People who own vicious dogs such as pit bulls have significantly more criminal convictions - including crimes against children - than owners of licensed, gentler dogs such as beagles, American researchers report.
A study of 355 dog owners in Ohio showed that every owner of a high-risk breed known for aggression had at least one brush with the law, from traffic citations to serious criminal convictions.
And 30 per cent of people who owned an aggressive breed of dog and who also had been cited at least once for failure to register it had at least five criminal convictions or traffic citations.
This compared to one per cent of owners of low-risk, licensed dogs such as poodles, beagles or collies, the researchers reported in the Journal of Interpersonal Violence.
"Owners of vicious dogs who have been cited for failing to register a dog (or) failing to keep a dog confined on the premises ... are more than nine times more likely to have been convicted for a crime involving children, three times more likely to have been convicted of domestic violence ... and nearly eight times more likely to be charged with drug (crimes) than owners of low-risk licensed dogs," said Jaclyn Barnes of Cincinnati Children's Hospital Medical Centre.
Barnes and colleagues used public records to check on the criminal pasts of dog owners.
They used agreed definitions of vicious dogs used in writing local ordinances. "A 'vicious dog" means a dog that, without provocation, has killed or caused serious injury to any person, has killed another dog, or belongs to a breed that is commonly known as a pit bull dog," they wrote in their report.
The definition excludes dogs used in law enforcement or dogs protecting an owner or property.
Aggressive breeds identified by the Society for the Prevention of Cruelty to Animals and some insurance companies include pit bulls, rottweilers, akitas and chows.
The most frequent low-risk breeds seen in the study included terriers, beagles, collies and poodles.
My friend had a pit bull that was great. My problem with them is that when they bite, they do not let go. From witness accounts I have read where an adult has tried to get a pit bull off a child or person, it is nearly impossible unless the dog is getting repeatedly beat with bats or shot. That is the problem. I remember one man on TV who saved a child talking about how he was beating the dog with a bat and the dog was unfazed. The man was hitting it as hard as he could to get the dog to stop and let go and simply could not believe the dog was unfazed from the beating he was giving the dog. Not that some go crazy or snap, but when one of the dogs attacks (and I have read stories of the pit bulls attacking children in homes, unprovoked, they were raised in) it is nearly impossible to stop it. When other dogs bite, they let go, I know there are instances where they do attack. The difference is in the ability to stop the attack once it is started.
Here are examples of these animals.
An 82-year-old Spotsylvania, Virginia woman was mauled to death by her neighbor's three loose pit bulls.
A two-year-old Huntington, West Virginia girl was buried yesterday, killed in a mauling by a neighbor's pit bull that was supposed to have been quarantined eight days earlier.
A St. Charles County, Missouri man who owned two pit bull mixes was killed in his home by one or both of them, due to crushing injuries to his neck from dog bites. Neighbors said he loved his dogs and had talked about how well-behaved they were.
ATTACKS BY PIT BULLS OF FAMILY AND FRIENDS
In another St. Charles County, Missouri incident, a man and his 13-year-old stepson were attacked viciously by a relative's pit bull in their home.
In Charlotte, North Carolina an eight-year-old girl waiting in a family friend's house to get her hair braided was attacked by a resident pit bull, and suffered an armpit puncture wound, a missing chunk from a calf, and a blood clot threatening a main leg artery.
A 23-month-old Portland, Oregon boy was in critical condition after being attacked by tethered pit bull owned by his family, in his own back yard, with an adult present.
In Albuquerque, a two-year-old girl was mauled by her family's three pit bulls, and was in critical condition with multiple bites and massive upper torso trauma.
A six-year-old Florida girl underwent surgery to have her left eyelid re-attached after it was bitten off by the pit bull of a family friend on a Mother's Day outing. She had approached the dog carefully, to pet it, when it snapped. The dog's distraught owner said it had never shown aggressive behavior before, and would be destroyed.
A Tennesee grandmother was shocked after her previously well-behaved pit bull mauled two of her grandchildren.
This study is flawed, and any newspaper or magazine that published about it without pointing out its many and major flaws has severaly compromised its journalistic integrity. It is hate propaganda. Substitute the name of a visible minority for "pit bull owner", republish, and watch the ACLU go ballistic on you.
To quote a friend...
"It's been interesting to watch the media play pass-the-word with the latest so-called study of dog ownership patterns. In Windsor (one of the worst Tin Pot Towns in Ontario) the other day, it was 'Troubled Owners, Troubled Dogs'. In Vancouver, they used 'Nasty Bowser Likely Has Nasty Owner'. In Ottawa we read "You Are What You Walk'. See how things get distorted, little by little? I can't wait to see what the next column headline says. How about 'Pit Bull Owners are Fair Game'? At least that one would be true.
Bulletin November 19: It's in Oz now with the headline 'Pit Bull Owners More Likely to be Criminals'. Man, this is some hurtful stuff these morons are spreading around. I really feel for my friends right now. When is this going to end?. How much of this crap are we supposed to take? Why won't anybody make them stop? I just don't know what to do anymore so I'm sitting here weeping at my keyboard typing on this little blog trying to fight back. Go ahead and laugh, I don't know what else to do.
I note that the Ottawa Citizen scribe decided to add a little oomph to the press release and in so doing, revealed that she either hadn't read the actual paper or didn't understand it. I have read it and would like to point out some of the language which I find concerning.
1. "People who own "high-risk" dogs are much more likely to be high-risk themselves, a new study says, with vastly more criminal convictions and traffic citations found among those who own dogs that don't play well with others."
While it is obvious that an abusive or neglectful dog owner will be more likely to have antisocial tendencies, to state that the above is shown in this study is simply untrue. Aside from the fact that the 'high risk' breed attribution is suspect at best, the study group used very narrow subject selection criteria and because of the method used, there is no indication that there were 'vastly more' criminal convictions or traffic citations in either the 'high risk' or 'low risk' groups. Dogs were selected based on citations from animal control or from licensing records which in Hamilton County, Ohio are filed by breed code, so it would be very easy to select reported breeds of dogs to review.
According to the authors, only 0.5% of the cited dogs had been involved in an aggressive incident (killed a dog) 1% were cited for 'unknown' reasons and 0.5% for failure to muzzle. In other words, 98% of the Animal Control citations were for other things, such as failure to license (44.7%), failure to keep dog confined to premises (41.5%), failure to obtain liability insurance (11.7%), well you get the idea. Careless? Yes. Criminal? Not in my world.
My impression of the paper on a quick first reading was that the authors had decided upon a conclusion and then used very specific and exclusionary data collection methods to prove it. Actually, despite the obvious bias throughout the article, the so-called 'high risk' dogs did as well, or better than the dogs in the 'low risk' group.
Put another way, 'pit bulls', of which the greatest number of dogs (153) overall were studied, showed a citation rate of 52% and non-citation rate of 48%. This compares favourably with Dalmatians (57% cited vs 43% not cited), Shih Tzus (83% cited vs 17% not cited) and Chinese Crested Dogs (100% cited) just to give a few examples. It's really all in how you work the numbers, ie, while 100% of Akita owners were cited, only three dogs were studied. When the percentage is much higher than the actual number it is always a statistical red flag.
2. "researchers examined the relationship between high-risk dogs and "deviant" behaviour in their owners."
Deviant is a sociological term in common use which means non-conformist, unconventional, or criminal. The fact that the columnist puts it into quotation marks shows she is unfamiliar with the term or is trying to show irony. Either way, it's telling.
3. "In a sample of 355 dogs and owners, the study authors categorized as "high-risk" all pit bull-types and dogs of other breeds that had killed or seriously injured a person without provocation, or killed another dog."
Actually, they used information from the insurance industry (which is dancing in the streets over this new line of business, a rare thing for them) and from a 1983 research paper about pack predation* which did not highlight the 'high risk' breeds, namely 'pit bulls', Rottweilers, Akitas and Chows at all. A few of the dogs cited, both 'low risk' and 'high risk' had been involved in aggressive incidents. The study group also used the municipal definition, 'pit bull' to determine what was or was not a 'high risk' dog. In the Cincinnati bylaw definition, a 'pit bull' type is considered 'vicious'. Even these authors state that the 'high risk' designation due to breed or type was basically meaningless in terms of actual behaviour.
I hate to be a wet blanket as I know how much certain members of the media like to play up anything to do with 'pit bulls', but not every study paper has merit.
I have serious problems with the table of breeds which is supposed to prove the premise.
First of all, we all know by now that 'pit bull' is a catch-all term which refers to at least three and up to five purebred dogs (and an unknown number of mongrels), depending upon where you live. It is not a breed. Oddly, those purebreds are missing from the table.
Second, there is no breed known as 'terrier'. Terrier is a group consisting of , in Canada, 26 breeds, at least two of which are dogs commonly and erroneously labeled 'pit bulls'. One Jack Russell and a few Schnauzers are the only terriers on the list and then there is the catch-all 'terrier', even though the authors state that terriers (24 generic), beagles (20), collies (18) and poodles (17)' were the most common breeds in their sample.
I note that the second-most popular dog on the Hamilton County licensing website (GSD) which is mentioned in the article as being 'dangerous or vicious', is absent from the table, as are a few other popular breeds such as the American Cocker Spaniel, German Shorthaired Pointer, Bullmastiff, Bull terrier, Malinois, Border Collie, Corgi (Pembroke or Cardigan) and so on.
The most popular dog in the County according to licensing statistics, Labrador Retriever, had only one dog reviewed in this study. It is also rather odd that there were only 8 'mixed breed' dogs studied in this part of Ohio, other than the 'pit bull', 'terrier' and 'hound' categories, all of which are generic, mixed breed terms that were shown separately.
From a housekeeping standpoint, there are errors in the figures, which calls the data presented into question and makes one wonder what other errors may have slipped through. This paper, in my view, is nothing but opinion disguised as research, performed by an obscure group, printed in a non-peer-reviewed journal and then released to the media in order to attract attention.
Who came up with the research question? Why is it important? Who funded the study? Why are there two authors from the local SPCA, one in the senior author position (last), when neither has any research experience?
Why is the thrust of the paper designed to instil fear into both the dog-owning and non-dog-owning public, starting with the sensationalized title? Will the implied threat in the conclusion, ie, that social workers should always determine the breed of a family's dog as a marker of the suitability of a home for children, achieve what straightforward breed bans have not? Will this be the final tool used to get people to dump their dogs, the fear that their children may now be taken away too? As I read through the paper, I detected the fine hand of the AR fanatics who now appear to be walking in lockstep with social engineers who cannot tolerate non-conformity or views that differ from their own.
A few observations for your consideration:
Hamilton County, Ohio, which includes Cincinnati, has a population of about 850,000. A reliable ratio for a dog population is 20% of the human population, so we could conservatively estimate that there are 170,000 dogs in Hamilton County.
As we also know that licensing compliance in any municipality, in the absence of active enforcement, typically runs at about 10%, we can estimate that the available pool of licensed dogs is around 17,000. However, the County's stats show 21,703 licensed dogs just among its top five favourite breeds (Labrador Retriever, German Shepherd Dog, Golden Retriever, Beagle and Boxer) and the study states that the subject pool was just under 50,000 licensed dogs (two year review). The licensing figure above revises the estimate upward to 217,000 at a minimum which is well within the range we would expect to see.
The authors selected one group of owners with citations from Animal Control and another group of non-cited dog owners from the same database divided by 'low risk' or 'high risk' breeds, ending up with a study group of 355 dogs in total. Of this very small group, 167 dogs were considered 'high risk', because of citations or their shape. High risk and low risk owners then had their names run through a police database to determine whether they had any criminal or traffic convictions.
OK, so let's see how everybody made out. Here's Table 2, sorry, it's blurry.
[sorry, table would not copy]
Bearing in mind that this is offense-related, not participant-related, it sure looks as though there is a problem. Due to the fact that the high risk cited subjects average 4 offenses (criminal, AC citations and traffic) per person, it is obvious that there are repeat offenders in the group. We see that low risk cited subjects average 1.08 offenses (convictions, AC citations and traffic) and again, there must be repeaters. In other words, we have a slew of convictions but that's about it, divided by a subjective dog categorization scheme.
What is bothering me is the inclusion of the citations in the totals. If I decide to divide groups by height, ie, short and tall, then subdivide them into short and tall without/with moustache, add traffic tickets and then do a table to show the result, isn't it pretty clear that I will have lower overall numbers of non-moustached people when I do my tally, since the absence of same is one of my selection criteria?
In this example, I am using 100 people in each subgroup and each subgroup has 50 offenses (O). LR with/without citations (C), HR with/without citations. I could add more factors but this is nice and simple and as you know, I am not a scientist or a math whiz.
[SORRY, TABLE WOULD NOT COPY - SEE HTTP://CAVEAT.BLOGWARE.COM]
These data, without qualification, are sensational but fairly meaningless. The only thing they prove is that if you seek, you will find. It reminds me of the missing breeds - why weren't they included?
Why was just one example of the most popular breed in the County, Labrador retriever (8,459 licensed) and not one of the next most popular, German Shepherd dog (4,915 licensed), reviewed?
Why were only a few dogs of each breed or type studied, except for 'pit bulls'?. After the catch-all 'pit bull' category, the next largest was the catch-all 'terrier' category at 24. Six times more 'pit bull' owners were reviewed than 'terrier' owners. One hundred and fifty-two more 'pit bull' owners were reviewed than owners of Labrador retrievers, Australian Cattle Dogs, Belgian Shepherds (Tervuren only), Shetland Sheepdogs, Redbone Hounds, Pointers, Lhasa Apsos, and so on. Where are the other hound breeds, the dogs so well known in Appalachia, the Foxhounds, Black-and-Tan Coons, Blueticks and the rest?
Was it because a lot of dog owners would not sign the consent form? Was there a consent form? Was any kind of inducement offered to participate? Maybe a few bucks, a bag of dog food?
Just asking...
Here's the beginning of the Limitations section:
Limitations
In discussing the results of the current study of the ownership of high risk dogs as a marker for deviant behaviors, specific limitations should be kept in mind. Recruitment of persons eligible to be classified as owners of high-risk licensed dogs was limited by the small pool available in Hamilton County. Thus we were unable to match this group on zip code that controlled for neighborhood characteristics.
Oh? That's funny, because you managed to scrape up 153 'pit bull' owners, 73 of them licensed, yet you say that 'pit bull' owners don't license due to fear. Gee, I wonder why that would be? There are only 4 Rottweilers in Hamilton County, 3 Akitas and 3 Chows and they are all unlicensed? This thing is starting to stink, call HazMat! At least they admit the 'high risk' dogs came from an atypical neighbourhood and I have a feeling it wasn't Cincinnati's version of Rosedale.
It seems that they had to keep looking until they got the numbers they wanted and the majority of offenses in the 'high risk' group which were traffic-related just weren't cutting it. Who includes traffic offenses or failure to license a dog when trying to predict criminality? I guess these busybodies do.
I suppose then that the 90% of people in Ontario who don't license their dogs are all hard-bitten crooks. Will this appear next? 'People Who Don't Buy Dog Tags Are Usually Criminals.'
It is also noteworthy that 'crimes against children' include 'allowing children to become unruly' (guess we're all guilty of that one) and 'failing to restrain child properly'. Now, obviously with the kinds of vehicles and the traffic we have now it makes sense to use a carseat and regulation is becoming more common but I'd hardly call not using one a 'crime', at most I'd call it foolish. It could be a result of poverty, those mandatory car seats are expensive and if you don't own a car and are just getting a lift from a friend, well.
There are no socioeconomic data to further qualify the results. It is known that crime overall and particularly violent crime is much higher among the poor than among the financially secure. It is also known that the poor are under greater police scrutiny and that their neighbourhoods tend to be less safe overall, partly because they are more crowded. Hamilton County, which includes Cincinnati, has a high crime rate. In 2004, Cincinnati ranked 8th in homicide, 2nd in robbery and 3rd in break-ins in comparisons among selected cities across North America.
Put another way, most people charged with some (but not all) of the offenses we see here are likely people of low socioeconomic standing. In fact, when you note the similar number of domestic violence convictions (HR=15, LR=11), as that is an offense that easily crosses socioeconomic lines, but then see a wide disparity in other offenses, it begs the question of whether the actual areas in the county or city were preselected as well. Those from more affluent areas are less likely to engage in break-and-enter, robbery, common assault and other crimes. They are less likely to be charged with having illegal drugs. They are less likely to have teachers calling social workers because of concerns about their children. They are less likely to have their neighbours calling them in to Animal Control or the police.
Is this because they are a better type of person? No, I think it is because the affluent live in their houses and cars more than in the neighbourhood so they are less in the public eye and are farther apart. They don't walk their dogs on downtown streets. The big reason, of course, is that they tend to be better educated and have the means to hire lawyers. Both of these qualities will cause the authorities to treat them with more deference. People who are aware of their legal rights and who have the money to hire lawyers are well protected from random harassment.
My point here is that if everyone were under increased scrutiny, I am certain that a lot more traffic tickets and other offense warrants would be issued across socioeconomic lines than we see now.
Did the investigators search for Staffordshire Bull, American Staffordshire and American Pit Bull terrier owners, or only the generic 'pit bull'? Purebred dogs are expensive and people who obtain them tend to take care of them. Was the exclusion of the three most commonly included purebreds in the 'pit bull' pot a deliberate or unintentional factor which further skewed the data?
I don't want to sound harsh, but in my opinion, anyone who has registered their dog as a generic 'pit bull' within the last ten years is not too swift. Criminals are also not too swift - in a US study conducted through the prison system to calculate the average hourly earnings of criminals, the figure was $3.00 per hour. Aside from the fact that you have to work alone, often at night and take tremendous risks, it really is true that crime doesn't pay. It's one field where the more successful you are, the less well known you are, starting with the justice system.
The most obvious reason to me for these results, aside from the huge number of 'pit bull' owners included in the study versus other types, is that, as we know, 'pit bull' owners are always being watched, are more likely to have complaints made against them and are more likely to merit attention from both Animal Control and the police.
If a 'pit bull' gets out of the yard, which happens to every dog owner at some time, it's a call to Animal Control, sometimes even the police. When one of my Brussels Griffons got out under the fence last year, a neighbour called to let me know he'd been wandering around my house for 15 minutes. If a 'pit bull' barks at another dog on the street or chases a cat, it's a call to Animal Control. So, right away, without using at least an equal number of 'low risk' or even subjects from other 'high risk breeds' in the study, you will have a biased outcome.
The only way, in my view, to have accurately studied this question, assuming you even wanted to do so, would have been to first check AKC registrations for the area to select only purebred dogs, then create a map (by zip code) of all of the just under 50,000 (two-year period) licensed dog owners and sort them accurately by breed (a Staffordshire Bull terrier or American Staffordshire terrier is not a 'pit bull'). Then you would distribute your study population evenly across zip codes, by breed. If you wished to include mixed breeds, which really demonstrates nothing if your goal is to single out a breed of dog and its owner as deviant, then you would put all mixed breed dogs (a 'Terripoo' is not a breed, nor is 'pit bull', 'hound', 'terrier', 'collie', 'poodle') together, then run the searches for citations and divide them by citation type matched to breed or type. Then, the names of all owners would be run through the police database to search for convictions which would then be matched to breed or the generic 'mixed breed' of dog owned.
You would, of course, still end up with biased results because approximately 80% of dogs are unregistered and therefore mixed breed pets but this could be corrected through either a weighting factor or a deliberate number choice, ie, 'we will look at every breed in the county to a maximum of 50 dogs and will also include 50 mixed breed dogs'. I feel the results would be more meaningful if 'mixed breed' were left out entirely and purebreds only were studied, assuming this type of 'what if' pilot investigation has any meaning at all, that is.
While this method would still profile only a portion of the dog owners in the county, it would be much more objective than what has been done here and would reveal more about dog ownership patterns in relation to antisocial behaviour, although as a friend pointed out,
"A robber is a robber, why does the kind of dog he or she owns matter? It's like determining what they watch on TV, what kind of clothes they wear or car they drive. It's a specious connection at best and can only be harmful to dog owners across the board, which I suspect is the intent."
I agree with my friend.
A small percentage of dog owners in Hamilton County, Ohio, a group categorized by offense, not participant, were considered socially deviant. We don't know whether breed alone made this group 'high risk' as there is no breed-by-conviction breakdown. We only know how many had bylaw infractions by 'breed' cited by Animal Control and we have percentages, not actual numbers for the rest.
The typical number we would expect to see in a jurisdiction, about 0.08%, or fewer than one in a thousand dog owners (regardless of breed) could be considered either criminal or incompetent or both, depending on the criteria used.
This study, in view of the unusual selection and more unusual exclusion criteria, the use of percentages instead of numbers, the disproportionate number of dogs of 'high risk' breeds (the so-called 'pit bull' at 153) compared with others (one Labrador retriever out of almost 9,000 licensed according to 2005 stats), the exclusion of locally popular breeds (German Shepherd dog, Am Cocker), the failure to enumerate the number of owners with repeat offenses and the failure to match by zip code, among other failings, makes me hope that its shelf life in our ever-watchful and honest media is a very short one indeed.
It also makes me wonder whether the police check was done first and was followed by the animal control citation check, if you know what I mean, but that's an unscientific way of thinking, isnt' it?
I imagine that there is some sort of desperate scramble going on to create some scientific evidence to support the banning and restriction of dog breeds (especially in Ohio, where Toledo's law was overturned on constitutional grounds and is being appealed by the dog-haters and others who persecute citizens there).
Well, this certainly isn't it. Back to the drawing board, folks.
The most obvious point of all - this 'research' took place in a wee corner of Ohio and just like all the other much-ballyhood US information, doesn't apply to Canada in any way. For example, in Ontario, available bite statistics show that there has never been a reported bite by the breeds erroneously dubbed 'pit bulls' by the government and the parrots in the media. A recent national survey in the US found that the average 'pit bull' owner is female, over 30, middle to upper middle class, employed and educated. I guess it depends who you ask and how you ask it.
How about keeping this junk science, like your pinhead dogcatcher Skeldon, in Ohio, guys. We really don't need the extra stigmatization right now. We have more than enough of it here in Dalton McGuinty's Ontario."
*Borchelt, P. L., Lockwood, R., Beck, A. M., & Voith, V. L. (1983). Attacks by packs of dogs involving predation on human beings. Public Health Reports, 98, 57-66
http://caveat.blogware.com
