"The thing I’m most interested in right now has become a kind of crusade against correlational statistical analysis—in particular, what’s called multiple regression analysis. Say you want to find out whether taking Vitamin E is associated with lower prostate cancer risk. You look at the correlational evidence and indeed it turns out that men who take Vitamin E have lower risk for prostate cancer. Then someone says, "Well, let’s see if we do the actual experiment, what happens." And what happens when you do the experiment is that Vitamin E contributes to the likelihood of prostate cancer. How could there be differences? These happen a lot. The correlational—the observational—evidence tells you one thing, the experimental evidence tells you something completely different.
In the case of health data, the big problem is something that’s come to be called the healthy user bias, because the guy who’s taking Vitamin E is also doing everything else right. A doctor or an article has told him to take Vitamin E, so he does that, but he’s also the guy who’s watching his weight and his cholesterol, gets plenty of exercise, drinks alcohol in moderation, doesn’t smoke, has a high level of education, and a high income. All of these things are likely to make you live longer, to make you less subject to morbidity and mortality risks of all kinds. You pull one thing out of that correlate and it’s going to look like Vitamin E is terrific because it’s dragging all these other good things along with it.
This is not, by any means, limited to health issues. A while back, I read a government report in The New York Times on the safety of automobiles. The measure that they used was the deaths per million drivers of each of these autos. It turns out that, for example, there are enormously more deaths per million drivers who drive Ford F150 pickups than for people who drive Volvo station wagons. Most people’s reaction, and certainly my initial reaction to it was, "Well, it sort of figures—everybody knows that Volvos are safe."
Let’s describe two people and you tell me who you think is more likely to be driving the Volvo and who is more likely to be driving the pickup: a suburban matron in the New York area and a twenty-five-year-old cowboy in Oklahoma. It’s obvious that people are not assigned their cars. We don’t say, "Billy, you’ll be driving a powder blue Volvo station wagon." Because of this self-selection problem, you simply can’t interpret data like that. You know virtually nothing about the relative safety of cars based on that study.
I saw in The New York Times recently an article by a respected writer reporting that people who have elaborate weddings tend to have marriages that last longer. How would that be? Maybe it’s just all the darned expense and bother—you don’t want to get divorced. It’s a cognitive dissonance thing.
Let’s think about who makes elaborate plans for expensive weddings: people who are better off financially, which is by itself a good prognosis for marriage; people who are more educated, also a better prognosis; people who are richer; people who are older—the later you get married, the more likelihood that the marriage will last, and so on.
The truth is you’ve learned nothing. It’s like saying men who are a somebody III or IV have longer-lasting marriages. Is it because of the suffix there? No, it’s because those people are the types who have a good prognosis for a lengthy marriage.
A huge range of science projects are done with multiple regression analysis. The results are often somewhere between meaningless and quite damaging.
I find that my fellow social psychologists, the very smartest ones, will do these silly multiple regression studies, showing, for example, that the more basketball team members touch each other the better the record of wins.
I hope that in the future, if I’m successful in communicating with people about this, there’ll be a kind of upfront warning in New York Times articles: These data are based on multiple regression analysis. This would be a sign that you probably shouldn’t read the article because you’re quite likely to get non-information or misinformation.
Knowing that the technique is terribly flawed and asking yourself—which you shouldn’t have to do because you ought to be told by the journalist what generated these data—if the study is subject to self-selection effects or confounded variable effects, and if it is, you should probably ignore them. What I most want to do is blow the whistle on this and stop scientists from doing this kind of thing. As I say, many of the very best social psychologists don’t understand this point."