Two Negatives Do Not Equal Two Positives
9th February 2010
Consider my Rule: If You Can’t Count It, You Can’t Change It.
Please realize this Rule does not mean: If You Can Count It, You Can Change It.
I can count the gray hairs in my beard, the wrinkles on my face, and the stretch of the skin over my elbow, and even if you color, nip, or tuck them, I’m still getting older. You can count it, but you cannot change it. Many smart people miss this distinction and believe that the power of counting provides the power of changing.
Consider various Credible Sources in the now running gag of calorie counts on menus as a way to change waistlines. A common sense reading of the scientific literature indicates that you can do a lot of counting on this problem, but Calorie Counts won’t change hips, thighs, and bellies. People do not change their diet enough from this intervention to create a practical, lasting change. But Cool Credible Sources keep counting into the wind.
Consider this NYT unsigned editorial and the scientific study they cite in the editorial. Both pieces argue that Calorie Counts are good, they work, and they should be expanded. Please try and read the new scientific study the Times references, because that study is a great illustration of Counting Ain’t Changing.
First, realize that this scientific study is not quite scientific in the usual sense of the term. It comes from guys who call themselves scientists (okay economists, but most of us here aren’t real scientists, meaning physicists, but we’ll ignore that distinction), working in a scientific program (a Graduate School of Business), at a university that does science (Stanford). So, the paper is scientific in that sense. But, the paper as cited by the Times is not scientific in the sense that it has been reviewed, accepted, and published in the scientific literature. It’s been published by PR release, no peer review now. Inspect it closely, but you’ll find no mark of consideration and acceptance from a larger field of experts.
Second, while these economists know how to count, they do not know how to change. They find that when Starbucks restaurants started using calorie counts on their menus, the amount of ordered calories decreased 6%. Working the way economists do with assumptions (assume this 6% always works with all menus and every restaurant uses them and Americans obtain about 25% of the calories from restaurants and on and on), the economists explain that 6% difference in the real world as:
If average daily intake is around 2,000 calories, the implied calorie reduction is 30 calories per day.
Thirty. Trente. Dreißig. Treinta. Si. San Jyu. A day.
Thirty calories is a real thing. You can make 30 calories in a chemistry experiment or by slicing a slice of Wonder Bread in half. Calories exist and you can count them. But, if by some miracle you could create an intervention that caused everyone to eat 30 fewer calories a day, it would have no practical impact on their waistlines. Their metabolism would adjust to this frip of a decrease in a few days or weeks and maintain the old body weight.
Yet, the study authors and the unnamed Times editorialist think that their counts means your change.
Consider, finally and briefly, another Rule.
There’s a Difference between Persuasion, and Smoke and Mirrors; With Persuasion the Illusion Lingers.
Persuasion makes a change. The Cool Credible Counters at the Times and Stanford only imagine that they do. The Times and Stanford offers the Image of Truth in their Mirrors, but not the Persuasion of Truth because their claims fail. Cassandra cautions these writers: Credibility for Attention is a Losing Trade.