Hacker News new | past | comments | ask | show | jobs | submit login
U.S. Postal Service releases dog attack national rankings (usps.com)
56 points by infodocket 13 days ago | hide | past | favorite | 86 comments

This part was disappointing: "Top 10 Dog Bite States" because it is probably driven by population and does not provide unique insight. It would be awesome to, minimally, scale this by state population -- or ideally, scale this table by unique mailing delivery addresses.

But the fact that Houston is on top with nearly 20% more bites than LA yet LA having a much larger population leans toward it not being directly population related.

You need to normalize to the dog population at least - rather than the human population.

If I were a postal employee, I'd want it normalized to bites per hour worked

Probably need to normalize both to dog population and people population (perhaps normalize to dogs per capita?)

Houston has less than half the population density as LA. Bigger properties / houses / yards almost certainly means more dogs per capita in Houston than LA.

Rockford IL is punching way above its league!

*biting above its league.

I'll see myself out

Yeah, my thoughts exactly. Also, some states (e.g. NY) may have a higher percentage of mail delivered to "mail hives" than other states (e.g. Ohio). I would think that delivering to a mailbox on the house-owner's property is a lot higher risk for dogbite than delivering to an apartment-renter's mail slot. Does that account for why Ohio is as high as NY? Maybe, but it's hard to say from this data.

I'll bet central mailrooms are a big driver for safety. In my NYC building, for example, the mailman does not even see any of the public space -- they enter the delivery room through the mailboxes' rear side and deliver en masse to all 110+ units' mailboxes. I imagine the higher the highrise the more efficient the delivery and the safer the delivery.

Now contrast that to crossing lawns individually house by house...

Scaled by population, at least for CA/TX:

39.51M / 782 = 72139

29M / 402 = 50524

Surprisingly higher rate, especially considering I'd -guess- more dogs in CA are kept indoors, although obviously the dog-ownership-rate is another important adjustment. This is all fairly meaningless data without loads more controls.

You swapped the results, and should be computing the reciprocal anyway.

782 bites / 39.51 M people = 19.8 bites per million people per year (CA)

402 bites / 29 M people = 13.9 bites per million people per year (TX)

Erm aren't these numbers people per dog attack? So the higher number means fewer per capita attacks? Seems like a weird way to calculate a rate. Regardless, label your units people!

Why would you guess more dogs are kept indoors in California? Much of Texas gets too hot for dogs to live outdoors during the summer.

Isn't Texas often hot enough that it would be unfriendly to leave dog outside?

I was thinking the same thing. It should be something like dog bites per 1,000,000 residents or something like that. Then, the rankings would probably change.

Per resident or per number of dogs? City populations are going to have fewer dogs and the dogs in cities are going to be leashed or in doors and less likely to bite.

This doesn't really matter from the point view of the victims of the bite. It should be normalised by number of man-hours worked.

Are they just trolling us by not doing a pro-capita version? "California is the most populous state, news at 11"

Per capita, or per letter delivered or something would be an obvious first step, but also it is measuring which cities have yards and driveways. Lots of people live in NYC, but not so many have enough space for a good dog attack.

> ...but not so many have enough space for a good dog attack.

Of all the reasons people cite for moving out of the city, I think this one is my favorite. :)

"Sure, Williamsburg was awesome when I was single and in my 20s, but now that I have a family, I'm looking for somewhere better suited for a good dog attack."

"Pet Peeve #208: Maps that are actually just population density"


Government job security challenge: https://ibb.co/vxj3BKf

Edit: Turns out I'm a dummy and can't figure out how google works. I am totally wrong on this one!

What I found odd is why Huston is number 1. Seattle is known for how many dogs it has and is a much larger city! I wonder why they have more.

Houston is the fourth largest city in the country. It has three times as many people as Seattle.

Yeah turns out I'm stupid. Sorry about that!


No it isn't.

Seattle metro is 3.9M, Houston is 7M.

Sorry yeah I'm totally wrong about that. My fault!

If you work for the USPS in California, you have a 1.23% chance of being attacked by a dog. This doesn’t adjust for repeat dog incidents, nor whther You have a desk job or are in the field. Likely much higher for mail carriers in the field which is nuts

In my country there is a simple rule: dog bites human; end of dog. This notion keeps owners a tad more involved. Not sure how that is in other nations.

While this is interesting, it would be far more useful if this was broken down by breed and zip code.

It would allow focused efforts on rescue and enforcement of aggressive dogs. And perhaps even give insight into crime in certain areas.

Too bad they don't track by breed

The breed doesn't matter, it's the owners that influence the temperament of the animal. The nonsense statistic you ask for simply causes people to associate certain breeds with attacks when in reality it's the owners who do not keep their dogs restrained who need to be analyzed.

I'll just leave this here for all the pit bull lovers: https://www.dogsbite.org/dog-bite-statistics-fatalities-2020...

One problem is pitbulls are not really a standardized breed so a lot of mixed breed terriers get mixed in here.

Though this is an important metric. I don't think pitbulls are inherently more likely to bite, but I do think they cause a lot more damage when they do bite. Minor bites are unlikely to always be reported. Note that all of the dogs listed are big powerful jawed dogs.

USPS doesn't have a set schedule to come to your house, and they have to enter your (perhaps fenced) yard.

It's very plausible that a responsible dog owner would have a dog out in the fenced yard when a postal worker came by, and then the breed is definitely going to be a factor.

Dogs have completely different personalities across breeds and yet you won't even entertain the possibility the breed has something to do with how aggressive a dog is?

This is why usps should've released that info.

There is already data out there about this

This is why the data would help.

Most people saying this have no proof of their claim.

Labs. I just bet.

(I've been bitten twice when riding, both times by labs. They're goofy dogs, but have a mean streak if they don't know you.)


Please don't. It's just very weird to compare dogs and humans. Dogs have been bred for certain behaviours, humans haven't. Which is why It's relevant to ask for breed types for dogs. I just find it to be in very poor taste to try to draw parallels between "dog discrimination" and human discrimination.

You can absolutely think broder collies are naturally driven to herd dogs, or Newfoundlands are naturally driven to fetch from water without being racist to your fellow human. I mean wtf!

Dogs have been bred, and certain breeds are nature-inclined to certain behaviors, but the parent is somewhat correct in that the influence of that nature is limited.

However, I think the more powerful correlation is that of owners selecting breeds and training for expected behaviors: Some Newfies can be trained not to jump in the water at every opportunity (mine is capable of this restraint, but will always let me know she wants to). Few Newfies will be afraid of the water. But dog owners who know they want their dogs in and around water may select Newfies, swim with them, and otherwise encourage that behavior. More relevant to the article, owners who want dogs to guard their property will select certain breeds who may have only a slight nature for guarding but will nurture that behavior.

I obviously agree that it's still a lot of nurture, and that breeds can totally deviate from their original "characteristics" or intended purpose. My point was more that it's still completely different to humans because broadly speaking, specific dog breeds are much more likely to instinctively behave in a specific way. Whereas let's say the difference between middle eastern and white men is almost purely "nurture", and the nature part is usually purely physical.

And while yes, dog owners do have a huge influence on a dogs character, it's undoubtedly harder to teach a collie how to hunt and a terrier how to herd than it is to teach them the role they are actually known for. That's not true for humans, and his comparison is just incredibly uncomfortable because it implies a similar difference between say, pitbull and a greyhound. Which is not true.

Modern dogs are so far from what they were bred for, the dominant influence by far is the influence of the owners. The only relevant statistic would be: percent of bites by dogs that are owned by humans who have not properly trained the dog. That will be a statistics that is more than trivia, the charts in that article are useless trivia paid for by tax dollars.

A useful study would be able to provide a recommended course of action an owner could take to reduce the likelihood their dog bites, which is what my proposed study above would do. TRAIN YOUR DOGS PEOPLE.

>certain breeds are nature-inclined to certain behaviors

Dogma. Prove it. Link to any amount of supporting evidence.

You are by far the most dogmatic person on this comment section. Are you actually saying there's no genetic element to dig behavior? Not only does that go against scientific consensus, I'm not even aware of any debate or controversy on the subject at all. There's certainly a debate around which one is the dominant factor, but afaik it's pretty fringe to claim there's no genetic factor. I assume any amount of citations and proof won't convince you but there you go:

1. Serpell, James, ed. The domestic dog. Cambridge University Press, 2017. (start on page 70, available on Google books, too long to quote)


2. "We found that a large proportion of behavioural variance across breeds (among-breed heritability) is attributable to genetic factors (figure 1a). The mean among-breed heritability was 0.51 ± 0.12 (s.d.) across all 14 traits (range: h2 = 0.27–0.77), and significantly higher than the null expectation in all cases (permutation tests, p < 0.001). These estimates are also significantly higher than those in previous studies assessing heritability of the same traits in large within-breed samples (mean difference = 0.37, 95% CI = 0.44–0.30; t13 = −12.25, p < 0.001; [44], but see [45]). Incorporating among-breed variance thus yields h2 estimates that are on average, five times higher (range = 1.3–25.5 times higher) than traditional within-breed estimates, which could be due to limited genetic and phenotypic variation within breeds." (MacLean, Evan L et al. “Highly heritable and functionally relevant breed differences in dog behaviour.” Proceedings. Biological sciences vol. 286,1912 (2019): 20190716. doi:10.1098)

So what you did here was google something then cherry pick examples that support your findings. However, if you dig into the sources you'll find the authors are part a network of psuedoscientist eugenicists.

General consensus among proper studies is the opposite: behavior is not highly heritable.


Table 1 is quite extensive. Nearly all of the two dozen studies show less than 10% heritability of the measured behaviors.

> Dogs have been bred for certain behaviours, humans haven't

Humans were. And are.

Dogs were engineered explicitly for various behaviors over the course of tens of thousands of generations. On top of that, treating one dog breed as a higher bite risk over another doesn't have a severe negative impact on society, and a dog can't feel marginalized or belittled.

The act of identifying dog breeds by behavior statistics and the act of socially discriminating human beings are nowhere near comparable - not in terms of real harm nor theoretical harm nor practical usefulness.

> If you do it to dogs then you'll do it to humans.

How did you come to that conclusion?

I used to have a couple dogs and now I have children. I walked the dogs on leashes and made them poop outside. I've never done either of those things with my children, or any human.

The assumption is that different dog breeds are more aggressive than others, independent of their upbringing. Their behavior is in their nature. This hasn't been born out by science yet is still used to justify policy. People care less when it's done to dogs because people are not dogs and it's easy to just say "sure, that sounds right" without digging any further.

Show a single study that finds that dog breed influences aggression. Just one. You'll find plenty that correlate the two and maybe even some that stop there and ignore the existence of other factors. It's The Bell Curve for dogs.


Correlation vs causation is irrelevant in most dog bite contexts, where the practical concern is harm avoidance.

Then why only look at one correlation? Wouldn't it make more sense to look any many correlations that could potentially correlate better, such as owner demographics? Or would it be better to only discriminate against dogs and not their owners? We all know how much the dogs use USPS.

If harm avoidance is the goal then why not take it on by promoting better dog upbringing?

If it turns out that pit bulls are 10x more likely to bite me than golden retrievers, it doesn't matter whether this is due to genetics or to correlated owner behavior. I'm walking down the street, see a pit bull, and have only that information to go on when assessing whether I should proceed cautiously or turn around and increase distance.

Also relevant is that the breed has an effect on the seriousness of injury if an attack does occur. Obviously I'm going to be less fearful of a chihuahua even if they're more aggressive than the terrier. It's the difference between antibiotics and stitches vs. facial reconstruction.

Who is only looking at one correlation? We all take into account the information we have. More information is better. If they had provided more information, it would have improved each reader's overall picture.

>Or would it be better to only discriminate against dogs and not their owners? We all know how much the dogs use USPS.

I don't know what this means.

>why not take it on by promoting better dog upbringing?

I do promote that. I don't know where you're getting this from.

Back to the idea of harm avoidance:

Let's say we discover that the only factors that determine bite risk are how the dog is raised and how you behave around it. And let's say there is a non-causative correlation showing breed X has a high bite risk. Then it would be reasonable when adopting a puppy to not consider breed X a high bite risk, but it would also be reasonable to consider breed X a high bite risk when encountering one you didn't raise yourself, and therefore e.g. not allow your child to interact with it (or whatever precaution makes one feel comfortable). More information is better.

>>Or would it be better to only discriminate against dogs and not their owners? We all know how much the dogs use USPS.

>I don't know what this means.

I'll explain my meaning with a hypothetical.

What if one breed were responsible for a high number of USPS delivery worker attacks? Promoting this information begs a policy response against this breed. What if most of this breed were owned by a certain demographic? From the outside this looks like a policy response against a demographic with window dressing to justify it. The outcome is what matters.

I agree with your later point. However, dog owners and parents are responsible for their dogs and children (respectively). More information is better, but selecting some specific information here is clearly targeting something or someone, regardless of intent. That is what I take issue with. Target the issue, not the demographic.

See my other comment but basically we need to know what percent of those dog bites were done by dogs owned by people who didn't socialize and train the dogs. That would be an actionable statistical analysis.

I understand your concern about dogs, but the question is why we would think what a person believes about a breed of dogs impacts how they approach humans. You said it well yourself:

> People care less when it's done to dogs because people are not dogs

I have actually seen children on leashes. Heck, it was semi popular during he 80s and early 90s because of the stranger danger scare.

> The assumption of nature over nurture has been continually used to justify unequal treatment. If you do it to dogs then you'll do it to humans.

I mean, okay.

> What information would that give?

Which breeds are more likely to attack mail carriers.

It would be a single correlation. If the assumption is then made that there is a causation and that is used to inform policy, then there are unequal prejudices informing public opinion and policy. Pseudoscience masquerading as science. Something bad masquerading as something good.

You do it for dogs because you're not a dog, no big deal. Maybe you'd do it for people with a different skin color because you are not their skin color. It has happened continuously in the past 100 years and continues to happen.

> Pseudoscience masquerading as science. Something bad masquerading as something good.

If a certain breed of dog is responsible for more attacks than any other breed, then breed-specific legislation is not pseudoscience.

Yes it is. This is the very crux of the issue. Simply showing a correlation is not demonstrating that behavior is linked to breed. Show a single study that even attempts to prove dog nature is linked to breed. You'll find that actual science finds the opposite.

Sheep dogs like to herd. Retrievers like to retrieve. Covering your ears and saying lalalala doesn't allow you to claim correlations are immune to causality.

What makes you say sheep dogs like to herd? Because that is what they do? If you trained a retriever to herd sheep and a sheep dog to retrieve, would the dogs not do those things? What would the dogs do if they were never socialized? Are you even considering these questions, or are you covering your ears and accepting the not-science dogma you've heard your entire life?

> What would the dogs do if they were never socialized?

Funny you should ask. I own a herding breed right now.

1. Socialization has really no impact on what a dog is bred to do or what it has an aptitude to do. Socialization mainly teaches dogs how to read "dog cues" and how to fit into the dog world. That is - to understand when another dog does or does not want to play, who the dominant one is, what the polite greetings are, etc.

2. My breed does certain things that are unique to the breed, and exhibits behaviors unique to the breed. She was not trained to do so. She lays prone on the outskirts of the field, anticipating a command from a human to move the stock in a given direction. She does this, despite never seeing a sheep/cow in her existence, nor being trained for herding in any way.

Whenever I see dogs of this breed at a dog park, they exhibit the exact same behaviors, down to the posture and eye focus.

The more I read your comments in this thread, the more I realize how little you actually know about dogs, nor did you actually read my comment before labeling it pseudoscience.

Partly, what I meant was that the causation is also irrelevant - if 1 breed is responsible for most attacks, and there is legislation against it that solves the issue - that doesn't mean anyone has been discriminated against. It just means there's going to be less attacks in the future.

> What makes you say sheep dogs like to herd?

Because they do. That's what they were bred for. This has nothing to do with human race analogies. Dog breeds influence behavioral traits heavily. I'm not convinced you know anything about dogs.

Have you ever, in your life, considered that you may sometimes be wrong?

> Pseudoscience masquerading as science. Something bad masquerading as something good.

So if a rigorous scientific study concluded that pit bulls are more likely to attack mail men than golden retrievers, normalized to the same starting conditions, is it ok to make discriminatory policies against them?

The volume of dog owners who can't control their animal is shocking to me.

My neighborhood it is pretty good. A few blocks over is complete chaos. Every house has a dog that seems frantic and freaks out over everything and they're all only held back by an invisible fence.

I don't get it. What do these people get out of a dog that is clearly frantic / stressed like that when any person comes by, or another dog or anything?

If you're not going to invest some serious time / effort to train your dog, please don't get one.

I don't get it either.

Rural VA/WV vs suburban VA is night and day. The former has ill-tempered dogs roaming at large - rare is a weekend I go cycling out there and don't get chased by at least one dog (and of those, 1/10 are willing to bite). Locally, the dogs are on leashes and generally more socialized.

And suburban VA compared with anywhere in Scotland is again night and day. Dogs everywhere in Scotland, mostly off-leash in open areas (on leash in where appropriate), but almost always well-behaved. At least when I've visited family (Edinburgh, various towns in Perthshire, plus some time vacationing in the Highlands). Totally different culture with dogs - allowed in pubs, people don't freak if the dog is off-leash like they do here, etc.

This is why Switzerland past very strict laws which require you to take several dog courses with your dog or you will not be able to own it (all dogs must be registered). Certain aggressive breeds are also illegal or are required to wear a muzzle when out.

All this after kids and people getting mauled by dogs back on the 90s became a common news item.

>All this after kids and people getting mauled by dogs back on the 90s became a common news item.

Again, this proves the original point of people not being able to control their dogs. I'm not a fan of certain dog breeds, but that's my hang up and I'm certainly not going to tell someone that the can't or shouldn't own a dog because of my own preference. I'm all for the mandatory training and registration, but the outright banning of breeds seems unnecessary if you trully are enforcing the training.

I think it’s probably fair to say a lot of people get dogs as pets without too much regard for training them.

Usually if the dog is well socialised then you’re unlikely to have too many issues. If not, it can be difficult to correct the behaviours and I’d wager most people in that situation probably just put up with it.

Yeah the contrast is amazing. All but maybe one dog in my neighborhood are happy to hang out in the front yard and they're cool with anyone who comes by.

Most don't even need a leash, they're content to just hang out, play, and socialize if someone shows interest.

Other neighborhood, it's dog chaos.

Some people keep dogs to discourage break-ins. We had a moderate level of crime in the neighborhood where I used to live, but my neighbor’s big black noisy dog ensured that our houses and cars were left alone. This is probably also the reason in many rural areas.

I have a huge American Bulldog. He's playful, fun, obedient and extremely well trained ... and outright terrifying when someone approaches the house. We say "I see them, thank you" and he's friendly again. He'll go find us to alert us to people in the yard. At night, we have to go walk through the house together if he's on edge.

Everyone is quite intimidated until they come visit and have a proper meeting. He also remembers guests quite well, and reacts differently to them. He learns mailpeople's faces / walks and gives a cursory "gruff" for old ones, and the full-throated threats to new ones.

That's a feature: Dogs are the best break-in / intruder deterrent around. My wife originally bought him for that reason. There's never been a problem with kids, friends, neighbors, or anything, but if you were walking through our neighborhood and saw him threaten you, you'd probably say he was a menace. He's not.

I don't doubt it works.

In fact I keept a chewed up tennis ball and dog dish next to my back door for ages. No break in attempts, despite it happening to neighbors ... no dog, but it seemed to be enough ;)

Having said that as a more practical point if I were some super burglar (granted I suspect most criminals are not) ... I really think most angry dogs would happy change their mind and be your best friend if you just dropped some treats for them. They really wouldn't care what you did after that.

Dogs are a great alarm, decent deterrent, and wonderful companion. They somehow react to the very human uncertainty that someone feels when a big-toothed animal is acting aggressively. It's unsettling on a primal level.

If we need more deterrent, there are other options in the house, but it's nice to have early warning at least, I'd imagine.

On a semi-related note, I bet the number of "frantic" dogs is up these days. As more people stayed home last year, more dogs were adopted, but the owners couldn't properly socialize them with other dogs or even different people in public places or just by having guests, which I reckon is a big part of the problem.

"Houston, we have a problem"

did you know the dog breed most implicated in dog bites is the golden retriever? they're bred to retrieve hunted water fowel so their bites are soft and rarely require hospitalization after an attack


somewhere on Dog Internet there’s a detailed ranking of states where homes are most frequently invaded by strangers bearing mysterious parcels

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact