I agree with the letter of Kevin Morris and his response to the letter of Gregg Calkins. With tongue in cheek, here is an example (mind you, a very bad example concocted for the purpose of emphasis) of what I think Mr. Morris is driving at.
Since the beginning of the 20th century, there has been a steady decline in the number of boys named Roy (source: www.babynamewizard.com). Over the same period of time, we are told that global temperatures have been steadily rising. It is possible to find a statistical correlation between these two variables — and the correlation could be relatively strong.
Proponents with the zeal of Al Gore would say that the proper conclusion is beyond debate: The steady rise in temperatures is the result of human activity and is being caused, at least in part, by parents who fail to name their boys Roy.
Opponents with equal zeal would maintain that the opposite is true. They would say that the declining number of boys named Roy is the result of rising global temperatures.
So, who’s right? As Kevin Morris would put it: Just because there is a statistically significant relationship between variables … “it’s anyone’s guess which caused which or whether a third unknown variable is causing both.” As is illustrated by the present bad example, I would add the possibility that neither caused either and that there is no variable that caused both.