Friday, September 25, 2009

murder and ice cream!

There's an article on MSNBC today that's making me a little stabby. The headline: "Study: Spanked kids have lower IQs."

I assumed the headline was a bit of sensationalism on MSNBC's part. Anyone who hasn't heard "correlation does not imply causation" over and over again in the classroom could be fooled by this headline into believing that spanking causes low IQ. (The example we were always given is the strong positive correlation between murder rates and ice cream sales - both of which rise in the summer). That can't possibly be what the study concludes. Right?

Except that is exactly what the researcher is trying to say. He claims that he controlled for socioeconomic status and other factors in his study, and therefore that the lower IQ scores were actually caused by the spanking. Says the researcher, ""You can't say it proves it, but I think it rules out so many other alternatives; I am convinced that spanking does cause a slowdown in a child's development of mental abilities."

I'm pretty sure this researcher was "convinced" that spanking causes lower IQ before he ever started his experiment. He goes on to explain how he thinks spanking lowers IQ, something his study could not possibly have told him. Then I get stabby and have to close the window before I hurt my computer monitor.

1 comment:

  1. You may be right, but I'm not buying any more ice cream. I don't want to get myself or someone else killed!