The Oregon State University newspaper (The Daily Barometer) had this to say last week:
According to a press release issued by the Women's Center, 2,000 rapes occur every five minutes.
Huh -- 2,000 rapes every five minutes. That would mean 2000 x (60/5) x 24 x 365 = 200 million rapes a year (presumably in the U.S.). Many people underestimate the frequency of rape. Still, one would hope that it doesn't happen 200 million times a year; at least a little bit of multiplication should have alerted the writer and the editor that something was wrong.
Something was indeed wrong; when I e-mailed the Barometer to ask what the source was, they pointed me to the press release they were citing. It reads:
About 2,000 rapes are committed daily at the rate of about one every 5 minutes.
Not 2,000 rapes every five minutes, it turns out, but 2,000 rapes daily, or one every 5 minutes. Off by a factor of 300 (5 x 60) from how the newspaper rendered it.
But wait! A "rate of about one every 5 minutes" would be about 300 daily ((60/5) x 24), not about 2000 daily. The Women's Center press release was also mistaken (on at least one of the statistics, and maybe both); and again a little multiplication would have helped catch this.
For those who are interested about what the real number actually is, the answer of course is that we don't know for sure. The National Crime Victimization Survey, a survey of noninstitutionalized Americans age 12 or over, estimates that there were 72,000 completed rapes in the U.S. in 2003, plus 45,000 attempted rapes, and 82,000 sexual assaults (completed or attempted attacks short of vaginal, anal, or oral penetration); the 72,000 number would of course translate into roughly 200 rapes daily, not 2000. On the other hand, other studies have reported considerably higher levels, including the 700,000 number that corresponds to 2000 daily (though this was from the early 1990s, and the rape rate has apparently fallen considerably since then). To my knowledge, there continues to be a hot debate about the number (though not about whether 60/5 x 24 = 2000).
I e-mailed the Daily Barometer and the Women's Center to ask what's up, and to suggest that a correction be published (or, as to the web site, simply made); no response from them yet, I'm afraid.
(Note that the NCVS site was down when I checked the link; I fortunately have a printout from which I read the data, but I wanted to alert people that they might have trouble accessing the data themselves.)