ATTS (American Temperament Test Society)


I think this is very interesting the percentage rate for dogs that passed where high for the breeds that are on the current UK DDA

TOSA 100.0%

Yes with the Dogo Argentino and Tosa they didnt test many be these dogs are not common at all, so you need to bare this in mind.
This is how popular breeds did that do not face being banned

Breeds being debated by the government if these breeds should be added too the DDA

Interestingly this is what the British Bulldog percentage that passed this temp test

Here is the link to what this Society is all about

It just goes to show ALL dogs cannot be reliable with humans but the breeds been branded as devil dog and dangerous have scored higher than most!


I don't think it's the temperament issue that's the problem. It's the fact that if they are nasty or they have a go the damage they inflict can be pretty horrendous. As you say no dog is 100% reliable, they only have to have an off day but that could be the day they kill someone.
Personally, most of the small breeds have probably be ruined by their owners letting them get away with murder so I don't think that is an accurate indication of the breeds 'temperament' rather an indication of how poorly trained they were!
However, I would like to see this kind of temperament testing in the UK.


The thing is Becky to me this shows that a. Not one breed is 100 percent reliable with humans, BUT the breeds considered unreliable and dangerous too the public are scoring high percentages compared to popular breeds never facing the DDA because of Image.
I think this is why the DDA and BSL does not work, there are no REAL dangerous dogs, only owners!

But Becky I do understand you point that if a Pitbull or Stafford was to attack a child (god forbid) it would do alot more damage quicker than a Yorkshire Terrier


The Atts test statistics are not particularlr reliable because the samples taken of the breeds vary greatly..

example = SEALYHAM TERRIERS, one dog tested =100% reliable.

If we are looking at Atts statistics some gundog breeds do much better than many bull breeds , here are just a couple
English Cocker Spaniels 93%
Flat coats 91%
Welsh Springer Spaniels 100%

Statistics can mean what you want them too..look at the high number of bull breeds in these fatal attack statistics


Again I suppose it;s how you look at it. Fair enough large breeds are going to be responsible for all the deaths but I'd be interested in dog bites in general because I would bet your bottom quid that the small breeds would be high at the top, particularly the westie in my opinion lol!


Minihaha could what you have just said about the ATTS not go for this too?

But i do understand your point, but where it really counts Minihaha are the breeds branded and then tested to show on mass these dogs are more reliable than some other breeds not branded as dangerous(which is the point im trying to make)

STAFFORDSHIRE BULL TERRIER They tested 61 52 passed 9 failed 85.2%
AMERICAN PIT BULL TERRIER 542 tested 456 passed 86 failed 84.1%

compared to popular breeds
BORDER COLLIE 230 tested 184 passed 46 failed 80.0%
MINIATURE SCHNAUZER 103 tested 81 passed 22 failed 78.6%

None of the dogs i above are dangerous but when tested the percentage showed that dogs considered dangerous did better to those breeds who would never be considered dangerous!

I dont think the ATTS is perfect, i think by testing the same amount each breed would have been better, but i can see its point and aim!


Interesting stuff here, and I agree statistics can be made to say anything really which is why I'm more concerned with the actual death quota's counted and the facts (however unsavory) are there for us all to see; from

human DBRF during the past 20 years. Pit bull-type dogs and Rottweilers were involved in more than half of these deaths.

Whilst I agree that all dogs can be dangerous, especially in the wrong hands, and that many bites are frequently administered by much smaller breeds, the sad fact is that when these breeds do attack the damage is colossal. I know there are many issues to be sorted out regarding 'Pit Bull type' and so on but, I think we all agree any death is one too many and all dog owners need to be educated and responsible for their dogs.


Hayley I don't know what point you are trying to make here, the statistics you have posted mean very little as far as I am concerned, how can you make a comparison when the number of dogs tested in each breed varies from 1 to over 1000. For a reliable comparison you would need to take an equal number of each breed .

If you want to look at statistics look at the ones I posted on fatalities from dog bites in the USA in a 10 year period,
page 2 here

this shows by far the highest number of fatalities were from pitbull types, 66 deaths in ten years, collies just 6 and no fatalities from Schnauzers Mini or otherwise .


A few more statistics for you Hayley, this time from the American Law Society

In one study it was reported that 42% of all dog related killings of human beings in the United States are caused by Pit Bulls attacks, yet Pit Bulls constitute only 2% of all dogs. 70% of those mauling deaths were of children.
According to a large federal government study conducted "only" a third of the fatal dog attacks in the United States. This study also cited the disproportionate threat these dogs pose to children. Children, according to the study, are the most vulnerable victims in dog attacks, with those under the age of 14 accounting for 42 percent of all dog bite injuries. Most of the mauled victims are between the ages of five and nine.

It just goes to show what these child friendly dogs are capable of doesn't it.

I still believe it is down to the individual circumstances of each incident but there is no getting away from the fact that when a pitbull bites it really can cause damage.


Some interesting stuff, thanks for the links.

I do think Hayley, your first link, the test is biased against breeds that have origins in guarding. Fighting breeds are naturally going to have quick recovery from stressfull situations and show more friendliness towards strangers. These breeds are also much more confident off their own territory. A necessary trait for fighting breeds but not so much found, particularly in the Flock Guardians.

Also herding breeds would be at a disadvantage with the guntest. These breeds have super sensitive hearing and most I think would fail on this. Unlike the fighting breeds you list, which would be bred to be bombproof.

And Mini's point. It's very hard to take seriously, stats that show 100% pass rate with only one dog tested for some breeds.

But probably the most significant part for me is the lack of any testing for behaviour towards dogs. Ie. there is no evaluation of within species aggression. I think if this were included we would see very different stats.

In general though, it's probably expected that fighting breeds show low aggression to humans as this is a necessary trait for a fighting breed but, as others have already pointed out, it's when their innate fighting qualities are directed at humans that serious damage is done and the reason why Pitbulls feature high in death statistics.