That’s rather like saying people named “Dave” are the most murderous, once you discount all the murders committed by people not named Dave.
Ah, but how many taxi drivers and chauffeurs have gone on murder rampages (and I don’t mean necessarily using the vehicle as a murder weapon)?
Few, I expect, but among people killed in the workplace under any circumstances, taxi drivers beat postal workers by a mile. The OP is biasing his results by picking the sample he wants. He even interprets it favourably without justification, i.e. there’s no reason to assume all of the three to four murders annually among postal workers are the result of someone “going postal”, i.e. on a wild killing spree. Sometimes it’s just Carrier A clubbing Carrier B because B was sleeping with A’s wife.
I’m kinda curious how many postal workers get fatally hit by cars while on their route. It wouldn’t surprise me if the number rivalled or exceeded homicides.
I understand what Cecil is saying about the low numbers problem, but the OP is right to question the value of the studies cited for the question at hand. What we needed was a study of worker-on-worker/employee violence. Including law enforcement officers whose job puts them at risk, and victims of murders that were “incidential” the the primary crime - robbing a hack or store clerk - really skews what data there is.
Reminds me of a newspaper article several years back where Wichita’s preeminent local historian was asked to compare the town’s cattle-town-days murders with the present. He pointed out that there were four murders in the first few years of the town’s history, compared with today, where we typically have about fifty murders per year. Well, he overlooked the difference in population (twelve to a couple of thousand in the beginning, 300,000 today) and thus the huge difference in murder rate. (I know from my own research that he also missed a murder occuring a few months before the first local newspaper began publication, but that’s another matter.)
Course, we’ve known since Melville’s Bartleby the Scrivner that working at a post office will mess with one’s mind. 
Looking back at the original question, it doesn’t ask about murder or deaths. It asks about violence by coworkers.
While you might be able to get some idea if you check injury statistics (OSHA for non-governmental employers, the governmental agencies themselves for governmental agencies) you might get a clue – assuming that you took the time to look deep enough to judge whether the injuries were in fact on-purpose, and not accidents.
And that doesn’t even count the number of times that violence insured, and wasn’t recorded as an injury.
Workplace violence does’t just include murder. It’s a vague concept, but I would suggest that any act that would qualify as an assault or battery, whether or not charges are actually filed, and whether or not injuries are reported, qualifies as workplace violence.
And as someone else pointed out, just because it’s between people who are coworkers doesn’t mean it’s work related. If you punch your coworker in the face because you caught him cheating with your wife, that’s not workplace violence in my book.
Nor does workplace violence have to happen at the workplace. If someone drives to their boss’ house, and kicks his ass there for something that happened at work, that’s still workplace violence in my book.
To me, “going postal” means someone reacted in a violent, physical way against a coworker or coworkers (including bosses!) because of something work-related. Murdering a coworker is just the most extreem case of “going postal”, but it’s not the only way “going postal” manifests itself.
Its anecdotal at best ,no cite,but I’m sure I read somewhere that there was a high incidence of violence amongst staff in restaraunt kitchens on hot days in premises that had higher temperatures as the norm all year round.
This is more an indictment of speed- and power-crazed drivers than anyone in a workplace. 
A few of the posts point out the major problem with this column: It does not address the original question. “Going postal” refers to violence against co-workers (or, ex-co-workers), not the general threat of violence in the workplace. This is the first time I have ever seen an answer by TPM to be so off base. The comments and stats regarding taxi drivers and cops have no relevance to the specific question at hand.
Sure, it does. It cites several violent postal-related incidents and gives some stats, then goes on to put the issue in perspective; although “going postal” has becomes a popular meme, they remain a fairly rare occurrence and a small fraction of work-related deaths overall.
What “address” to the original question would you have expected? Summaries of every incident until he reaches 600 words?
The original question was: “Is it true U.S. Postal Service workers are more likely than other occupations to “go postal” on coworkers?”
So it depends on how you define “going postal”. Does violence that doesn’t result in death not meet the definition of “going postal”?
The title of the article was: “Are employees of the United States Postal Service more likely to be violent towards coworkers?”
So the actual question was reframed in a way that would suggest that “going postal” equals being “violent towards coworkers”, and not just killing coworkers. But then the body of the answer side-steps this, and addresses only homicide by coworker.
So perhaps we first need to stop and define just what “going postal” means. And I would suggest that the more general idea that includes all co-worker on co-worker, workplace related violence is a more valid definition of “going postal”, as the article’s title suggests. Perhaps some people much more narrowly define “going postal” as only being those cases when a co-worker kills other co-workers, but if we’re going to narrow the definition that much, that needs to be stated, and the article needs a more appropriate title.
FROM SEINFELD George: Let me ask you something… What do you do for a living, Newman?
Newman: I’m a United States postal worker.
George: Aren’t those the guys that always go crazy and come back with a gun and shoot everybody?
Newman: Sometimes.
Jerry: Why is that?
Newman: Because the mail never stops. It just keeps coming and coming and coming. There’s never a letup, It’s relentless. Every day it piles up more and more, but the more you get out, the more it keeps coming. And then the bar code reader breaks. And then it’s Publisher’s Clearinghouse day.
Jerry: All right, all right.
Okay, Rusty. Why don’t you write 600 words on it and we’ll have a debate on which essay is the more entertaining and informative.
Why? Because I disagreed with you? Has the discussion really already degraded to taunting?
My point isn’t that I could do a better job. My point is that the term “going postal” was framed one way in the article title, and framed another way in the answer to the question. I did not assert that the answer wasn’t entetaining or informative, nor did I assert that I could do a better job. I simply disagreed with your assertation that the answer addressed the question.
If you disagree with me, I’d appreciate it if you could support your opinion with something other than an unrelated taunt.
Taunting? Hardly. It’s a serious invitation for you to share what you think would have been a relevant on-point answer to the question given to Cecil Adams. I see claims that CA did a bad job, with no indication of what criteria would identify a good job.
I already have:
If the answer to a question about “going postal” can’t include data on how common it really is and how it compares to other workplace slayings, then I guess we do have to agree to disagree. The mistake is to call what I’m doing “taunting” as though I had some motive to attack you personally. I don’t even know you.
Indeed, anyone who has done any research knows that the vast majority of postal workers are merely junk mail toting crazies, who commit their lives to the delivery of the mail and the ongoing battle with the Polar Bears. 
As far as school violence goes, from what I understand, the year of the Columbine shootings was also the year with the lowest incidence of school violence, as it turns out.
And has anyone considered that postal worker shootings may in fact be caused DIRECTLY by violence depicted in things delivered by mail? Take, for instance, that Tom Clancy book you ordered from Amazon.com!
What we’d like to know is whether postal workers are more likely to murder a co-worker. Specifically, I’d love to see the numbers compared to the national average and to the closest runner-up (or leader).
For anyone who is interested, the USPS report cited can be found here: http://www.casacolumbia.org/Absolutenm/articlefiles/33994.pdf
This report claims that:
Seems like a cop-out to me. I’d like to know how things compare in the 54% of cases where they can tell.
You can’t make a meaningful statement with a sample size of only 16. Samples of at least 30 or 40 are considered barely accurate and samples of at least several hundred are needed for significant accuracy. You’re demanding Cecil make a definitive statement but the information just isn’t there. It’s like looking at a baseball team’s performance over a two-week period, insisting someone predict their standing at the end of the season and being disappointed when they can’t.
FWIW, I was saying the USPS was copping out, not Cecil. None of us are in a position to make demands of TPM.
I demanded that he provide me the recipe for a perfect Margarita, but he never came through. I figure anyone is in a position to demand such a recipe from anyone who is Perfect. 