The Use of Statistics
To the Editor:
Whatever other faults Ramsey Clark’s Crime in America may have, misrepresentation of statistics is not one of them, despite James Q. Wilson’s remarks in his review [Books in Review, March]. When Clark claims that the statistic “one person in 400 is the victim of a crime of violence in a given year” implies that the average individual’s chance of being a victim is once in 400 years, he is substantially correct. If the probabilities remain the same, this average person can expect to be a victim only once in 400 years, though the attack could come at any time during this period.
Mr. Wilson then goes on to commit a very common blunder in his own calculation of probabilities, though in this instance the numerical difference is not great. If the average person has a one-in-200 chance of being a victim of a violent crime each year, then over a seventy-year period his chance of being a victim is one minus the quantity 199/200 raised to the seventieth power, or about 0.296, not seventy times 1/200 or 0.350 as Mr. Wilson claims. For a black female, the lifetime chances are slightly lower than one in two, not better than one in two. But all these calculations depend on the probability of attack remaining the same from year to year, which itself is an unlikely event.
Edward J. Seligman
Broad Brook, Connecticut
James Q. Wilson writes:
I am happy to defer to Mr. Seligman’s grasp of probability theory but not to his reading of the implications of Ramsey Clark’s statistic. As Mr. Seligman notes, my crude estimate of the lifetime probability of victimization (.350) is not very different from his exact one (.296). But both figures, I persist, are very different from the soothing suggestion by Mr. Clark that the average person has only a one-in-400 chance of victimization. Without further explanation, which Mr. Clark does not provide, that statistic will be read by the average person as meaning that the odds are 399-to-one against his victimization. He will conclude that public concern over events so “rare” is simply misplaced. I have spoken to many people who have read the figure in precisely this way, just as I have spoken to persons who have read Mr. Hoover’s “crime clock” to mean that on every street corner there lurks a mugger. I am sure Mr. Seligman would join me in asking that both sides shun the use of “clever” statistics.