It Ain't Necessarily So by David Murray, Joel Schwartz, and S. Robert Lichter
It Ain’t Necessarily So: How Media Make and Unmake the Scientific Picture of Reality
by David Murray, Joel Schwartz, and S. Robert Lichter
Rowman & Littlefield. 249 pp. $24.95
Despite its clunky subtitle, this is a readable and even an entertaining book. And despite its breezy title, it is a work that deserves to be taken seriously. Its three authors have set out to explain why the media persistently get things wrong, especially when reporters and editors are dealing with science-related policy issues like global warming, the health dangers allegedly presented by coffee or hamburgers, the interpretation of opinion surveys, or the incidence of hunger, rape, domestic violence, and child abuse. Their emphasis is on the print media, and heavy lumps are taken along the way by the New York Times, the Los Angeles Times, the Washington Post, and the weekly newsmagazines.
That something is seriously wrong with coverage of scientific/ technical matters is not exactly controversial. Many of us recurrently sense that the accuracy and reliability of media reports seem inversely proportional to our own personal knowledge of any particular subject. I remember this hoary problem being called the “curse of journalism” by Hedley Donovan decades ago, when I worked for Time and he was editor-in-chief. It is inevitably most severe when complex research data are being addressed.
One index of the problem is the way story lines keep changing. One year it is presented as a given that children living near electric-power lines are at an increased risk of contracting leukemia; then the “given” turns out to be false. Or it is presented as a given that 25 percent of all women have been raped; later on, this too is exposed as nonsense. Wobbles like these can reflect genuine advances in knowledge, but on the evidence assembled here, they are more often the result of shortcomings in reporters and editors.
The authors of It Ain’t Necessarily So are all Washington-based social scientists with academic backgrounds. Murray teaches at Georgetown and is the director of the Statistical Assessment Service, whose newsletter regularly points up journalistic mangling of research data. Schwartz is a political scientist, earlier based at Michigan and Virginia and now at the Hudson Institute. Lichter is president of the Center for Media and Public Affairs and the co-author of ten books, most of them critical examinations of the media at work. Still often cited by grouchy conservatives, including me, is Lichter’s finding, back in 1982, that most journalism students at Columbia rated Fidel Castro a better leader than Ronald Reagan.
The authors’ plan of attack in It Ain’t Necessarily So is to reveal “the inner workings” of the media, zeroing in on biases and institutional arrangements that recurrently lead to foul-ups, and offering detailed case studies as illustrations. They begin with a chapter tackling the interesting problem of unreported news. Thus, in the midst of the AIDS epidemic, the New York Times simply ignored an April 1996 finding of the Centers for Disease Control that new diagnoses of the disease had significantly declined. Similarly unnoticed by almost all news organizations was a Justice Department report, issued at about the same time, showing sudden and sharp declines in sexual assaults.
Why do such things go unreported? The authors have several answers, one of them being the built-in media bias in favor of “bad news,” which is viewed as more exciting and therefore more salable to “news consumers” (a term, sprinkled throughout this book, that I could do without). This bias is often fortified by the left-of-center predispositions of media people in general, for whom bad news constitutes further evidence of social injustice and of the need for government intervention to remedy it.
The authors’ analysis at this point is not exactly original; indeed, their telling of it is buttressed by citations from a 1984 book by Ben Wattenberg, The Good News Is the Bad News Is Wrong. But they later delve more deeply into the issue, showing that scientists themselves increasingly tend to prefer bad news to good. In a curious illustration of a “feedback” mechanism, more and more researchers, having figured out what brings publicity, strain to present their findings as evidence of some new and growing threat to society.
Still, the lion’s share of the failures elaborated in It Ain’t Necessarily So are unquestionably traceable to enduring patterns of journalistic behavior. These patterns include a lazy-minded preference for news that seems to conform to some widely accepted “template,” i.e., a familiar, ongoing story to which the reporter feels he is bringing new material. A striking example is the avalanche of publicity given in 1996 to a short, understated article in Nature about the apparent northward migration of a particular kind of butterfly in the American West. What grabbed the reporters was the sense that they were confirming the reality of global warming. It turned out, however, that temperatures had not risen at all in the areas where the butterflies were migrating.
Another such pattern is mindless indifference to the definitions underlying claims about survey data and other studies. A memorable case in point is Dan Rather’s lead-off bombshell on an evening newscast in 1991: “One out of eight American children under the age of 12 is going hungry tonight.” What Rather did not say, and presumably did not know, was that an outfit called the Food Research and Action Center had established the presence of child hunger by asking a sample of Americans whether, in the past year, there had “ever” been a “food shortage problem” in their household. By that definition, I’ve been suffering from hunger myself.
The authors of It Ain’t Necessarily So offer many more reasons why the press gets things wrong, and I found myself nodding in recognition of just about all of them. But at some point, I also began to wonder whether these multifold bad habits might not be more parsimoniously subsumed under just two categories: (1) rampant innumeracy and (2) Left-liberal bias.
It seems odd that the authors nowhere focus on the sheer ineptitude of many journalists in handling numbers. Based on long years of association, I would venture that most typically start out aspiring to be not mathematicians or accountants but writers. Often possessing overdeveloped verbal skills (many of them come out of college bearing degrees in “communications”), they are nevertheless totally unclued about correlation coefficients or standard deviations or basic laws of probability. So they are hopelessly mismatched when assigned to explain—or even to summarize—an epidemiological study centered on the degree of association among different variables. Or about the levels of risk built into different kinds of behavior. Or the likelihood that this or that will happen. Inevitably, they botch it.
My own favorite error of this kind (not mentioned by the authors) is always guaranteed to surface during election campaigns. When polling results put the candidates only three percentage points apart, and the margin of error is said to be four percent, the media will inform you that the result is a statistical dead heat. In fact, the candidate who is ahead by three points is still favored to win. The only inference to be drawn when the lead falls below the margin of error is that the “confidence level” associated with the forecast has been reduced.
The treatment of political bias in It Ain’t Necessarily So is odder still. Although (as I have indicated) the authors do cite Wattenberg on the media’s liberal propensities, they do not lean on this theme, and nowhere squarely identify politics as a prime cause of the phenomenon they have set out to explain. This seems increasingly peculiar as one turns the pages of their book, since the great majority of the distortions they bring on stage clearly have had the effect of tilting stories leftward.
Possibly the authors underplay liberal bias because they are themselves identifiable as conservatives and are wary of being dismissed for partisanship. Possibly they felt that liberal bias would look to readers like a stale issue. But at least one reader sees it as a missing issue. Especially when he thinks about those Columbia journalism students now on the loose out there.