Why Do Americans from the Smallest States Die at a Higher Rate in Iraq?

In a recent piece in USA Today ("Why Smallest States Suffer Most In Iraq," 28 March 2008, at http://blogs.usatoday.com/oped/2008/03/why-smallest-st.html), Al Neuharth claims that a disproportionate number of deaths in the Iraq war come from states with the smallest populations and that two reasons explain this phenomenon: first, Americans from the smallest states earn less, and second, they are more patriotic. As a result of these two factors, Americans from the smallest states are, presumably, more likely to join the military and therefore to be killed in Iraq.

Neuharth provides numbers demonstrating a huge disparity between the "death rates" of the largest and smallest states. I have no way of independently analyzing these numbers, so I'll accept that they are correct. However, I don't buy the two supposed reasons for the disparity.

First, Neuharth claims that Americans from the smallest states earn less than those from the biggest. However, according to the Census Bureau (http://www.census.gov/hhes/www/income/income06/statemhi2.html), the median household incomes of the union's five smallest states:

State	Median Household Income Rank
Alaska	6
Vermont	16
Wyoming	28
South Dakota	34
North Dakota	39

compare quite well with those of the five biggest states:

State	Median Household Income Rank
California	11
Illinois	18
New York	21
Florida	33
Texas	38

Second, Neuharth claims that residents of the smallest states are more patriotic than those from the biggest, but he provides no evidence for this statement. Perhaps he is basing it on intuition? I suppose that there are many ways to measure patriotism—the number of Americans who fly U.S. flags on their homes or who have memorized the pledge of allegiance, perhaps—but doubt that there is any single method that most people would agree on. In any case, I—based on intuition—doubt that residents of the union's smallest states are demonstrably more patriotic than those from the biggest.

Yet even if we accepted both of Neuharth's premises, his argument would still be flawed. He appears to suggest that the disparities in income and patriotism lead more Americans from the smallest states to join the military, resulting in a greater number of deaths in Iraq. However, while a greater proportion of Americans from the smallest states do, in fact, appear to join the military, this moderate difference does not seem to account for the much greater death rate disparity.

The Department of Defense's "Population Representation in the Military Services (2002)" (http://www.defenselink.mil/prhome/poprep2002/chapter2/c2_geography.htm) provided "representation ratios" (fractions of 18-to-24 year olds who enlisted in the military, normalized so that a value of "1" represented the average state) for all 50 states. We find that only three of the five small states had above average representation ratios:

State	2002 Representation Ratio
North Dakota	.8
Vermont	.9
South Dakota	1.3
Alaska	1.5
Wyoming	1.6

while two of the five big states also had above average representation ratios:

State	2002 Representation Ratio
New York	.8
California	.8
Illinois	.9
Florida	1.3
Texas	1.3

Although the 2005 version of that document did not provide representation ratios, we can estimate those by dividing the number of active component enlisted accessions (which the report did provide, at http://www.defenselink.mil/prhome/poprep2005/appendixb/b_10a.html) by each state's population. I then multiplied these fractions by 1000, simply to avoid working with very small numbers. (Two caveats. First, I could not easily find 2005 population figures, so I used the Census Department's 2007 figures, available at http://www.census.gov/. I assume that states' populations did not change much between 2005 and 2007. Second, these numbers are different than the 2002 representation ratios, since they are not normalized and represent fractions of states' entire populations rather than of their 18-to-24-year-old populations.) Here are the results for the smallest states:

State	FY 2005 Active Component Enlisted Accessions	2007 Population	More Recent Representation Ratio
Vermont	228	621,254	.37
North Dakota	287	639,715	.45
South Dakota	495	796,214	.62
Alaska	470	683,478	.69
Wyoming	364	522,830	.70

and for the largest states:

State	FY 2005 Active Component Enlisted Accessions	2007 Population	More Recent Representation Ratio
New York	6,978	19,297,729	.36
Illinois	5,763	12,852,548	.45
California	16,845	36,553,215	.46
Florida	9,339	18,251,243	.51
Texas	17,070	23,904,380	.71

Again, while residents of the smallest states do seem to join the military in greater proportions than those of the largest states, the difference does not seem to be big enough to account for the much greater disparity in the death rates that Neuharth cites.

I do not know why the percentage of Americans from the nation's smallest states who die in Iraq is so much a greater than the percentage from the biggest states. However, what I do know is that Neuharth's analysis does not seem to explain this phenomenon.