The Volokh Conspiracy - "A CBS News Investigation Uncovers A Suicide Rate for Veterans Twice That of Other Americans" -- Or Does It?

The quote is drawn from this CBS News story, with results that CBS labels "stunning." And the story claims that the stunning results indeed remain robust when one controls for sex. Here's the data on the Methodolgy page:

Results for 2004

Overall Rates
Veterans: 17.5 to 21.8 per 100,000
Non-Veterans: 9.4 per 100,000

Male Rates
Veterans: 30.6 to 38.3 per 100,000
Non-Veterans: 18.3 per 100,000

Female Rates
Veterans: 10.0 to 12.5 per 100,000
Non-Veterans: 4.8 per 100,000

Results for 2005

Overall Rates
Veterans: 18.7 to 20.8 per 100,000
Non-Veterans: 8.9 per 100,000

Male Rates
Veterans: 31.5 to 35.3 per 100,000
Non-Veterans: 17.6 per 100,000

Female Rates
Veterans: 11.1 to 12.3 per 100,000
Non-Veterans: 4.5 per 100,000

Yet here's something odd about the data: For the overall rates to correspond to the male and female rates, the veteran pool would have to be 62% to 64% female, and the nonveteran pool would have to be about 66% female. Check it out, for instance, with the lower bounds on the 2005 data: 11.1 x 0.63 (female) + 31.5 x 0.37 (male) = a bit under 18.7 (overall).

Or, if you prefer, consider a veteran pool of 23 million people. You'd need:

14.5 million (23 million x 0.63) women, or 145 hundred-thousands, with 1610 suicides (145 x 11.1) and
8.5 million (23 million x .037) men, or 85 hundred-thousands, with 2680 suicides (85 x 31.5) to get
a total of 4290 suicides (a bit under 18.7 per 100,000) for the whole 23 million.

That can't be right. The VA reports that the veteran pool is only 7% female, which means that if the CBS News overall veteran and female veteran numbers are right, then the male veteran numbers would be 19.27 to 21.4 in 2005 and 18.06 to 22.5 in 2004, not far from the male suicide rate of about 17.7 per year (see WISQARS). Of course, we can't tell which of the CBS numbers are right — but it does seem like they can't all be right. Plus of course you can't have a population that's about 51% female but at the same time 62-64% female among veterans and 66% female among nonveterans.

Or am I missing something? Please let me know.

Thanks to Ares (an Aviation Week & Space Technology blog) (linked to InstaPundit) for pointing to this story, though Ares doesn't seem to discuss the specific numbers listed on the CBS Methodology page. Note also that CBS reports, "One age group stood out. Veterans aged 20 through 24, those who have served during the war on terror. They had the highest suicide rate among all veterans, estimated between two and four times higher than civilians the same age. (The suicide rate for non-veterans is 8.3 per 100,000, while the rate for veterans was found to be between 22.9 and 31.9 per 100,000.)" The suicide rate for males age to 20 to 24 is about 20 to 21 per 100,000; what the suicide rate is for male veterans age to 20 to 24 seemingly can't be determined from the CBS numbers, given the apparent flaws I identified above.

	[Eugene Volokh, November 14, 2007 at 7:34pm] Trackbacks "A CBS News Investigation Uncovers A Suicide Rate for Veterans Twice That of Other Americans" -- Or Does It? The quote is drawn from this CBS News story, with results that CBS labels "stunning." And the story claims that the stunning results indeed remain robust when one controls for sex. Here's the data on the Methodolgy page: Results for 2004 Overall Rates Veterans: 17.5 to 21.8 per 100,000 Non-Veterans: 9.4 per 100,000 Male Rates Veterans: 30.6 to 38.3 per 100,000 Non-Veterans: 18.3 per 100,000 Female Rates Veterans: 10.0 to 12.5 per 100,000 Non-Veterans: 4.8 per 100,000 Results for 2005 Overall Rates Veterans: 18.7 to 20.8 per 100,000 Non-Veterans: 8.9 per 100,000 Male Rates Veterans: 31.5 to 35.3 per 100,000 Non-Veterans: 17.6 per 100,000 Female Rates Veterans: 11.1 to 12.3 per 100,000 Non-Veterans: 4.5 per 100,000 Yet here's something odd about the data: For the overall rates to correspond to the male and female rates, the veteran pool would have to be 62% to 64% female, and the nonveteran pool would have to be about 66% female. Check it out, for instance, with the lower bounds on the 2005 data: 11.1 x 0.63 (female) + 31.5 x 0.37 (male) = a bit under 18.7 (overall). Or, if you prefer, consider a veteran pool of 23 million people. You'd need: 14.5 million (23 million x 0.63) women, or 145 hundred-thousands, with 1610 suicides (145 x 11.1) and 8.5 million (23 million x .037) men, or 85 hundred-thousands, with 2680 suicides (85 x 31.5) to get a total of 4290 suicides (a bit under 18.7 per 100,000) for the whole 23 million. That can't be right. The VA reports that the veteran pool is only 7% female, which means that if the CBS News overall veteran and female veteran numbers are right, then the male veteran numbers would be 19.27 to 21.4 in 2005 and 18.06 to 22.5 in 2004, not far from the male suicide rate of about 17.7 per year (see WISQARS). Of course, we can't tell which of the CBS numbers are right — but it does seem like they can't all be right. Plus of course you can't have a population that's about 51% female but at the same time 62-64% female among veterans and 66% female among nonveterans. Or am I missing something? Please let me know. Thanks to Ares (an Aviation Week & Space Technology blog) (linked to InstaPundit) for pointing to this story, though Ares doesn't seem to discuss the specific numbers listed on the CBS Methodology page. Note also that CBS reports, "One age group stood out. Veterans aged 20 through 24, those who have served during the war on terror. They had the highest suicide rate among all veterans, estimated between two and four times higher than civilians the same age. (The suicide rate for non-veterans is 8.3 per 100,000, while the rate for veterans was found to be between 22.9 and 31.9 per 100,000.)" The suicide rate for males age to 20 to 24 is about 20 to 21 per 100,000; what the suicide rate is for male veterans age to 20 to 24 seemingly can't be determined from the CBS numbers, given the apparent flaws I identified above.