…as long as the things they are gullible about confirm their prejudices.
I’ve got a newfound – interest – in military welfare these days. So I pay particular attention to news items that cross my computer screen that touch on issues about the welfare of our military.
A few weeks ago, I saw the release about rising suicide rates in the military, and to be honest I was concerned.
Army soldiers committed suicide last year at the highest rate in 26 years, and more than a quarter did so while serving in Iraq and Afghanistan, according to a new military report.
It just made the Washington Post, and was picked up by lefty blogs TalkLeft and Greatscat!, who say – respectively – “This war was not worth the price. We have a President who is unable or unwilling to acknowledge his mistakes. Experts agree the war in Iraq cannot be won militarily. So why are the troops still there? Let’s stop the funding and bring them home now.” and “Um, I wonder if Rove left spin instructions for this report’s release?”
Gosh, guys. It is a damn serious issue – because the rate is rising, and because one suicide is unknowably painful to the circle of the person who kills themselves.
But you know, the first question I asked was “Gosh, I wonder how that compares to normal rates of suicide in the US?
So I spent a whopping ten minutes looking up stats and building a small spreadsheet.
So in 2004, there were a total of 14,328 suicides in the US in the age group 20 – 44 (the group that I think pretty well covers the population in Iraq – some are younger, some are older). the total population in 2004 in that age group was 104,259,000 – so the rate/100,000 population was 15.25.
And since the rate in the military is higher – significantly higher at 17.3/100,000 overall and 19.9 for those serving in Iraq and Afghanistan – I was darn concerned.
And then I asked one more question.
Well, the suicide rates by sex are pretty different, I recalled. I wonder what happens if I sex-norm the suicide rates in the military?
Here’s an approximation (because the %of women in reserves is slightly higher, and I didn’t find the % serving in Iraq).
According to the DoD, approximately 17% of US active-duty forces are women.
According to the CDC, the 20 – 44 population had 14,328 suicides in 2004. Of those, 11,460 were men, and 2,868 were women. The census gives an estimate for 2005 population from 20 – 44 as 52,513,000 men and 51,746,000 women.
By my math, this gives a suicide rate of 21.82/100,000 for men, and 5.54/100,000 for women.
If I norm the suicide rates by multiplying the sexes rate by the population in the military, I get (21.82*83%)+(5.54*17%)=19.06/100,000.
So the suicide rate among all active-duty troops is lower than the 2004 norm – even at the current high point – and the rate among combat troops is slightly above the norm.
Does this mean it isn’t serious and that we shouldn’t put resources into PTSD treatment or that each suicide isn’t itself a tragedy? What do you think I believe…come on, of course.
But is this a symptom of a military so brutalized by the horrors of service that they are killing themselves at an incredible rate?
What do you believe I think? Why can’t people do some freaking homework before the leap to the Isle of Conclusion – that’s what I think.
Lithuania had a male suicide rate of 75.6 per 100,000 in the last World Health Org figures. Isn’t it time they all moved to the Russian Federation, where it’s only 70.6?
A soldier serving in Iraq is less likely to kill himself than a civilian in Australia (21.2), France (26.1), Germany (20.2), Austria (27.3), Belgium (29.4), or Switzerland (26.5).
So this “all time high” puts him at about the same risk of suicide as a Canadian (19.5 – the leading cause of death among Canadian men aged 25-29).
Several articles are quoting James F. Dunnigan, who attributes the higher rate to the high number of deployed reservists. That makes sense, because reservists face greater disruption to their personal lives than active duty soldiers do. But not as much disruption as being a Belgian, apparently.
The measure of morale is morale, not suicide rates, and “there is a history of distorted reporting on morale,”:http://www.strategypage.com/dls/articles2005/200585193232.asp as this article from Dunnigan’s site maintains. The suicide rate in Iraq plunged to 8.5 in 2004, at a time when anti-war sentiment was ramping up and the situation in Iraq was “deteriorating”.
And we have some very recent history of poor reporting on morale, as in outright fabrications – but let’s not get into that.
Suicide rates vary and fluctuate without obvious cause (military suicide was an incredibly low 4.6 in Operation Desert Storm) and low enough in any event that it is absurd to try to correlate them with current events.
In part, it is because the military has historically had lower suicide rates then the at large population. I believe 12.3 (from the article) vs. 17/100k, with valleys much lower.
This article has nothing to do with comparing military suicide rates to the population as whole – at best, to do that, this information can be reframed as “historically low military suicide rates are approaching the rest of the population”. It would be good if they put this information into the article, and compared to how national trends have gone.
While the difference between 2005 and 2006 is very close (88 total to 99), the key information is here – 9.1/100k in 2001, 17.3/100k last year. While I can’t answer if it is brutalized by horrors, something is going on. Fatigue, despair, loneliness, separation, I pulled one of the quotes.
“Failed personal relationships, legal and financial problems and the stress of their jobs were factors motivating the soldiers to commit suicide, according to the report.”
This does not sound like standard PTSD (though PTSD could contribute). This sounds like what, well – what the population as a whole faces. What does worry me is:
“In addition, there was a significant relationship between suicide attempts and number of days deployed”
and
“was limited evidence to support the view that multiple … deployments are a risk factor for suicide behaviors,”
Dave while I think the trends are likely to be real, I also wonder if the Law of Small Numbers is in play. 40 people in a population of 2.5 million isn’t many, and I wonder – a lot – about statistical variance.
Having said that, my two points are
1) people who seize this as evidence of a broken military are being silly;
2) we do need to work hard on support for the troops – in all spheres…
A.L.
More likely than not, this spike is mere “random variation in the data”:http://www.outsidethebeltway.com/archives/2007/08/army_suicide_rate_and_random_variation/.
As Dave [#2] correctly points out, the relevant comparison is between previous suicide rates in our military, regardless of where those stand relative to the general population, here or in Belgium.
A.L. and James [#3,4] are also correct in thinking that this could be random variation. Maybe yes, and maybe no.
Science gives us some useful guidance. Science includes (at least) two phases of reasoning: hypothesis generation and hypothesis testing. Hypothesis testing is rightly conservative, with the null hypothesis (typically “random variation”) being given a significant advantage that has to be overcome by any successful alternative. Hypothesis _generation_ on the other hand, is no holds barred. (August Kekule is said to have generated the ring-shape hypothesis for benzene after a day-dream of a snake eating its tail. He followed this up, undoubtedly, with rigorous hypothesis testing before publication.)
So what does this mean about suicides in the military? If I see a disturbing up-tick, in a situation where a mechanism is operating that could potentially cause such a thing, it is perfectly reasonable to _generate_ the hypothesis that this mechanism caused that change in the data. What you can’t claim is that the hypothesis has been rigorously _tested_.
And then it’s perfectly reasonable to discuss such a hypothesis with friends and associates, including especially ones who are likely to be critical, because they may be able to find alternate explanations or flaws in this one. A useful outcome of such a discussion is that, sometime, we discover a way to do a rigorous test of the hypothesis using existing data.
And sometimes, depending on what’s at stake, it might be better not to wait for the data necessary for a conclusive test. When the canary dies in your mine, you probably shouldn’t ignore it, on the grounds that data from one canary is not statistically significant.
AL
“But is this a symptom of a military so brutalized by the horrors of service that they are killing themselves at an incredible rate?”
In the links you provided, I didn’t find anyone who made this claim.
“Why can’t people do some freaking homework before the leap to the Isle of Conclusion ”
Are you referring to the Army report? Should its authors have put it into the context that you believe would prevent people from the hypothetical leap to conclusions that you fear will happen without the context. Or is that the APs job? or WaPost’s? (On the other hand, since we haven’t seen the full report yet, perhaps it did put the stats in the context in which you believe they ought to be seen, and the AP/WaPost simply supressed the context.).
But the puzzling thing to me is that you seem to think the report, on its face, (as does seem the Army) describes a situation serious enough to merit some sort of attention, some sort of attempt to address the underlying cause. You seem to want it both ways. Yes, the Army report needs attention, but No, the press shouldn’t report on the report because someone might get the wrong idea.
I think you need to take into account that those who oppose the war believe that all the death is an unneccsary waste, whether soldiers are killed in vehicle accidents, by bombs, helicopter crashes or suicides, because they don’t believe that the underlying mission is either likely to succeed, worth the price or a combination of both. So it would be natural for them to add one more card to a growing deck of reasons to withdraw. Even if the increase in the rate of suicide was very slightly higher this year than last, that in itself would quite reasonably and naturally be seen as one such card.
And I do agree that if anyone were to argue that a slight rise in the rate of suicide among soldiers were, by itself, a sign of a broken military, he or she would be silly. However, so far, I haven’t heard of anyone making that argument. All I have seen is news reports that are mostly comprised of quotes from an army report (that seems to express some concern over the siutation) and two blogs that report the report of the report, followed by a short commentary that doesn’t really even address the report(s), but makes personal attacks against Bush and Rove.
I guess I just don’t see the deal being big enough to justify the post. But then this comment is probably already longer than your post and so there you have it.
I think the real story is the suicide rate among women:
bq. _In 2006, the overall suicide rate for the United States was 13.4 per 100,000 people. It was 21.1 per 100,000 people for all men aged 17 to 45, compared to a rate of 17.8 for men in the Army._
bq. _The overall rate was 5.46 per 100,000 for women, compared to an Army rate of 11.3 women soldiers per 100,000._
“UPI”:http://imedinews.ge/en/news_read/58288/
mark –
Jeralyn says:
“I blame President Bush. Every day he keeps our soldiers in this war, more of them are going to die. The ones that survive will come back with post-traumatic stress disorders that will take years if not decades to overcome. Some of them are bound to take their own lives as well.”
You don’t think that counts??
A.L.
Alternatively, they could all join the U. S. military.
Female suicide _attempts_ always far outstrip men. I dont have stats but i would imagine women in the army are simply more likely to suceed. I’d like to see female suicide attempts in the military vs the US at large, I bet its also lower than the average population. Easy access to a handgun does increase the statistical probability of succeeding in suicide (obviously).
AL et al,
Excellent point about statistics, comapring Army suicides to population as a whole, distorting effects of greater male to female in populations under study, the Law of Small Numbers, radnom variation, and so forth.
Outstanding analysis.
Too much reporting and commentary on Iraq involves partisans and advocates that take whatever sliver or biased amalgam of data to back up a predetermined and prejudged (prejudiced) assessment. The rhetorical ends justify the means, and facts are merely instruments of demagogery.
To those who make the argument that reporting has been neutral and without the implicit “the army is broken” and “morale is causing soldier to kill themselves,” you’d have to be very naive not to see the obvious storyline this reporting falls into. Watch the reaction of Lefty blogs for confirmation, they all scream what those of us rightward detect as subtext within the media reports.
A.L.: The article indicate that this is an Army report, so I am not quite sure why you used gender figures for the entire military service (17%). The data in your link indicates that women made up 14.6% of the Army in 2004. Using that figure, multiplying the sexes rate by the population in the military, you get (21.82*85.4%)+(5.54*14.6%)=19.44/100,000.
Slight quibble which actually strengthens your point, but the more interesting observation is that since 2001 the percentage of women in the Army has steadily decreased (from 15.7% in 2001 to 14.6% in 2004) This is contrary to the overall trend in other services.
If this trend continued into 2006, then the changing demographic composition of the Army might be a significant part of the uptick in the suicide rate. The Army is becoming more male after decades of the opposite trend.
PD – good catch, I didn’t notice that it was one service only…and yes to your trend idea…I’ll do a sensitivity measurement and post it up.
A.L.
While the suicide ratios for active military and the general populace may be statistically moot, I wonder whether the suicide ratio holds for those having been deployed and those w/o one or more deployments? And while the overall ratio may be comparable, what is truly alarming is the doubling of the number of active military suicides between 2002 and 2006.
I also note that the statistics cited for military suicides cover only ACTIVE military (incl. active Guard & Reserve)… which leads me to ask: given a reliance on these statistics to disprove a causual relationship between this war and suicide, since most soldiers and Marines suffering severe mental conditions that might lead to suicide or attempts are [medically] discharged or retired — exactly where are those statistics recorded?
dadmanly,
The report is an Army report, being released to the media for publication. How is it an example of liberal media bias if they publish it?
A.L.
“You don’t think that counts??”
Depends, I guess. Counts as what? A silly statement? Yeah, okay, I’ll give you that. Is one silly statement worth being the central focus of your post? Not so sure. Does it justify the headline you chose? Again, not so sure.
Is this all about Jeralyn, then? One person? One comment?
Why is the Army releasing this report? Possibly they/it believe it’s newsworthy. Possibly its just routine. Possibly they hope to make the impression that they are aware of and doing something to combat an increased suicide rate. In any case, I think your statisical/context gripe is with them.
I’m just wondering if the Isle of Conclusion is getting crowded due to people jumping onto it from more than one direction. It might be worth a peak on the rightward shore and see if there’s a gathering there, too.
Perhaps there WILL turn out to be a widespread use of this Army report by war critics to argue that the army is broken. I could be wrong. It wouldn’t be the first time. By my count it would be the 439,722nd time.
But I am suggesting that it might be better to wait till a few more than this Jeralyn actually make such an argument before “People are so gulliable.” headlines get posted. “Jeralyn is so gulliable (he/she/it accepts Army reports w/o doing his/her/its homework)” might have been more suitable to the present state of affairs.
–mark
By my count it would be the 439,722nd time.
I think your count is off, which, by your count, brings the total to 439,723.
LOL, Rob. Very good.
Interestingly, Rob’s figures are actually logically impossible. If mark’s count were off by one (too high) then he would be wrong in his count, which would bring the figure back up to the same number (which would make it right, which would invalidate it being wrong and make the count wrong again, and then the universe would implode). If his count were too low by one then he would still be off by two, since being wrong on the count would add one more.
My wife puts the count considerably higher. Coincidentally, her count is remarkable close to the number of minutes that have expired since our wedding day.
A hard (but good) day at work, and you guys just made it a good day.
I’ve talked a lot about wanting meaningful discussions here, but I also want – very much – for those folks having the discussions to appreciate each other the way mark, Rob, and Robin seem to…
…so thank you all. You made my day.
A.L.
I would think any sentient bloggers would realize the comparison is meaningless. The idea that Army or Pentagon numbers are anything but a tiny fraction of military suicides is insane. When does the Pentagon tell the truth? Ever? Pat Tillman ring a bell? Anything?
This number is just the ones they couldn’t hide, mislabel, or misrepresent. These guys must of done it in front of twenty witnesses at high noon during chow, after writing suicide on their forehead with permanent marker…puh-lease, our troops are fighting an armed occupation where they are required to kill people, even women and children. And all while expecting to be blown to shit at any moment. Ever notice most of the homeless are mentally ill war vets?
These guys need as much help as we can give, probably decades of therapy and emotional support. Billions of dollars anyway, and the sooner they can come home the better, obviously. Since there is no imminent threat to our country of any serious magnitude(though I admit the neo-cons have seriously increased the danger), more death and damage for the glory of Bush seems to be gross stupidity…
No, AL, we thank you for providing the means and opportunity. It’s very much appreciated.
–mark
Some people are gullibe because it fits their agenda.
SSQuirrel’s comment is pretty much beneath contempt.
There are indeed serious strains on soldiers deployed multiple times in some parts of Iraq, on some kinds of missions. I know they’re there — I’ve heard returning soldiers talk about it in some cases.
So far in this thread what I haven’t heard is any mention of the fact that the Army started to really concentrate on this issue with suicide prevention outreach (and training for all soldiers and leaders) a year or more ago.
Molon, re-read #16. It was mentioned.
Squirrel, you’re full of squirrel poo. There are a number of real studies that show that very few of the mentally ill homeless are war vets; we don’t have nearly enough wars to fill that roster.
And given that the military and government and powers that be pretty much lie about everything all the time, I’d be deeply concerned about the giant space lizards who are really running things in the world capitals. I hear they think of us as appetizers.
Want to try again with some homework and facts?
A.L.
The Washington Post/Associated press journalist Pauline Jelinek apparently has the same math skills that were revealed in the medical study I saw a while back, published by an international news distributor and eventually a San Diego newspaper I read. The medical study concluded that chocolate was overwhelmingly beneficial to your longevity and heart health —-the more you ate, the better your overall health.
However, the sample set for the study was about six or eight Germans. And they were serious. Many journalists apparently failed 3rd grade math, and never took another math course after that.
Yet, there is an underlying and yet sad issue raised by the suicide rates that perhaps has less to do with the military (especially considering that their suicide rates are lower than society at large), and more to do with a degradation of the sense of purpose in our lives and the general emotional indifference and apathy that our culture pressures us to accept.
For individuals susceptible to depression, depression can be triggered losing a sense of worth in our life, loss of purpose of our actions, loss of hope against the struggles in their life, of our planet.
Yet, if anything, becoming a man or woman of purpose by helping save others from tyranny’s viciousness, the brutality of chaos, and pain of violence may in fact provide the exact purpose and courage that the rest of our culture (and apparently that Washington Post journalist) would prefer we destroy. And that’s called becoming a leader, becoming a member of the US Military.
As we are reminded by this Washington Post journalist, apparently some people wish to destroy our sense of purpose and the crush our hope by providing bad information and encouraging general a blind disgust at people with purpose. Like the pot shots from the bench from the mathematically incompetent.
Perhaps the cause of this sad societal increase in suicide has nothing to do with joining the military (as that AP journalist who obviously failed 3rd grade math implies), but instead has more to do with too many weak individuals in positions of influence whose actions destroy truth and purpose. Which this journalist, Jelinek, is obviously a case in point.
Clearly, the statistics show that the military has a greater success at combating that cultural malaise. And this post’s corrections to the poor conclusions of Jalinek reminds us all that any journalist who fails to do his or her research at the expense of truth, deserves to lose his/her job.
But then again, Pauline Jelinek is probably spooning large bowls of fudge into her mouth as if it were oatmeal believing that there is overwhelming evidence that all forms of fudge are the newly discovered fountains of youth and that she won’t even gain a pound. Seriously, Pauline, better do that math homework you missed back in grade school.
What on earth has this to do with Jelenik’s math skills? What conclusions are you talking about? All she did was write a short news article based on an army report. All the figures come from the report…..which was released to the press in order that these very stories get published. Jelenik is just a mouthpiece here for the Army. There’s no independent reporting…not a paragraph that isn’t a paraphrase or a quote from the report…just a simple straightforward relay of the Army’s report. It probably took her all of 45 minutes to knock this one out. A more innocuous piece of “journalism” would be hard to find.
#29. True enough as you state. However, let’s face it. Methinks that particular “journalist” would have trouble following math of any kind. She didn’t go deeper because she COULDN”T go deeper. She is a journalist. Statistically, that pretty much puts her much to the left side of the fabled Bell curve.
Da Coyote,
I see. So your criticism amounts to this: she failed to do what you claim she is incapable of doing. Clever.
So a lack of context is no longer the fault of the reporter? I’m getting less impressed with journalism as a profession.
Exactly what context is missing in the WaPost article? The army has become concerned about a rise in suicide. They’re working to figure out why and instituting programs and measures to address their concerns. They issue a report to the press. The WaPost publishes a brief news (not analysis) article summarizing the report’s findings the day it is issued. Why are we even discussing this? Talk about looking to confirm your prejudices. It’d be difficult to find a better example of this phenomena than many of the comments in this thread. “Liberal media summarizes army report; refuses to give full social context of slight increase in army suicide rates in bald attempt to discredit war in Iraq.” “Stupid female journalist regurgitates statistics given her by Army and refuses to do indepth interviews with sociologists to make sure other stupid people don’t misunderstand Army report and come to conclusions that undermine Iraq war effort.”
mark – try this (made-up) news story out –
“Crime rate rises in Torrance. Property crimes in Torrance were up 6.5% in 2006, raising concerns among city leadership and citizens. The City Council is asking the police department to look into it’s practices and policies, while reminding citizens that the property and personal crimes rates in Torrance are substantially below the mean in Los Angeles County and among comparable cities.”
That’s both good press relations (bad Army for not including it in the report) and good journalism. You provide context to help people make sense of isolated facts.
Better?
A.L.
YOU KNOW OF COURSE THAT THE PRESENT DAY MEDIA IS NOT VERY BRIGHT. TAKE FOR
INSTANCE THE NEW YORK TIMES. THIS WAS ONCE A GREAT PAPER BUT THEN IT STARTED
TO LOOK FOR LIBERALS THAT WOULD PRINT THE NEWS THAT LIBERALS WOULD WANT TO
READ…..so they got plagerists, they got liars who made up stories, and they
got really stupid nutty types (just out of school) whom they could train to
write the news with the LIBERAL SLANT! TODAY THEY ARE REAPING THE REWARDS OF
THE NEW YORK TIMES POLICY FOR WRITING.
ON THIS STORY WE HAVE AN INDIVIDUAL THAT HAS DONE HIS HOMEWORK AND REALIZES
THAT THE FIGURES QUOTED ARE NOT RELIABLE. HE HAS GIVEN THIS NEWSPAPER THE
TRUE RESULTS AND WE ARE SURE IT DID NOT TAKE HIM TOO LONG TO COME UP WITH HIS
NEWER FIGURE.
ISN’T IT A SHAME THAT THE LIBERAL PAPERS ARE SO LETTING THE AMERICAN PEOPLE
DOWN SO BADLY. ISN’T IT A SHAME THAT THE LIBERAL PAPERS WILL DO ANYTHING AT
ALL TO MAKE OUR MILITARY LOOK LESS EFFECTIVE THAN WE KNOW THEY ARE…..!!
The Army’s press release seems to provide the proper context. Compare the ledes:
_Though its number of suicides increased in 2006, the Army’s suicide rate still is lower than the rate for the same age and gender group in the overall U.S. population, according to a report the service released today._
“Press Release”:http://www.defenselink.mil/news/newsarticle.aspx?id=47074
_Army soldiers committed suicide last year at the highest rate in 26 years, and more than a quarter did so while serving in Iraq and Afghanistan, according to a new military report._
WaPo
BTW/ Still looking for the actual report. Got report?
Heh, I think Mark, in an attempt to defend her, just whacked Jelenik harder than T did.
_”All she did was write a short news article based on an army report. All the figures come from the report…..which was released to the press in order that these very stories get published. Jelenik is just a mouthpiece here for the Army.”_
Nice. I guess she could have just saved everyone a lot of time and just linked the pdf of the report? Of course, then they give up editorial control, and well, that just can’t happen. And I doubt the Army generates reports primarily so the press can report on them. I’m guessing that they actually generate them to assist in policy making and decision making.
_”There’s no independent reporting…not a paragraph that isn’t a paraphrase or a quote from the report…just a simple straightforward relay of the Army’s report.”_
Ever heard of creative editing? Facts can still lie, if you arrange them just so.
_”It probably took her all of 45 minutes to knock this one out. A more innocuous piece of “journalism” would be hard to find.”_
Easy, yes I’m sure. Innocuous? Not so much.
It’s not like journalists don’t know anything. I’ve been playing with the “Wikipedia Scanner”:http://wikiscanner.virgil.gr/ and I’ve noticed that some reporters from The New York Times and The Washington Post spend a lot of time editing the “Doctor Who” entries. They sure do know a lot about Doctor Who.
douglas, don’t get me wrong here. I’m not trying to defend her..at least not in the way you are implying. I’m just suggesting it’s run-of-the-mill daily paper obligatory reporting: army issues report–here’s what it says. I’m not claiming she should win the pulitzer prize…i’m just saying that by covering an army report, neither she nor the WaPost are impugning the army or suggesting that it is broken or that there is some crisis….it’s just basic news of the day reporting.
Glen, how do you know reporters are editing the entries and not someone bored manager in say, the circulation or advertising department?
AL,
“while reminding citizens that the property and personal crimes rates in Torrance are substantially below the mean in Los Angeles County and among comparable cities.””
Sure. Absolutely. If that’s what the city council is doing…or in this case the army. Look, I’m not claiming it’s great journalism, deserving of a pulitzer. I’m just claiming that your post was both an overreaction to a run-of-the-mill report and a jumping to conclusions about both the motive behind the report and uses to which it might be put by war critiques.
Mark,
You do have good points about my comments, and thank you for pointing them out. I appreciate that. You are right on target in that technically there aren’t faulty calculations in her report. However, the real story here is she missed the far more important, bigger story–that simple mathematics reveals.
The Army has the guts to face the hard facts regarding the difficult and painful statistics of suicide within their organization. And they reported them to the public, as they should. It is deeply tragic, and we all have deep sorrow and compassion for the families who are facing a terrible loss.
Yet, like any great organization the Army never settles for “at least we are better than the rest of society” or “at least we didn’t increase suicide as much as the rest of you guys.” Instead they are attempting to understand the increase and hopefully reduce it even more than the deeply troubling (and higher) suicide rates we see in our society at large.
Tonight this “Army Suicide Rate” story appeared on national evening news, with the context that the Army was the cause of this and it implied, by disregarding context, that society’s average was much better. However, it’s mathematics that indicate that Army experience is likely not the primary causal variable in itself, because the suicide rate is HIGHER in the society at large. Apparently, if you join the Army the probabilities for suicide go down. But again, you have to understand causal variables to draw appropriate mathematical conclusions.
A basic understanding of math and doing a little homework would have revealed to her that there is a larger, more deeply troubling problem here.
The tragedy of suicide, the deeply painful result of a long struggle with depression, can never be easily summarized, the causes and struggles are complicated. The condition of depression can sometimes be very difficult to treat. We should always have compassion for individuals who face that struggle. At least the Army is reporting and diagnosing the tragic issue.
Meanwhile the media apparently chooses to disregard the larger even more tragic context that all of us should be evaluating why our culture is experiencing an increase in suicide. In other words, Mark, if you want to show compassion you will have a greater probability of finding a person in desperate need of your compassion in your own neighborhood, than if you were in the Army.
Because many journalists are apparently too numbers adverse to run some simple calcs, they have missed the opportunity to let us ALL know we have tragedies right under our nose, of which the Army seems to be the only one with courage to report and address directly and publicly. Gratefully, the Army doesn’t want to be just the like the rest of us. They want those numbers lower. They are taking care of their own. We should do the same.
#35 (Therese):
Please don’t shout. Also, putting lots of carriage returns into you posts makes them harder to read.
mark –
Good question.
“Here’s an edit done from an NYT IP”:http://en.wikipedia.org/w/index.php?title=Jennifer_8._Lee&diff=33044116&oldid=31623940 to an article on NYT reporter Jennifer 8 Lee. (“8” is actually her middle name.) The guy added:
I assumed that only a another reporter would bother to do that, to get back at that snotty bitch from the Metro section who thinks she’s such hot stuff. After all, it doesn’t say “Many circulation managers think she is very rude.”
This is the same NYT IP that was used to deface entries on “George Bush”:http://en.wikipedia.org/w/index.php?title=George_W._Bush&diff=prev&oldid=33031414 and “Condoleeza Rice.”:http://en.wikipedia.org/w/index.php?title=Condoleezza_Rice&diff=prev&oldid=15725782 On the positive side, here’s his contribution to “Brylcreem.”:http://en.wikipedia.org/w/index.php?title=Brylcreem&diff=prev&oldid=91251688
But what I really don’t understand is why he changed “Larry Bird’s”:http://en.wikipedia.org/w/index.php?title=Larry_Bird&diff=prev&oldid=111672440 weight from 220 lbs to 2200. That was it – add a zero, save and exit.
I suppose you have a good explanation for that.
Most reporters are horribly ignorant.
Zero statistical understanding (Just One Minute blog took apart another NYT report “here”:http://justoneminute.typepad.com/main/2007/08/they-cant-mean-.html
Look to their training: Thought-indoctrination, “the narrative was correct but the facts were wrong,” “fake but accurate” and so on.
A sports reporter on a regular beat on say, the local football team (College or Pro) would be kept honest by readers who know at least as much as he would about football. But outside sports that check on reporter ignorance has been absent up until now and blogs.
Even so … some reporters still act stupid. One asked Lovey Smith where he was “when he learned he was going to the Superbowl?”
IF THOSE questions get asked it’s too much IMHO to expect reporters to be numerate. Yes of course they spend most of their time editing Wikipedia and other stuff.
Jeralyn also says:
bq. Experts agree the war in Iraq cannot be won militarily. So why are the troops still there? Let’s stop the funding and bring them home now.
Oh, really? We won the WAR in 3 or 6 weeks, depends on which source you wish to cite. The aftermath has been a tad more of a problem but to say it is unwinnable is pure BS. We could win the Civil War in 3 weeks too but that would give all the gentle souls like Jeralyn the VAPORS. Maybe those experts she quotes are just the voices in her head?
If you read the words of the Iraqi bloggers and the reports from Totten and Yon, we can win the aftermath of the war by doing a few simple things – provide electricity, provide security, provide opportunity. The current surge strategy is laying the GROUNDWORK and groundwork only for the first item in the coming year. The rest may take some more time. BUT the short-attention-span US public cannot concentrate thru a 30 second sound bite much less a protracted military and civil works program. It begins to sound like the tantrum of a 4 y.o. – the sounds coming from the Left side of the aisle – I WANT THEM OUT NOW!!!!!!!
I know a LtC in the USA who was there early on and is back for his 2nd tour. He was in charge of special projects which means helping build and restore infrastructure. The only areas that had 100% coverage and service for electricity before Saddam fell were those of his special enclaves. The rest of the country had to have the infrastructure BUILT from the ground up – i.e. (for you simple minds) there wasn’t any BEFORE we went in, much less any to repair. Get a clue folks.
The rates of suicide in the armed forces are lower than the general population because they are highly motivated and well balanced individuals. But even highly motivated and well balanced people have issues sometimes that they cannot resolve. It does not mean that there is some HUGE problem that needs to be addressed but that there is an issue worth reporting and investigating. The USA is doing just that.
SHEESH – don’t publish this and they are going to be found guilty of hiding something. Publish it and they are guilty of being heartless taskmasters of the poor, hapless, and weakminded US soldier who cannot be anything but a pawn of the EVIL and WICKED BUSHHALLIBURTONROVIAN blood for oil cabal. Damned if they do and damned if they don’t.
/okay = rant off and I will disappear back into the troll works for a time. Just that stuff like this gets me worked up.
([knock-knock] anyone home Misha?) (waves good-bye)
_9.1 per 100,000 in 2001 and is 17.3 per 100,000 today._
Double your pleasure,
Double your fun,
*Somebody* here is Doublemint Dumb…
8)
It doesn’t particularly make sense to compare army suicide rates with the general population. The army really is a select group. Lots of the people who tend toward suicide can’t get into the army.
imagine this overweight depressed guy, George. He tries to enlist in the army. The third morning he’s there, he doesn’t get out of bed. His sergeant comes to him and asks in a deceptively mild voice, “George, what seems to be the problem?” And George says, “I just don’t feel like it’s worth getting up. I’m going to lay here in bed today and I’ll join in with you guys tomorrow.” Does George seem like a good candidate to finish training? If he fails, and if at some later time he commits suicide, he’ll be a civilian committing suicide and not a soldier.
If suicide rates in the army approach civilian levels it’s a sign that something is very wrong. For comparison, if our soldiers were generally as sedentary and out of shape as the civilian population, it would mean something was _wrong_.
Still, the numbers are low enough that they aren’t a big issue in themselves and they wouldn’t be even if they were a lot worse. Say the suicide rate went up by 100 soldiers a year, or even 150. I wouldn’t be surprised if you lose more than that to traffic accidents and you could save more lives by driving courses and enforced traffic regulations. To the extent it means much it’s like that canary in the coalmine. It’s a measurable statistic that might have implications for things that are harder to measure. I don’t know how the army will use it, though. Civilians mostly don’t have the context to draw meaningful conclusions. We have to depend on what the military tells us, and we know they lie a lot.
But I don’t think there should be a lot of controversy here. It isn’t controversial that long deployments and repeated deployments are bad for the army. That’s unarguable. Where there’s a question is whether the results are worth this wear-and-tear on the army. That question *is* debatable. My own view is that if we intend a counterinsurgency victory, we can expect it to take five to fifteen years starting last january, and in that case we’d better pace ourselves. If we can’t win with troop levels we can maintain for ten years, we’d do better to disengage. Sure, the troops put up with a lot more in WWII, but for us WWII lasted around three years and nine months. If this was WWII it would have been over sometime around last january.
Can we fight this as a sustainable COIN war for ten years? We can’t afford to burn out the experienced veterans too fast.
SSquirrel: Your last post was a drive-by. Drive-bys are frowned on, though we cut regulars a little more slack.
Care to prove you can actually contribute substance? I strongly recommend it.
We need a leader capable of helping people make the jump back to reality from the Isle of Conclusion. What better way to fight the demons of ignorance?
Milo Giuliani for president!!
J Thomas:
It makes a very poor canary in the coal mine. It makes a very good mine for misleading statistics.
Suicide statistics, and their variations from year to year, are somewhat meaningful when you’re dealing with a nation of many millions of people. When you’re dealing with a couple million people, the numbers fluctuate so much from year to year that analysis is meaningless. You cannot compare changes in the suicide rate of the Armed Forces (about 2,000,000 including reserves) with changes in the suicide rate of 300,000,000 civilians.
When the military has a low rate or suicides for a given year, which it often does, it is possible for a single person to drive the suicide rate up a full point or even more. In small countries with low suicide rates, it’s possible for a single suicide to DOUBLE the national suicide rate.
The smaller the sample, the less stable the statistics, and the less you can deduce from them. The suicide rate in San Diego bumped up in 1997, because 38 UFO cultists staged a mass suicide. That doesn’t tell you much about the quality of life in the rest of San Diego.
Here are a couple of facts about suicide:
– It is not proven that PTSD causes people to commit suicide. Some researchers think so, but others don’t and it is not established scientific fact. The myth of abnormally high suicide rates among Vietnam veterans has been debunked. Vietnam veterans actually have lower suicide rates than the general population in some studies.
– The major cause of suicide seems to be depressive disorders, which put some people at permanent risk for suicide. Your claim that such people would not succeed in the military is wrong – such people have succeeded at almost everything. Environmental factors, including PTSD, could prompt a person with a depressive disorder to commit suicide, but are unlikely to cause a person without such a disorder to commit suicide. Some people with depressive disorders kill themselves for no apparent reason at all, surrounded by wealth and friends.
Glen,
I just cannot agree with anything you have said.
First off, J Thomas is absolutely correct that there is a selection bias toward healthier – both physically and mentally – in the military. Therefore it makes no sense to compare military suicide rates to non-military suicide rates. These are very different populations.
A significant trend of increased military suicides should be taken at face value. It means something and it is irrelevant what is happening – suicidewise – in Compton.
“When you’re dealing with a couple million people, the numbers fluctuate so much from year to year that analysis is meaningless.”
Wrong. An N of a couple million is sufficient for reliable statistical analysis and meaningful trending. The statistical methodologies for testing significance can take population (Or sample) size into account and an adjusted error is factored in. In my “day job” I work in the actuarial field for a major healthcare insurance plan. If what you said were to be true we would be unable to perform our jobs.
And we don’t look at our Commercial HMO rates, compared them to Medicare rates and decide there’s nothing to worry about because the HMO is still lower than Medicare…….nice try AL, et al 🙂
“…..it is possible for a single person to drive the suicide rate up a full point or even more.”
Only a dunce looks at % change when decision making if that is what you’re talking about. It is meaningless. From 1 event to 2 events is a 100% increase. Sounds like a big jump, but so what?
If you were refering to the potential for a single event to drive the rate up 1/1,000 or per 100,000 or whatever in a population of a couple million or so, then you are mathematically incorrect.
I don’t know the raw volume of military suicides, but based on the rate it has to be around a couple hundred per year; more than enough to test for significance on rate changes.
Additionally, I’d like to see some finer cuts of the data before any conclusions as to causality are drawn (e.g. what is the rate among those having served combat tours as opposed to supply/support? How does number of combat tours interact with suicide rate?, How do married troops compare to unmarried? etc).
“…Only a dunce looks at change….”
should read % change
Avedis, there were 99 suicides in the military in 2006. I’d be very interested in your professional take on fluctuations in a population that small.
Absolutely, when you’re talking tens of thousands of events you move into a level of statistical certainty. But my – admittedly college-level – stats experience is that once you’re below a certain level, the complexity of prediction goes up dramatically and the certainty of prediction goes down dramatically.
A.L.
avedis –
You’re right; I meant what the “dunce” said – a percentage increase of a full percentage point or more. For example, the roughly 12% percent increase in military suicide from 2005 to 2006 (from 87 to 99) was caused by 12 people.
You’ll find plenty of such dunces in the flood of news stories about this report. “Reuters”:http://www.alertnet.org/thenews/newsdesk/N19303337.htm for example:
The operative words in this post are “gullible” and “military welfare”.
A casual reader might at first be drawn into the notion that AL thinks the media are gullible and don’t really care about military welfare (as evidenced by their alleged misrepresentation, or misinterpretation, of suicide data based on an anti-military bias), while he himself is not gullible and is genuinely concerned for troop welfare, because he has “done some real digging” and has come up with a different take, thereby defending the honor of said troops while basking in the glow of his computer monitor.
What a load of hooey. I’ve rarely seen so much made of so little, and for no real purpose.
For example, even a casual reading of the news will produce many serious affronts to the troops welfare, both here (PTSD) and in Iraq (see below) that are enough to make any normal, rational person livid with anger. Yet these seem to escape the notice of AL, who says he has a concern for folks in uniform.
I think by his choice of topics and arguments (weak as they are) he demonstrates rather convincingly that he is more concerned with the perception of the war by the US public (the real topic of this thread, wrapped in a “concern for the troops”) than anyone’s welfare.
So here’s what gets me angry about the treatment of our soldiers over there, not this hooey:
http://www.cbsnews.com/stories/2004/08/17/eveningnews/main636644.shtml
New Fuel To Halliburton Fraud Fire
(CBS) When it comes to logistical help for U.S. troops in Iraq, Halliburton is the biggest game in town. Under a wartime contract that’s $7 billion and growing, it’s serving the needs of 200,000 troops.
But the Houston-based conglomerate once headed by Vice President Dick Cheney is neck-deep in allegations of waste and fraud involving millions of taxpayer dollars, reports CBS News Correspondent Sharyl Attkisson.
The U.S. Army is threatening to partially withhold payments to Halliburton for the logistical support the company provides for troops in Iraq. The reason: allegations of millions of dollars in over-charges for food, shelter and services.
“There was no regard for spending limits,” says former employee Marie DeYoung.
Some of the most compelling accusations come from people like DeYoung, who worked for Halliburton subsidiary KBR.
She recently told Congress that while troops rough it in tents, hundreds of preferred Halliburton KBR employees reside in five-star hotels like the Kempinski in Kuwait with fruit baskets and pressed laundry delivered daily.
“It costs $110 to house one KBR employee per day at the Kempinski, while it costs the Army $1.39 per day to bunk a soldier in a leased tent,” DeYoung said.
“The military requested that Halliburton move into tents, but Halliburton refused.”
Documents obtained by CBS News show an auditor repeatedly flagged improper fees for soldiers’ laundry. At one site, taxpayers reportedly paid $100 for each 15-pound load of wash – $1 million a month in overcharges.
Halliburton insists it doesn’t waste money, it saves it. But overcharging is the subject of one federal investigation and there are separate probes for alleged bribery and kickbacks.
Nobody from Halliburton agreed to an interview, but officials have said the criticisms are politically motivated in an election year. Halliburton also “questions the factual nature” of many assertions from the ex-employees, but is looking into them “because we take all allegations seriously.”
With the tab for the war in Iraq now topping $150 billion, all the investigations could help determine whether some Halliburton employees paid by taxpayers to make life better for soldiers are, instead, putting themselves first.
J Thomas: I agree i was thinking of this statistic the other day, and agree that we’re looking at a select population. Of course, 99 is too small to look at accurately. It would be better to look at other symptoms/problems that carry a wider sample set.
So things like PTSD, severe anxiety, severe depression etc would make for a much better (more accurate) report than merely suicide. Maybe this report exists, maybe it doesn’t, however the reports that I have heard are not good.
For example: A recent study in JAMA finds that spouses of Iraqi soldies are more likely to “abuse”:http://www.npr.org/templates/story/story.php?storyId=12385667.
bq. The study published in the Aug. 1, 2007, issue of JAMA shows that some military spouses don’t do as well as Erving. The Army commissioned the study, and it examined all confirmed reports of abuse and neglect collected by the Army from Sept. 11, 2001, through the end of 2004… Martin says that overall, rates of maltreatment of children in military families is low. But during deployment, rates jump, outstripping civilian abuse rates.
bq. “We found that their rate of maltreatment increased three times… [Neglect] was four times higher… during the deployment than non-deployment. And physical abuse was almost twice as high.”… The researchers identified about 3,300 confirmed cases of abuse or neglect in 1,800 families. Two-thirds of the incidents were moderate to severe — the vast majority were neglect.
So better study, also better reporting. NPR had an interview with the researcher for an hour. Definately gave the feeling that stress on the families was causing this problem (that is highly unusual) for the military. And that the military is responding by hiring hundreds of extra counselors to better monitor emotional stress in families of deployed soldiers.
Still, study is showing that there is an immense amount of stress on the military and military famillies.
All,
99 out of a population the size of the US military (which apparently – based on the raw nuber of suicides compared to rate – has fewer personnel than in my day) is NOT too small a number to perform statistical analysis for significant change.
The statistical methodolgy one would use would probably be the Poisson – this depending on the nature of the distribution as well as some other aspects of the data.
Poisson is used where the number of events under scrutiny is relativels small and not normally distributed and its calculations take into account the random flux inherent in small number data.
When you think about it, there are events – you can probably identify some in your own life – that generally only occur with a low level of frequency; say once or twice in a given time frame or spacial frame. If, all of the sudden, you observed the event occurring ten times in that same temporal/spacial frame you would instinctively sense that something was going on; something had changed. Poisson deals, mathematically, with these sorts of situations.
Example: My household buys grain for horsefeed. On average out of every 200 bags purchased 1 contains meal worms. Then I switch stores. I observe that the grain I purchase at the new store has a meal worm infestation rate of 6 out of 200 bags. Is there a significant difference in the quality of the grain sold by the 2 stores?
Some of you here would say the numbers are too low to tell. Yet, I’d wager that any of you would feel that the new store was ripping you off, inferior, etc. You’d probably go back to the original store.
But this is exactly the type of situation where poisson distributions can – and do – test for significance.
Look this methodology up…..
Glen, “You’ll find plenty of such dunces in the flood of news stories about this report. Reuters for example…”
Yes. The media has consistently proven itself utterly unable to understand or convey data in a scientific manner.
Sometimes it is just painful for me to read.
This is especially true around election times when the abuse and misuse of “confidence intervals” becomes rampant. The way the media – and even the “pundits” – talk about these things….they might as well just wave a flag bearing the words, “I am a Dunce”…….
Does the military screen recruits for any history of mental illness or depression? If it does, it would explain why the suicide rate in the armed forces is lower than the population at large, for much the same reason that test scores in private schools are better than public schools. If the military is already rejecting the people most likely to commit suicide before they can enlist, it shouldn’t be surprising that the suicide rate for the population as a whole is higher.
Glen, going way back to #44, since you asked…I’m guessing either typo or mischief. I did notice that your original “reporters” at NYT has become one person you suspect of being a reporter.
#60: SS: Yes, they screen for a lot of things. Having never been in the armed services, my specific knowledge is limited. However, I know they screen mental health, intelligence, criminal records & background history, health & hygiene, as well as a personalities aptitude to survive bootcamp & extended military service.
One problem the military (more specifcially, army) has had lately is the need for growing recruitment (Actually, AL, it would be interesting to see if this number grew it all due to recruitment changes). This has led to decrease standards in health, age, intelligence and aptitude requirements. It used to be that boot camps were designed to weed out those who couldn’t cut it. Instead, there is alot of pressure to hang on to every potential recruite, wether they are making the grade or not (these candidates can be retrained if they fail completely).
It also wouldn’t shock me if areas of the country with higher service records also have lower rates of suicide… generally due to extended familial networks through church, close-knit families and smaller town environments. For these reasons, comparing the military suicide number to the general US number may be misleading. Instead, controlling for income, religous affiliation and home environment might give more accurate results (But I’m just making guesstimations here)
Alchemist;
Yep, what we want is definitely an unstressful war. Just like … refresh my memory. Dang, when WAS that stress-free war we had?
I guess the only stress-free option is pacifism. We’ll just tell all those nasty stressor-type people to leave us the hell alone.
Ahhh … I feel my brain relaxing and going limp already …
Sorry, Alchemist — not actually picking on you. Just sayin’.
BTW, just to give a bit of feel for these numbers: a “17” per 100,000 is = 1.7% of 1%. And I noticed in one of the tables that about 8% of the suicides were classified as PTSD cases, which is 2/25.
Brian: You’re right, every war is stressful. However, early studies have shown that serious PTSD problems exist in higher numbers than have ever been seen before. This could come from a number of different causes that are unique to this war (or this time in history), including:
1)Better understanding of PTSD by Pychiatrists
2)Soldiers increasing awareness of PTSD
3)The unique type of guerilla warfare found in Iraq
4)Mild – Heavy Braindamage from IED’s occurring alongside High Stress events
Or a number of other issues not yet considered.
Although suicide is not a major factor, it may be an indicator of other problems, and my understanding is that the miliary is investigating that possibility. Kudos to them.
During this war, the Military has done an exemplary job of healing physical injuries. What remains to be seen (and we won’t know for sure until 5-10 years after we leave) is how succesfully the millitary will handle psychological injuries. Already several Iraqi veterans with severe PTSD have had disability checks canceled because the military claims they had “pre-existing conditions”. Additionally, Veterans Hospitals have DECREASED the funding they have asked for, despite the fact that an increase in physical and mental damages are expected over the next 16-20 months.
Salon has done alot of reporting with the help of military families, and spoke of facilities like walter reed well before the story broke nationally. Again, military appears (at this moment) to be working hard to solve the problem, and hopefully they will. However, it’s something to consider when analyzing the long term ability/stability of our forces in Iraq.
Alchemist;
Yeah, the everlasting bureau-pressure to define problems out of existence by fiddling an application form. Blech. There will always be a few they can point to who try to ride any such programs like a gravy train, of course, but that can’t be allowed to become a justification for denial and neglect.
Especially if this is going to be the Long War after-care and rehab is hugely important. My inclination would be to partly take things out of the hands of the “professionals” by constituting a well-funded self-help association with access to both purpose-built therapy facilities and wider resources. There would, IMO, be more motivation to keep eyes on the ball rather than building admin and lobbying empires.
avedis – sorry not to have weighed in earlier, I’m finishing a roof and getting ready to leave for NYC (and as a sidenote, if you don’t believe in global warming, try standing on a roof in Los Angeles in August for an hour) – of course you could use a Poisson analysis – hence the “law of small numbers”, which AFAIK is another term for using one.
The interesting question is what it would say…
Now this is a sidebar, because my claim is essentially a ‘political’ one based on raw numbers – it’s what I call “incident or problem” at work. Some things are “incidents” – which means they are in the queue for someone deal with and get some share of individual attention – and somethings are “problems” and are part of the team focus and get a set amount of attention every day.
“From my piece on risk”:http://www.windsofchange.net/archives/003110.php
One way we decide where something – like suicide in the military or mine deaths – on the incident/problem scale is to create some kind of comparisons.
Mine deaths increasing, mine deaths/ton increasing, and mine deaths/worker increasing? Clear call that you have a problem. Note that Dave’s #’s in the link here show declines (I’ll discuss in the thread above).
Military suicide increasing and statistically higher across and per population than the civilian norms? You have a problem. Military suicide rates increasing, but at or below civilian rates – still a problem but a less urgent one. One question is whether they are increasing year-to-year (a trend) or we have a spike. I’d love to get pointed to the dataset.
The issue of course is that you only have room on the plate for so many problems, so you have to make decisions all the time as which will be treated as ‘incidents’ (local, incident-specific response) and which will be treated as ‘problems’
Given that you’re in the biz, it’d be fascinating to go through an analysis with this data – I’d be happy to make it a post if you’d include the steps for folks to follow and some explanation…
A.L.
“Given that you’re in the biz, it’d be fascinating to go through an analysis with this data – I’d be happy to make it a post if you’d include the steps for folks to follow and some explanation…”
I’d be happy to, but only agree to if we have sufficient data to do it right. And, BTW, I have no preconceived notions of what the results would be.
Also, I generally agree with your last in theory, altough I disagree completely concerning your assertion that military suicides rates only become an action item if they somehow compare unfavorably to civilian rates. I re-assert that the populations are totally different and cannot be compared. Each should be examined separately.
Whatsmore, your differentiating between incident and problem is appreciated. Believe me, there are members of certain Boards of Directors that have trouble understanding this concept and much resources are spent unraveling their misunderstandings.
That being said, one does have to establish dashboard metrics that serve – as someone up-thread put it – as canaries in the mineshaft. These metrics serve to identify emerging problems out of the sea of incidents. I would think that the suicide rate among our armed forces personnel would appropriately be one such metric. A better one – maybe more of an umbrella metric of which suicide would be a drill down – would be rate of diagnosis of serious mental illness.
A quick down and dirty way to look at this issue would be to obtain the armed forces suicide rates, by year, for the last 30 years and calculate the mean and standard deviation. We could then do a visual – like a control chart – setting the upper and lower limits at, say, 3 sigma (+3 upper bound and – 3 lower bounds). Then plot the thirty points in a series corresponding to each year. Where there are spikes through either the upper or lower bounds we could say that something significant probably occurred; particulary if the spikes consisted of more than one point and/or were part of a general trend. We could then investigate what happened during that time frame in military affairs.
I still like using poisson or a variant thereof.
Military suicide rates aren’t a particularly good measure for morale problems etc because the numbers are small so it takes a large increase to be sure it’s real.
But diagnosed PTSD is a *bad* measure. First, we can hope that diagnoses get better with time. That means data from more recent years will be systematicly more precise than from earlier years.
Also, there’s more emphasis on early detection and detection of milder cases. It’s reasonable to think that early treatment is more effective. So the numbers would go up even if the problem is no worse — as part of improved treatment that might make things better.
Also, there’s an effort to put less of a stigma on admitting symptoms. It used to be that soldiers were supposed to tough it out and not admit their symptoms. For a short war that can delay lots of the problems until after the war is won. But of course, having to hide the problems tends to make them worse. When soldiers get permission to admit mild symptoms and get some help with them while still performing their duties, recorded rates will go up even while disabling disease goes down.
Also, someone on a different WOC thread pointed out that soldiers now get *benefits* from admitting to PTSD. It isn’t something to do while you want to continue a military career — of course it doesn’t look good and might interfere with promotions etc — but when you’re leaving the military why not get all the bennies you can get? In the past these benefits were smaller and also there was more of a stigma about lying about being disabled.
So the situation could actually be getting better even while it looks worse. Or it could be getting worse and not look as bad as it is.
Suicides suffer from random fluctuation but there’s less reason to expect them to be reported falsely. Probably there won’t be many murders reported as suicides, and if they are they maybe won’t change much, and if they do they might be something we’d want to count. In wartime there’s more chance to report suicides as enemy action if someone wants to do that, but more likely they’ll call it suicide. Technicly it’s suicide when somebody jumps on a grenade to save his buddies but I don’t think they count it that way.
It’s a pretty unambiguous number, without a lot of interpretation. They don’t change the diagnosis from year to year and they don’t start finding mild cases.
It’s hard to get accurate data, and when you do get it then it’s hard to be sure what it means. Military suicide rates are at least less ambiguous than most military data. It’s still hard to be sure what it means.
Even a military study can be political.
It is interesting that the report is being pushed at the very moment that General Casey is saying that the army can’t take it anymore.
The point being missed, of course, is the military is not comparable to society as a whole.
Until fairly recently, the military did not take folks with a history of mental illness or those under treatment for various mental illnesses. So, comparing military suicide rates to those of the public as a whole is dishonest.
Additionally, the military screens out those who may have drug or substance abuse problems.
In short, the military is supposed to be probably the healthiest segment of our society. If we begin to see a higher suicide rate–that indicates a problem somewhere.
“Until fairly recently, the military did not take folks with a history of mental illness or those under treatment for various mental illnesses.”
In males, the onset of Paranoid Schizophrenia is 16-25. http://www.esquizo.com/schizophrenia/
If one is recruting 18 year olds…then the military will certainly end up with some percentage of paranoid schizophrenics that were asymptomatic at the point of recruitment.
1 in 10 people who suffer from schizophrenia end up committing suicide.
http://www.world-schizophrenia.org/publications/23-suicide.html
Soldier’s Dad: Afraid you’re still missing the point.
Do mentally ill get into the military? Yes, they do. However–and it’s a big however–the military (until fairly recently) has rejected anyone with a documented mental illness past. Thus, the military (until rather recently) would have seen far fewer folks with mental illness than society as a whole.
Additionally, as I noted, the military also screens for other contributory factors related to suicide: drug and substance abuse.
Yes, Clearly, the statistics show that the military has a greater success at combating that cultural malaise. And this post’s corrections to the poor conclusions of Jalinek reminds us all that any journalist who fails to do his or her research at the expense of truth, deserves to lose his/her job.
Norming the statistics by sex to compare military suicide rates with civilian suicide rates is like comparing the effects of wearing fluorescent socks on SAT exam scores to those who don’t wear them.
In normal everyday civilian society, women can fufill almost all the roles men fill, from construction worker to police officer to fire fighter to doctor to butcher, baker, or candlestick maker, so norming the statistics based on gender makes perfect sense.
Not so in the military. You have to take into account the number of roles women can fufill (which does not include combat roles) to the number of roles men can fufill. You do that and you’ll achieve 2 things:
1) Your statistics will actually be valid. Putting them in a spreadsheet does not validity achieve.
2) You’ll find your numbers come out a lot different.