Monday, January 26, 2009

Repeat After Me: Correlation Does Not Necessarily Mean Causation



A good example of that is this study about child neglect and breast feeding:

Breast-feeding has well-documented benefits. Studies have shown it nourishes babies while fighting off infections and even boosting IQ. Now a study in Monday's Pediatrics suggests nursing also may protect infants from neglect.

In a study of 6,621 Australian children over 15 years, researchers found that those who were breast-fed were far less likely to be neglected or abused by their mothers. Babies who weren't breast-fed were more than 2½ times as likely to be maltreated by their mothers as those who were nursed for four months or more, the study shows. There was no link between breast-feeding and the risk of maltreatment by fathers or others.

I really think (as does one psychologist also interviewed in the quoted article) that the first sentence in the second paragraph sounds a lot more credible if written like this:

"researchers found that those who were neglected or abused by their mothers were far less likely to be breast-fed"

To find a correlation between two variables doesn't tell us anything about causality. In this case I doubt that we are even talking about causality as such but about two variables that may go together.

This study is also a good example of a whole new generation of studies which determine what 'good mothering' is. There might be nothing wrong with that if we saw equally many interesting studies about what 'good fathering' is, but because we don't see those studies, we assume that only what mothers do matters. Note that breast-feeding had no correlation with neglect or abuse by the children's fathers. But something else may well show such a correlation (alcohol consumption? unemployment? religious affiliation? depression?) and it might be useful to learn about that something else.