Rejoinder: Better Training, Deeper Thinking, and More Policing
Xiaoli Meng · The American Statistician · 2010
ends with what I consider as a good example of selfpolicing: “The results shown here do not rule out a physical origin, but point out that a result obtained after a statistical analysis carries, in addition to the physics behind, the spurious byproducts of the method applied.” Compared to Hoerl and Snee, von Collani gave me too much credit. Von Collani labeled my article as a “milestone in the development of science,” and credited me as a revolutionary in policing science. While flattered, I must confess that I am at least a mile away in seeing the pictures von Collani is painting, and I am not sure if I would make a half turn or full turn in my grave if someone puts “Chief of Science Police” on my tombstone. Von Collani apparently is questioning the entirety of modern science and statistics and wants to replace everything by “stochastic thinking,” a discussion topic that is the furthest from my original piece, certainly beyond my reflection antenna. But my statistical thinking compels me to express skepticism of just about any claim of one size fitting all, especially when it comes to matters as complex and grand as science and statistics. THE POWER OF COLLECTIVE WISDOM AND ACTION The experience starting from reading Brown and Kass (2009) to preparing this rejoinder reminded me once more of the power of collective wisdom. No matter how thoughtful, articulate, and well-intentioned each of us is, our individual contributions are inevitably idiosyncratic and can even carry ironies that are obvious to everyone but ourselves. For example, I almost choked on the great wine I was enjoying with the discussions when I read Kotz’s question “Why did Meng stop with scientists and policy makers?” Indeed! Although I probably would not go as far as to include Easterling’s car dealers and shoe salesmen, how could I forget to include “whole generations of future teachers” in an article that is mostly about teaching future generations? This brings me to my concluding point, the same as the one in my original piece, and on another historic occasion, the first days of the new decade. I share Fox’s and others’ enthusiasm and optimism that our future is very bright, but to ensure that such enthusiasm and optimism will be carried over to future generations, we need collective action. So please do anything you can to help build the “statistical leadership” that Hoerl and Snee articulated: one lecture at a time, one speech at a time, one consultation at a time, and one publication at a time. And a toast to the new decade: may we all have that sickto-our-stomach feeling at least once before lecturing/speaking/ consulting/publishing! [Received January 2010. Revised January 2010.]