The question of whether to ban cell phone use in cars has been a fairly hot issue lately. It seems plausible to many people that using a cell phone while driving increases the driver's likelihood of being in an accident, but not a whole lot of formal studies have been done on this issue. One study that has been done and is very frequently cited is from the New England Journal of Medicine. The way that study's findings have been treated provides a helpful example of how easy it is to commit a causal fallacy.
Here is the central result from the study mentioned above:
"The risk of a collision when using a cellular telephone was four times higher than the risk when a cellular telephone was not being used." (Donald A. Redelmeier & Robert J. Tibshirani, "Association between Cellular-Telephone Calls and Motor Vehicle Collisions," (Abstract) New England Journal of Medicine, Vol. 336, No. 7, Feb. 13, 1997.)Here's a reference to that study from the newspaper USA Today:
"The 1997 study, published in the New England Journal of Medicine, concluded that talking on a cell phone while driving quadruples the risk of an accident." ("Cities, states, debate cell-phone bans" USA Today, Feb. 10, 2000)Something goes wrong between the study and the reference to it, but what?
What goes wrong is that the USA Today article makes the assumption that cell-phone use causes an increased risk of an accident, but the study merely makes the claim that cell-phone use is correlated with an increased risk of an accident.
What's the difference? For one thing, even if the correlation cited in the New England Journal study really exists, it's possible that the increased risk of an accident is actually caused by something other than cell-phone use. For example, it may be that people tend to use cell-phones when they're distracted by some business or personal issue. This distraction may cause both the cell-phone use and the increased risk of an accident.
Why does this mistake matter? Because studies like this are often used to justify legislation. Here, the theory might be: banning cell-phone use while driving will prevent accidents. But if, for example, the actual cause of the increased risk of an accident is the distraction, this theory may be false and the legislation may do no good.
The moral: Causal fallacies are common and often hard to notice, but failing to notice them can cause serious problems.