I thought this article was interesting enough to warrant posting and commentary:
QUOTE
Thursday, March 23, 2006
Damned lies and statistics
Probably everybody has read RedState's analysis of US Active Duty military deaths. It says:
Take a look at the actual US Military Casualty figures since 1980. If you do the math, you wil find quite a few surprises. First of all, let's compare numbers of US Military personnel that died during the first term of the last four presidents.
George W. Bush . . . . . 5187 (2001-2004)
Bill Clinton . . . . . . . . . 4302 (1993-1996)
George H.W. Bush . . . . 6223 (1989-1992)
Ronald Reagan . . . . . . 9163 (1981-1984)
Even during the (per MSM) utopic peacetime of Bill Clinton's term, we lost 4302 service personnel. H.W. Bush and Reagan actually lost significantly more personnel while never fighting an extensive war, much less a simulaltaneous war on two theaters (Iraq and Afghanistan). Even the dovish Carter lost more people duing his last year in office, in 1980 lost 2392, than W. has lost in any single year of his presidency. (2005 figures are not available but I would wager the numbers would be slightly higher than 2004.)
Winds of Change took exception to this analysis and posted this.
Occasional WOC contributor Captain Midnight took a moment to review Proud Kaffir's numbers, The Captain writes, "He is citing raw absolute numbers. They are the result of two things: A 27% drop in the size of the armed forces, and a 54% drop in the death rate from accidents. If you plot the death rate per 100,000 and break it out by cause of death, you get
(from Winds of Change)
As the Captain notes, the graph clearly shows a decrease in deaths due to accidents and a significant upsurge in deaths per 100,000 troops due to hostilities. Misuse of statistics is tendentious no matter what the position being defended.
The comments section at both Red State and Winds of Change raised some interesting points. A Little to the Left at Red State says:
That leads to an average number of deaths per 100,000 soldiers of:
Reagan: 43.22
Bush I: 31.75
Clinton: 27.75
Bush II: 37.05
So as you can see a greater percentage of soldiers die each year under George W. Bush than under either George H.W. Bush or Clinton or Carter. And please don't even try to claim that all that matters is the absolute number of soldiers killed.
But J Steele at Red State retorts:
The number that counts is the number the press uses, the absolute number of deaths in theatre regardless of the cause. That's the number you guys on the left like to use to scare the American people into believing that we can't beat the terrorists so we might as well give up and come home and hide under the bed.
Patrick Brown at Winds of Change raises the interesting point that the deaths per hundred thousand make small armies look bad.
But here's another way to look at it. Suppose that the US armed forces had been cut back to 100,000 men and that only 500 of them had been killed in the Iraq war. That death rate per 100,000 would have been much greater than the actual current rate - but it would have been a substantially better situation for us (setting aside any other consequences of reduction in strength).
Comment
A statistic is a number computed from a sample to describe some attribute of interest. Whether or not a statistic is a good answer depends on what question it is intended to answer. If the question is 'how many Americans have died while on active duty over a certain time frame', then Red State's figures give the arithmetic totals. If the question is 'how many deaths per hundred thousand active duty servicemen have there been' then Captain Midnight's quotients are correct. If the question were 'what percentage of military manpower did active duty deaths consist of' it would be different again because the population would have been growing since the 1980s.
For a relative who answers a notification knock at the door, it is absolute numbers that matter. "Is my son alright" and not "how many deaths per hundred thousand" are what concern him then. To a man thinking of a career in the services, deaths per hundred thousand in your military specialty matter very much indeed. The big red spike in combat deaths under present conditions means that you are safer in the Navy or the Air Force than in the ground forces, and if in the ground forces, safer anywhere except in the combat arms. For the public, it is what is not on the graph or the tables that matters most of all. It is how much peace and security this sacrifice has bought. In each of the Cold War years there were those who reviled the military as unnecessary though they brought us home through the Valley of the Shadow of Nuclear Death. And many who claim that the lives spent in fighting terror have been wasted drive straight to their offices each day increasingly confident that September 11 will never happen again. What answer would they want, I wonder, to which question?
posted by wretchard at 1:42 AM
It is undeniably true that statistics can be made to say anything you care to have them say, provided you are willing to be, um, 'selective' enough.
But I think myself that the underlying story here is in comparion to the oft-proclaimed sentiment that we are somehow losing "too many troops" in Iraq compared to what the effort to rescue that country is "worth."
Looking at the accidental death raw numbers, it should be amazing that there was no similar sentiment during peacetime regarding losing "too many troops" for what to many would seem to be "NO purpose."
For me, the bottom line is this: Every one of use dies, andwe each only die once. For an individual soldier, the death statistic that REALLY matters is whether or not it is your own, and whether or not that risk was taken, and the price ended up being paid, in service of something you believed in.
The press has focused with relentless intensity on the raw number total of casualties, without regard to the size of the force or comparisons to the toll paid annually just in training to be able to fight. And in doing so, they have performed a disservice to anyone who lost their life wearing the uniform, but not facing an actual enemy.
Look at the SUMS of deaths during 'peacetime' years, and compare them with the total number of combat casualties since the war began.
Totals, both in raw numbers and in rate per thousands, of deaths due to combat over the last 25 years is MINISCULE compared to accidental deaths ovet the same period.
For a 25-year period including two major ground wars, that's unprecedented.