Nov 17 2009
Bitter About Chocolate Benefits Report
A ScienceDaily news release about the benefits of eating dark chocolate put me in a temporarily bad mood (maybe five seconds). No, I didn’t eat a candy bar to pull myself out of the transient funk.
Let’s break down the first paragraph. Sentence #1:
The “chocolate cure” for emotional stress is getting new support from a clinical trial published online in ACS’ Journal of Proteome Research.
Okay, so this is the lead sentence. The enticer. We’ll give it some slack. Had chocolate cure not been in scare quotes, however, I would have jumped all over it. A treatment is not a cure.
Sentence #2:
It found that eating about an ounce and a half of dark chocolate a day for two weeks reduced levels of stress hormones in the bodies of people feeling highly stressed.
There is something important missing from this second sentence. If that something were provided somewhere else in the article, it wouldn’t be a big deal. That something is how much the levels of stress hormones were reduced. Good science depends on precision; good science writing shares some of that precision. Besides, there are not two possible outcomes for treatments: Worked and Didn’t Work. I believe we’ve got to stop promoting this simplistic notion. We misinform and miseducate when we do.
Sentence #3:
Everyone’s favorite treat also partially corrected other stress-related biochemical imbalances.
As a personal aside, chocolate is not my favorite treat. I’d opt for a bite of savory steak or a sip of a salty martini every time. I guess I don’t have much of a sweet tooth.
Speaking of steak, my beef with the third sentence is similar to the second. At least the writer added the qualifier (quantifier?) partially.
Does the how much of a finding really matter?
Consider this analogy: I work at a department store selling “failure insurance.” One day You buy a USB flash drive for $29.95. I talk you into buying the optional “failure insurance” for just fifty cents more. USB drives do fail, so why not buy it?
The next day you buy a top-of-the-line adjustable wrench for 29.95. Again I try to talk you into buying my failure insurance. It’s just fifty cents. Believe it or not, I tell you, wrenches do fail. Would you buy it?
In each of the above cases you would be remiss not to inquire about the actual failure rate.
Do numbers matter; does precision matter? Only if you want to be better informed.
Oh wow! I knew it! This is good news for me… I love chocolates… Before this news came out, it used to be that I eat chocolates when I’m under stressed, I’m not sure if it’s psychological, but I always feel better after eating one bar.