How We Know What Isn't So (18 page)

This is not to suggest, however, that a sense of plausibility does not foster the dissemination of false claims that have more serious consequences. It does. This is seen most clearly in the field of health, where bogus claims about the effectiveness of various practices in warding off ill health—claims that have the ring of plausibility—are frequently spread through the media and through everyday social exchange. These claims are discussed in detail in Chapter 8.

Upon reading this chapter, the reader may feel in a bit of a bind. The implication of much of the earlier part of this book is that our habitual ways of evaluating evidence are subject to error, and that therefore we can be misled by the apparent lessons of everyday experience. Thus we should place a little less trust in what our personal experience tells us, and we should rely more heavily on hard evidence and objective facts—on information that comes from beyond our own personal experience. The data of our own experience are often biased and incomplete, and we cannot always be counted on to evaluate them fairly. Consequently, those who study human judgment and decision-making urge us to give less weight to our own impressions and to assign more weight to the “base rate,” or general background statistics. For instance, in contemplating the odds that our own marriages might end in divorce, we should attach less significance to our present passion and current conviction that we have found the right person, and we should pay more attention to the overall divorce rate of approximately 50%. To be sure, we should not discount our current feelings and self-knowledge altogether; we just need to temper them a bit more with our knowledge of what happens to people in general. This is the consensus opinion of all scholars in the field. Because personal experience is not an infallible guide to the truth, we must augment it (augment it more than we apparently do) with relevant background statistics.

That is all well and good, but how do we get these background statistics? How do we know they are accurate? Indeed, what does the oft-cited 50% divorce rate mean anyway? Is it that 50% of all
marriages
end in divorce (in which case the total would increase for each and every one of the divorces filed by people like the Gabor sisters, making the odds of divorce seem worse than they are), or is it that 50% of all
people
get divorced at least once (so that no single person affects the total disproportionately)? It is hard to incorporate the overall divorce rate into one’s personal assessment without knowing which it is.
*

More generally, the information presented in this chapter indicates that it can be extremely difficult to get a truly accurate estimate of the relevant base-rate. We generally do not collect the base-rate data ourselves; it must be obtained from secondhand sources. Moreover, few people have the wherewithal to look up (and decode) the relevant data in scientific journals, and so their exposure is limited to the summaries presented by various media outlets. But alas, as we have just seen, the summaries presented for mass consumption are often terribly distorted.

The portrayal of the heterosexual AIDS risk provides a good example. If you are an exclusively heterosexual, non-IV drug-using, middle-class person in the United States, how worried should you be about contracting the AIDS virus? Here are some of the things we have been told: “Research studies now project that one in five heterosexuals could be dead from AIDS at the end of the next three years. That’s by 1990. One in five. It is no longer just a gay disease. Believe me.” (Oprah Winfrey).
³¹
“By 1991, 1 in 10 Babies May Be AIDS Victims.” (USA Today).
³²
“The AIDS epidemic is the greatest threat to society, as we know it, ever faced by civilization—more serious than the plagues of past centuries.” (A member of the President’s AIDS commission).
³³
Using information like this, the apparent base-rate is frightening indeed. One in five heterosexuals dead by 1990! One in ten babies with AIDS by 1991! To justify such predictions, the virus clearly must have been spreading wildly in the heterosexual population at the times these estimates were made. The risks to heterosexuals of a single sexual episode must have been enormous at that time. Anyone who followed the advice of decision theorists and gave considerable weight to the base-rate should have sworn off sex with anyone other than a longtime, faithful spouse.

Fortunately for those who did not rein in their sexual habits, the base-rate implied by these alarmist accounts was way off the mark. It is now 1990 and nowhere near one in five heterosexuals are infected, let alone dead. It is still the case that the
overwhelming
majority of AIDS cases involve gay men, intravenous drug users, and the heterosexual partners of the latter. 1991 is close at hand, but there is not even the faintest hint of an upsurge in babies with AIDS that would justify an expectation of 1 in 10 in the next year. Anyone who looked to the media to try to establish a sensible base-rate was not well served.

In marked contrast, people’s personal experience, fallible as it may sometimes be, would have given them a much more accurate estimate of the heterosexual AIDS threat. The vast majority of the U.S. population cannot think of a single person who has contracted AIDS through heterosexual intercourse, nor can they think of someone who knows someone who has. (The one-night-stand nightmare that was described earlier does not qualify because that story does not specify whether the “victim” actually contracts AIDS). By the light of personal experience, then, the heterosexual AIDS threat seems overblown. How overblown is hard to tell just yet, and the question is exceedingly controversial. The profile of the epidemic that is shaping up, however, indicates that the true threat is much closer to that intuited by personal experience (“It cannot be that pervasive, no one I know has it”) than that implied by alarmist media accounts (“One in five heterosexuals could be dead in the next three years”).

On what, then, should one’s judgments and decisions be based? Personal experience, or what we are told are the base-rate statistics? What happens when the two conflict? Decision theorists have developed formal procedures for combining these two sources of information into one overall assessment. Applying these procedures can be problematic, however, when the base-rate information is prone to inaccuracy. When both personal experience and the relevant background statistics tell us the same message, things are easy and we can be quite confident that our understanding is correct. When they conflict, however, we should understand that our assessments are particularly prone to error. We must learn to distrust personal experience a bit when it conflicts with the base-rate; but when the base-rate comes from an uncertain secondhand source, we must also distrust it when it conflicts with personal experience. Conflict between these two important but imperfect sources of information should temper our judgments and beliefs. Sometimes this note of epistemological caution is all that can be said about integrating divergent sources of information. But even that can be helpful because distinguishing what we know well from what we only think is true is itself an important advance. Sometimes it is not “the things we don’t know that get us into trouble; it’s the things we know that just ain’t so.”

The issue of how to think about the evidence of everyday life so as to avoid erroneous beliefs will be addressed more extensively in the final chapter. For now, it may be most important to consider how one should evaluate secondhand claims reported in the media. How can we know whether to trust a given claim? How can we know whether to accept a reported base-rate? Fortunately, there are several helpful guidelines.

Consider the Source
. One of the most important things to consider is something we all recognize in theory, but sometimes overlook in practice—the need to consider the source of the message. We all seem to honor this principle when we discount what we see in the
National Enquirer
and put more stock in what we read in
The New York Times
. But who or what is the source being cited by a reputable newspaper? With respect to the coverage of AIDS, we should know that we ought to give more credence to the words of epidemiologists than sex therapists, rock stars, or actors. Epidemiologists spend their working hours trying to understand and predict the spread of infectious diseases. No one is more equipped than they to issue projections about the spread of AIDS. Sex therapists are presumably quoted on this issue because of the connection between sex and AIDS. But however much a sex therapist might be helpful in overcoming sexual dysfunction or dealing with problematic sexual feelings, they are not expert in the complicated business of predicting the course of an epidemic. Throughout all of the massive, confusing, and often alarmist coverage of AIDS, it has been the epidemiologists who, on the whole, have issued statements most consistent with the facts as we now know them.

Attaching special significance to the words of the true experts, however, is not as easy as it might seem because reporters will often distort what an expert really said. A common way of doing so is to place an innocuous quote by a credible person next to an outlandish claim, and thus make it appear that the former endorsed the latter. For example: “One source claims that one in three teenagers could be addicted to cocaine within the next five years. Says Elliot Ness, a member of the President’s commission on drug abuse, “There are no easy solutions in the drug war.’” Mr. Ness might indeed believe that there are no easy solutions to the drug war, but nonetheless want to have nothing to do with the “one in three” estimate. The proximity of the two statements in the same paragraph, however, makes it easy to think he is responsible for both. Look carefully at what is actually quoted and what is only implied.

Trust Facts, Distrust Projections
. Predicting the future is risky business, even for those experts whose job it is to do exactly that. Just think of how often meteorologists are wrong about tomorrow’s weather, or how frequently economic forecasters misread the leading economic indicators. This means that we should give more weight to statements of fact by experts, and less credence to their projections of the future. Although the projections of epidemiologists for the spread of AIDS are certainly worth noting, we should pay even closer attention to their reports of current facts, such as the number of current AIDS patients, the percentage of cases among homosexuals, heterosexuals, and IV drug users, or the rate of seropositivity in blood samples donated to blood drives. Again, the message is one of epistemological caution: Be wary of those who claim to know the future.

Be on the Lookout for Sharpening and Leveling
. Scientists rarely make exact predictions. For instance, rather than stating that “54% of the electorate favors a tax on imported oil,” they will say that “54% plus or minus four percent favor an oil import tax.” Scientific predictions are almost always given as a range or “confidence interval.” Thus, the Center for Disease Control might say that “we estimate that somewhere between 500,000 and 1,500,000 people are infected with the HIV virus in the United States.” The larger number is obviously more newsworthy, however, and so news reports will often drop the range and report only the higher figure: “The CDC reports that as many as one and a half million people…,” We should be aware that any statement of the form “as many as” means that an extreme end of a confidence interval has been sharpened and presented for our attention. We must learn to scale such estimates down and put more faith in the downsized estimate.

Be Wary of Testimonials
. Often the media tries to impress us with the seriousness of some problem by presenting a vivid testimonial of an individual who suffers from it. These accounts are enormously successful in getting us to imagine what it would be like to be in similar circumstances, and they make us much more sympathetic toward those who suffer some bad fate. And well they should! By themselves, however, there is no reason they should have much influence on our sense of the
prevalence
of some malady. Their impact on our compassion is perfectly justified; their influence on perceived commonness is not. Any testimonial, no matter how moving, represents the experience of only one person. Often there is little reason to believe that that person’s experience is any more informative than one’s own in estimating overall prevalence. We should not allow the depth of our feeling toward any one person to influence our assessment of how many such people there are. Be wary of testimonials that urge us to do just that.

Summary
. Implicit in the discussion throughout this chapter is the idea that many of the inaccuracies that are part and parcel of secondhand information have an unfortunate impact on what people believe. It can hardly be otherwise. A person’s conclusions can only be as solid as the information on which they are based. Thus, a person who is exposed to almost nothing but inaccurate information on a given subject almost inevitably develops an erroneous belief, a belief that can seem to be “an irresistible product” of the individual’s (secondhand) experience.

*
This analysis is not meant to be an exhaustive account of the goals people try to fulfill in the process of communication. I discuss only those goals that are most likely to introduce bias and distortion into the content of the communication. For a more complete account of the goals of conversation and communication, the reader should consult D. Cushman & G. C. Whiting “An approach to communication theory: Toward consensus on rules.”
Journal of Communication
1972,
22
, 217-38; H. P. Grice, “Logic and conversation.” In P. Cole, & J. Morgan (Eds.),
Syntax and semantics
(Vol. 3). New York: Academic Press. 1975; or J. R. Searle, (1969)
Speech acts: An essay in the philosophy of language
. Cambridge, England: Cambridge University Press.