Tag Archives: Journalism

What, me worry?

John Tierney of the NYT does something unprecedented for a Journalist; he tells you about 10 things you DON’T need to worry about. It’s well worth reading, as is his recent follow up.

I would love to see an IT Journo imitate Mr. Tierney and publish a list of the 10 things a CIO or CSO doesn’t need to worry about. It would be a refreshing change.


Bad Science Journalism

Sometimes Bad Science Journalism is viewed as a tautology. There is no subject journalists seem to understand as poorly as science. This Reuters article by Michael Kahn illustrates why. There is no link to the actual underlying study (a common failing in science journalism) so it’s hard to tell how much fault lies with Mr. Kahn or with the study authors. But I have my suspicions.
Regardless of where the errors arise, the Mr. Kahn makes three clear basic mistakes; being incorrect on what is being measured, lack of indication of statistical significance of the data, and confusion of correlation and causation.

First and worst is the title of the article:

Gene explains why breast-feeding makes kids smarter

Wrong. As Doc Searls points out here, IQ tests do not actually measure intelligence. They measure how good you are at taking IQ tests. So the headline should have read:

Gene explains why breast-feeding makes kids do better on IQ tests

But do they? The article mentions a 6-7 point increase in measured IQ by among children who had the specific gene they were studying. But how statistically significant is 6-7 points in the sample size? What was the range and variance of measured IQ in the study population? Did all the children get tested at the same age? Did the get tested by the same examiner? None of this appears in the article. I doubt the reporter even looked into these questions. So the headline should have read:

Gene explains why breast-feeding might make kids do better on IQ tests

But is that even true? The journalist jumps right in with assumed causation. The unasked question is why did some mothers breast feed while others did not? Could there be a genetic factor? So the headline should have read:

Gene may explain why breast-feeding might make kids do better on IQ tests

Why does all this matter? First it gives people the wrong impression of how science should work. But in this case it is far worse than that. Breast-feeding is a very personal and difficult decision for mothers to make. Some choose not to breast feed and some simply can’t for physical reasons (or at least need to supplement with formula).

Mothers who are facing this choice, or who made it already, don’t need to be bombarded with articles claiming that they are harming their children. Not unless there is clear scientific evidence that is the case. There doesn’t seem to be in this article.

(Mirrored from TalkBMC)