve been told by people in network television that often they eliminate attributions simply because of the compression of time. It’s convenient to just say, “Administration sources say ….” If we went back and looked at the tapes, I suspect a lot of these references to anonymous sources would not be on stories where the information was somehow closely held. In cable, which is a more extemporaneous medium with very few news packages and most of the time is spent in live interviews, only nine percent of the stories have anonymous sources. On the Internet, where the stories are largely from print sources—very heavily wire copy—we looked just at the Web page’s four or five lead stories and 19 percent of the stories contained anonymous sources. What I think this tells us is that the format, and not always the journalism, determines how much transparency there is about sourcing.

Gene Policinski, who is executive director of the Freedom Forum First Amendment Center, described findings from a recent public opinion survey about the use of anonymous sources and also from a questionnaire that was given to journalists to determine the frequency and utility of their use of these sources.

Since 1997, we’ve taken samplings through our annual State of the First Amendment reports to look at how the public feels about the use of confidential sources. We found that public support for the right of journalists to have, as part of their toolkit, the use of the confidential source, remained very high—70 percent. That’s down from 85 percent when we began the surveys in 1997. A few months ago, working with the Investigative Reporters and Editors and the Radio- Television News Directors Association, we asked reporters to fill out a form about how they use sources. We had 711 journalists, reporters, editors and news directors who responded. Fifty-nine percent told us they’d used confidential sources in less than 10 percent of the stories they had written. When they were required to use anonymous sources on stories that would have not come to their attention in any other way, the number goes way up. Reporters told us that 60 percent of the stories they’d gotten from a confidential source would not have come to their attention had that source not had that protection. So journalists want to reserve that right. It’s interesting, however, that nationwide as opposed to Washington, the number of stories using such sources might be less. That might be hopeful for those who would like to limit that use—that the problem isn’t as big across the country as it may be in Washington. It doesn’t speak to whether government secrecy is creeping up. I also wonder whether those numbers will change upward as state and local governments model themselves more on the federal government and insist on that kind of behavior.