Bad Science: Spanking Kids Leads to More Aggressive Behavior—Or Does It?

Time Magazine has published an article on a new Tulane University study that purports to show that children who are spanked leads to more aggressive children in the long run. On the surface, it makes sense—”aggressive” physical punishment leads to more aggressive children. Obviously it would be best then not to spank one’s children unless you want them to become bullies. The only problem? The study doesn’t hold water.

For one, the study made no distinction between various “types” of spankings. That in itself is seriously problematic, as not all spankings are equal. It is one thing for a parent to spank a child after having made a commitment never to do so in anger (if angry, wait until not angry); the point then is simply a matter of consequences rather than whether the parent is angry or not. The alternative behavior—parent gets angry, child gets whipped because of it—is a wholly other thing, training the child that the appropriate action when one is angry is to aggressively lash out at another. But this study made no distinction between “lawful” and “angry” spankings. My guess is that the latter would lead to the child sharing similar aggression, while the former would not. But studies are problematized by not accounting for this difference.

Another problem—everyone repeat after me: “Correlation does not imply causation.” “Correlation does not imply causation.” “His name is Robert Paulson.” Oh, sorry. Got off track there. The study notes that the students who were spanked more often were more aggressive later on.

… spanking remained a strong predictor of violent behavior. “The odds of a child being more aggressive at age 5 increased by 50% if he had been spanked more than twice in the month before the study began,” says [study lead Catherine] Taylor.

So a correlation was established. Congratulations. But here’s a question: why were these kids spanked more often than the others in the first place? Could it be, perhaps, that they were spanked more often because they were more aggressive to begin with? Oh, but we can’t ask that question if we have an axe to grind, as it appears was the case in this study. This study is a poster child for bad science,  a great example of all too many “scientific” studies done in service of a preconceived bias, with no regard for possible sources of error. And yet again, the “correlation does not imply causation” error, which is marker #1 of a biased study.

In discussing this issue over the past week, one interesting theme did continue to come up with those categorically opposed to spanking: a notion of the “right of physical inviolability” shared by all human beings, which apparently also extends to children and their parents. I find this a fascinating response, especially since I’m not sure how diaper changes and wiping an older child after a trip to the toilet would seem to violate such a notion. In addition, I’m not sure how imprisonment (or its childhood equivalents, “time-out” and “grounding”) is any less a violation of a person’s “right to physical inviolability,” as they restrict a person’s body to a specific place, removing the freedom of movement. (I would think Foucault’s Discipline and Punish should have divested folks of the idea that imprisonment is somehow the result of more humanitarian concerns. Frankly, I am a firm believer that imprisonment is a far less humane punishment than corporal punishment, as it effectively entails taking away a person’s life, whereas corporal punishment is more about humiliation than anything.) Thirdly, punishment is reserved for those who have committed crimes—disobedience to established authority involves waiving one’s rights commensurate with one’s actions. A person who commits a heinous enough crime (or a child who insolently disobeys to a certain degree) has forfeited his “right to physical inviolability”; that is the nature of law and punishment.

Regardless, there still is no scientific study that has done any sort of adequate job showing spanking to be an inherently bad tool in a parent’s disciplinary toolbox. I would strongly argue that no one should ever spank a child out of anger, but there is no reason to think that dispassionate spanking as a clearly defined punishment is somehow going to ruin one’s child.

*EDIT: Apparently, another study applied the same methodology used in this study to other punishments, finding the same results:

The only problem was that we got the same apparently harmful child outcomes for grounding, sending children to their room, and even for child psychotherapy.

This led us to conclude that something is wrong when the “strongest evidence yet against the use of spanking” is based on statistical analyses that would make psychotherapy for children look as harmful as spanking!

Our study shows that these results are biased because defiant children lead parents to use all disciplinary tactics more often as well as to seek psychotherapy for their child.

More on that study and some excellent analysis of child discipline (including quite a lot on spanking) here.