What Occupation Has the Highest Homicide Rate?

(Or at least had it in 1990-92; my quick search couldn't find more recent data.) We're talking here about generally accepted Department of Labor job categories, which are limited to legal occupations, are nationwide, and apparently exclude the military.

The answer is here; the homicide rate for this occupation was 30 times the average overall occupational homicide rate. Of course, none of these approach the nearly 2% yearly homicide rate racked up by this perfectly legal job.

That's funny! Well not that, but I was just talking to my wife the other day about reading statistics years ago about highest-casualty jobs, and that I remembered taxi drivers were high, and so were store clerks. And then I pointed out that they all must have been ignoring US presidents.

Good to see those numbers again hehe.
1.8.2007 2:33pm
Speaking the Obvious:
I believe the homocide rate among dictators is also relatively high. I wonder what the connection might be?
1.8.2007 2:57pm
DustyR (mail) (www):
Where does the 2% come from?

I think it is closer to 0.92% if I did my math correctly. Lincoln, Garfield, McKinley and Kennedy. That's four 50% years (I didn't check to see if one died and the successor took the oath the next year but I doubt I am incorrect.) The rest were 0% years. That's 200% divided by 217(2006-1789) years, or an average annual homicide rate (not yearly homicide rate) of 0.92%.
1.8.2007 3:18pm
Eugene Volokh (www):
Four homicides over 217 years of service = roughly 2%.
1.8.2007 3:20pm
Ahmed (mail):
Dusty, I don't think it's relevant that only half the presidents in an assassination year died. What's relevant is that over 200 years 4 have been murdered on the job. (and eight died total). The way you are working the problem would be difficult to extrapolate to taxi-drivers (who knows how many were replaced).

If you would rather base your statistics on quantity of presidents, the numbers would of course be even more drastic, as nearly ten percent of presidents have been murdered on the job (and eight out of forty-three died on the job, but these are old folks so perhaps that's not as significant).
1.8.2007 3:33pm
These stats for the occupations of homicide victims are interesting, but wouldn't you like to know the occupations of the perps (among those who are classifiably employed)?
1.8.2007 3:38pm
chrismn (mail):
This is actually relevant to the "chickenhawk" argument: Old men sending young men to die while safe themselves. My guess is that if there were a market for "dying of homicide" on Tradesports for both President Bush (or Chimpy Bushitler depending on your politics) and a randomly chosen US serviceman in Iraq, or perhaps even a randomly chosen US Marine in Iraq, the market prices would be close.
1.8.2007 4:16pm
KactionJ (mail):
Of Course, with only one person holding the presidency at any given time there is the slight problem of ample size...
1.8.2007 5:11pm
KactionJ (mail):
Umm, sample size that is
1.8.2007 5:11pm
KactionJ (mail):

Of Course, with only one person holding the presidency at any given time there is the slight problem of ample size...Umm, sample size that is

William Howard Taft to the rescue!
1.8.2007 5:59pm
Oh, I dunno. Taft was of pretty ample size. Didn't he get stuck in the bathtub?

I was hoping the occupation with the highest homicide rates would be a quirky surprise, but those rankings pretty much square with my (everyone's?) intuition.

As for the president: I'd be rich if I had a nickel for every poster on dailykos who hinted someone should do Bush in. He doesn't, you know, respect the rule of law.
1.8.2007 6:04pm
matthew (mail):
I'm not statistician, I'm actually notoriously bad at this kinda stuff. BUT - - I don't know what the "Rate" column means--- So, it says

Taxicab services had the highest rate of work-related homicide during the 3-year period 1990--92 (41.4/100,000). This rate was nearly 60 times the national average rate of work-related homicides (0.70/100,000)

[ 41.4/100,000 ] - that's a cab driver's chances of dying in 3 years? or in any given year?

This later section makes me think "rate (per 100,000)" means annual rate (or, 'in any given year')...

When detailed occupations were analyzed for 1990--92 (Table 9), ... Compared with previously published data for the 7-year period 1983--89, these data indicate that rates increased more than two and a half times for sales counter clerks and nearly two times for motor vehicle and boat sales workers and sales workers in other commodities. . .

SO, assuming that's right, and we are saying "what are a random worker's chances of dying in any particular year?"

and I think maybe they calculated it as
[(d / yrs) / TotalPop ]

d= number of homicide deaths
yrs = duration of study
total pop = total population in that line of work over the subject period

so, something like this for the Presidents:
[(4/217) / 43]
[.018433 / 43]


1,843 (per 100,000)

Is that right? Suddenly 7-11 or the Kwik-E-Mart doesn't look like such a bad deal.

BUT, the study only goes back to 1980 data, so the rate by industry is something like this:

0 (per 100,000).

And what about Zachary Taylor? I suppose his death certificate doesn't say "homicide" so he wouldn't be a positive case on the study. But still.
Also. I wonder if the labor statistics would count someone who went to a different company as an additional worker? Should we cound Grover Cleveland twice? Because maybe I should call the number 42....
1.8.2007 6:26pm
matthew (mail):
1,843 (per 100,000)

Which is pretty much the same as EV's
nearly 2% yearly homicide rate
1.8.2007 6:32pm
Houston Lawyer:
What? No mention of postal workers even after the opening mail query posted below?
1.8.2007 6:48pm
DustyR (mail) (www):
Ahmed, the stat for presidents is good in itself as are the others in that study, even for comparison with each other, but to compare the former with the latter ones is not. Heck, the study, done by the experts, dares not even calc the rate for barbers and for good reason.

My earlier offering was a joke and if I thought Eugene was serious, I would have just invoked Twain. Besides, Lincoln and Garfield were definitely not "workplace homicides" while McKinley's and Kennedy's are borderline and debatable.
1.8.2007 7:45pm
John T (mail):
We thus see why taxicab drivers are incredibly leery of going into bad neighborhoods or picking up "suspicious" passengers. Not that it isn't absolutely terrible for law-abiding folks who live in such neighborhoods or look suspicious or whatever.
1.9.2007 12:50pm
markm (mail):

Besides, Lincoln and Garfield were definitely not "workplace homicides" while McKinley's and Kennedy's are borderline and debatable.

However, they were "work related". You can't get much more work related than being shot by someone who is upset at how you did your job. I'd also argue that for Presidents (and many lesser politicians), since they really can't go off-duty anywhere, anywhere they go becomes their "workplace" for the moment.
1.9.2007 1:43pm