AGCI Session II: Characterizing and Communicating Scientific Uncertainty
Session Chairs: Dr. Richard H. Moss and Dr. Stephen H. Schneider
July 31 to August 8, 1996
What's a Journalist to Do? Challenges and Approaches to Reporting Scientific Assessment
Sharon Dunwoody
University of Wisconsin - Madison
Madison, Wisconsin
In considering how the media deal with issues of scientific uncertainty, and science in general, it is important to consider the difficulties that afflict journalists as they attempt to convey scientific issues to the public. Some of those difficulties are a product of the individual reporters themselves, of their training, and their personal opinions. Others are built into the occupation, with its deadlines and focus on events over process. A third set of difficulties comes with members of the audience, who tend to transform information to fit their existing notions. Some of the roadblocks facing journalists are the very same issues that face all people when confronted with evidentiary issues.
Journalists bring a widely inconsistent set of skills to the task of covering global warming and other scientific issues. Their profession typically values behaviors that do "end runs" around validity issues. And there is some evidence that audiences value "fair play" more than reporters' judgments about what is right or wrong. In fact, readers or viewers can react angrily to journalists who try to tell them what the "truth" is.
What can be done to help journalists develop the analytical skills needed to deal with issues of evidence and probability? Universities are now beginning to teach basic reasoning courses to undergraduates, beginning the process of familiarizing students with statistics, probability theory and other tools for evaluating evidence. In addition, there is a slow but steady accretion of statistical and other evidentiary training within journalism schools. Working journalists are increasingly products of journalism school training (the proportion of journalists who majored in journalism or communication has risen from 34 percent in 1971 to 56 percent in 1992, according to national surveys). Thus, a concerted effort to improve evidentiary training in journalism schools should have a pronounced effect on the occupation in the future.
Media routines
corral the behavior of journalists and mediate against the full
telling of the story.
Journalists' occupational routines and norms present another set of barriers to the accurate presentation of scientific uncertainty. Media routines corral the behavior of journalists and mediate against the full telling of the story. Journalists work under tight deadlines and cover such a variety of stories that it is difficult for them to ascertain the "truth" of any issue. In this fast-paced environment, the attempt to report "truth" has been replaced by the concept of balance: making sure your story contains the variety of available perspectives on an issue. The theory behind this seems to be: if you can't know who's telling the truth, then at least provide a plethora of views. Over time, then, the truth will emerge. As a general strategy, the concept of providing multiple views has merit, Dunwoody says. Its primary failing is that it absolves journalists from validity checks. It can also be problematic when handled simplistically; "balance" shouldn't mean finding two people on opposite sides of an issue and giving them equal space in the story.
The competition between the need for explanation and the need for brevity presents additional problems for journalists. Journalistic accounts are short and getting shorter as editors decide that consumers don't easily tolerate lengthy text. In a 200- to 300-word story, explanatory detail is the first thing sacrificed. Even when journalists put details in, editors often cut them out. To make matters worse, journalists receive little formal training in how to explain things.
In discussion, the point was raised that there are also certain cultural differences between journalists in the U. S. and those in other countries. Most American journalists eschew an investigative, critical role, seeing themselves more as interpreters and transmitters, while journalists in some European countries tend to reflect critically on and comment upon scientific subjects.
Professional subgroups of journalists also exercise some control over news making. Through interacting with one another, reporters develop a kind of shared sense of who the appropriate sources are and how to "play" a particular issue in their stories. Members of the Society of Environmental Journalists, for example, have engaged in a long -running discussion of the debits and merits of taking on an advocacy role in environmental coverage. Currently, the group is urging its members to avoid patterns of source use or topic selection that might be interpreted by others as advocacy reporting.
The competition
between the need for explanation and the need for brevity presents
additional problems for journalists. Journalistic accounts are short
and getting shorter as editors decide that consumers don't easily
tolerate lengthy text.
Elite newspapers, especially The New York Times, play an important agenda-setting role as other publications "key" on them. For example, the Love Canal contamination story originally appeared in a small regional newspaper in upstate New York but only after The New York Times picked it up, months later, did it become a nationally significant story.
The role of the audience is another important factor in how science is reported. Audiences generally do not want to be told what is true. Rather, they seem to want to read information that shows respect for all sides and allows them to make their own decisions. People don't like to be preached to. It is important not to appear to be an elitist journalist telling the audience what to think. For example, Dunwoody talked about a "haunted house" story covered by several Wisconsin journalists. Town residents, asked later about which paper provided the better coverage, selected the one that treated the frightened residents of the house and the psychic who arrived to diagnose the problem as sources with similar legitimacy to that of the skeptical scientists and police. People seemed to resent the reporting that debunked the haunted house theories. People felt the journalist who had provided all views had behaved most responsibly, informing them of other perspectives and allowing them make up their own minds.
Audiences
generally do not want to be told what is true. Rather, they seem to
want to read information that shows respect for all sides and allows
them to make their own decisions.
Scientists are increasingly getting the jump on the peer review process by holding press conferences to release key results as they begin to understand the legitimizing function of the media. There is increasing evidence that having scientific research results published in elite newspapers confers benefits to the scientists involved. For example, data show that scientific papers published in peer-reviewed journals which are also featured in elite newspapers are cited more often by other researchers than those which are not reported on in newspapers.
In discussion about where reporters get their information for scientific stories, participants noted that journalists often use searches of databases such as Lexus and Nexus. These research tools point them to other newspaper stories on the subject, compounding the errors that are sometimes made in these stories, creating a feedback effect of inaccurate or inadequate news stories being relied upon as information for new news stories.
Media Coverage of Global Warming, 1985-1993
Dunwoody discussed Craig Trumbo's 1995 study (Trumbo, 1995) which examined media coverage of global warming from 1985 to 1993 in five major newspapers (The New York Times, The Washington Post, The Los Angeles Times, The Wall Street Journal, and The Christian Science Monitor), three major news magazines ( Time, Newsweek, and U. S. News and World Report), television network news shows (ABC, CBS, and NBC), and the science press (55 publications including New Scientist, Science News, the news sections of Science and Nature, BioScience, Scientific American, Discover, Sierra, and Audubon). In addition, the study includes time series data for the same period on policymaker attention (gauged by the number of entries in the Congressional Record containing mention of either "greenhouse effect" or "global warming"), public opinion polls (the number of poll questions containing either of those terms), and public concern (poll respondents indicating a very high level of concern about the issue).
In 1981, only 38
percent of respondents had heard of the greenhouse effect. By 1987,
that percentage had risen only to 40 percent, but over the next few
years it reached near saturation at 86 percent.
When these seven variables were plotted over time, they rose and fell together, "like a school of fish," rising to a coverage peak between 1989 and 1990 (coinciding with the extreme summer heat waves of 1989) and then declining steadily through 1993. Among newspapers, The New York Times reported most frequently on global warming over the eight years (195 stories), while The Wall Street Journal came in last (49 stories). Among the magazines, Newsweek had 23 stories on the subject while U. S. News and World Report had 11. ABC World News aired 34 stories compared to 25 on the CBS Evening News and 22 on NBC Nightly News. Among the science press, New Scientist led the pack with 207 stories, followed by Nature with 174; Science News had 78 stories and Science , 75.
In 1981, only 38 percent of respondents had heard of the greenhouse effect. By 1987, that percentage had risen only to 40 percent, but over the next few years it reached near saturation at 86 percent. Level of concern rose steadily between 1987 and 1989, such that, in 1989, there were many more "extremely" concerned than "moderately" concerned individuals. By 1993, though, the number of extremely concerned had fallen so there were about the same number as the moderately concerned. Still, global warming was consistently ranked as the most important environmental problem by only 5 to 9 percent of respondents throughout the period; that percentage peaked at 12 percent in November 1989, and coincided with high levels of media coverage.
Based on the study results, Dunwoody discussed who set who's agenda on this issue. The science press led both newspapers and TV, and newspapers led TV. This inter-media agenda setting was strongest early in the time series, suggesting that the mass media seek the kind of issue legitimization that coverage by others provides. The agenda -setting power of the science press diminished as time went by. At any given time, TV coverage and public concern about global warming were closely related, but none of the media variables predicted the rise and fall in public concern over time. "If public concern is being shaped by television coverage, the effect takes place in a shorter time span than that measured here," the author of the study says. The mass media and policymakers seemed to go back and forth, influencing each other, with the media's influence fading over time. Finally, and interestingly, an increase in policymakers' attention seemed to correspond to a decrease in public concern. Colleagues in the audience cautioned that the data on this last effect should be analyzed seasonally before drawing conclusions.
Did the global
warming issue peak too soon in the media? In 1988-89, media interest
peaked, long before the current "certainty peak."
Did the global warming issue peak too soon in the media? In 1988 -89, media interest peaked, long before the current "certainty peak." Now that it may be time for action on this issue in the policy arena, media coverage could be important in getting the attention of policymakers.
In closing, Dunwoody provided a few tips for scientists wishing to improve the accuracy of science stories for which they are sources of information. It is helpful to provide something in writing for the journalist, in addition to an interview. Data show that accuracy ratings go up when such written information is provided. Insist on face-to-face rather than telephone interviews, as this also improves accuracy ratings. At a scientific meeting, make time to meet with reporters and discuss your research in language they can understand. If there's an important caveat, say it in the middle of a very quotable sentence so it's less likely to be cut. Provide graphics to reporters and use familiar metaphors. Share good metaphors among scientists and think about how to convey uncertainties and other scientific issues visually.
Reference
Trumbo, C., 1995, Longitudinal Modeling of Public Issues: An Applica -tion of the Agenda-Setting Process to the Issue of Global Warming. Journalism & Mass Communication Monographs, No. 152.
Scientists are
increasingly getting the jump on the peer review process by holding
press conferences to release key results as they begin to understand
the legitimizing function of the media.