Monday, August 29, 2005

Pundit Innumeracy

In today's LA Times, Rachel Shteir notices that there sure is a lot of theft going on. How bad is it? Why, "a tweedy dealer of antiquarian books and maps" was recently charged with stealing valuable maps from the Yale library. But Shteir doesn't depend on anectdote alone. She reports that "according to the latest FBI Uniform Crime Reports, a theft occurs every three seconds."

Having established that theft is common, Shteir glides into stating that "theft has exploded today . . ." because of a variety of social forces, including the workings of market capitalism, the real estate boom, habits derived from music "file sharing," and moral laxness (referring to habitual stealing as "theft addiction.") But how much evidence does she supply that theft is actually increasing? None, zilch, zero.

So Citizen Cain will help out by checking the data from the Bureau of Justice Statistics National Crime Victimization survey. What does it show? Theft victimization rates of more than 400 per 1000 households in the mid-1970s, plummeting to just over 100 per 1000 by 2003. Gee, sure doesn't seem like theft is increasing.

Shteir is fond of the Uniform Crime Reports. Citizen Cain prefers the National Crime Victimization survey. But, let's look at the UCR too-- the UCR index of property crime (burglary, larceny-theft and auto theft). This measure shows a 23 percent decrease between 1994 and 2003.

Like many innumerate pundits, Shteir doesn't seem to know the difference between asserting that there's a lot of X, and stating that X is increasing. She's writing a book on kleptomania. Perhaps before she finishes it, she should learn this difference. While she's at it, she might want to think about the implications for her argument of the decrease in theft. If theft is decreasing, it's hard to see how the real estate boom and the overly forgiving term "theft addiction" are promoting theft.