What I love about those two reports (I read the 2005 one earlier) is how they skew the figures by working out the percentages based on number of registered dogs. Everyone knows that people stopped declaring there pitties to council or lying about the breed when the threat of seizure started to become a real problem. So, even though there are breeds responsible for many more actual attacks than pit bulls, pit bulls come out with a higher per capita percentage rating. Lets face it, they want to bring in the legislation so they'll skew the results anyway they can.
I'm sorry but I don't understand your logic. And it isn't "skewing the figures", it is providing a more accurate depiction of the data. Even if they listed the breeds without numbers of dogs (i.e. not representing the prevalence of the breed within the total canine population) the pit bull would still come in 3rd, 4th, and 7th in those tables in the first report. Therefore, even if ALL breeds were identically distributed, you would be justified in saying the pit bull was more dangerous than most according to the statistics.
Representing the rate of attack makes MORE sense. If you had only had two breeds of dog in Australia, A and B, and there were 50 000 of A, and 100 of B, and in one year there were 500 attacks by A and 100 by B, which is the most dangerous breed? Clearly we need to work out per capita attacks to account for the difference of the breed's representation in the population. By your logic it should be listed as A being the most dangerous due to the gross number of attacks being highest. This would be misleading, and although what I have presented is an argument in the extreme, I'm merely trying to illustrate that per capita attacks are a more relevant measure than gross numbers of attacks.
Statistics are fundamentally flawed as the report of dog attack relies on what a person thinks the bred of dog is, lots of dogs are unregistered and this is for a variety of reasons.
Also, in regards to misidentification, I do not think that in the majority of dog attacks the dog attacks some one on a street and runs away, with only their opinion on what the breed was becoming the identifier. If you look at the second study, I think, you can see the actions taken after the attacks. Sometimes the dog was put down, other times penalties were imposed etc. Clearly the victim knew where the dogs were located in order to have these actions taken, and upon actions being taken the investigating body would be made aware of the breed with certainty, either by visual identification or identification by the owner or identification by a veterinarian etc. So, given that in around 70% of attacks some sort of formal action was taken, we can be reasonably sure of the breed of the dog being correct. The other 30% either no action was taken or it was not certain what action was taken. Even if we assume that in all 30% of cases this was due to no dog being identifiable, we can at least be sure that 70% of the statistics are reliable. What about the other 30%? We would have had to take part in the report to know that. But still, I think that any pit bull owner who claims that the breed is not more dangerous than the average dog is absolutely kidding themselves. It is almost irrefutable.
That being said I do not support a ban on pit bulls. I think that their ownership needs qualification to ensure they don't fall into the hands of irresponsible owners.
Sorry
kaotikjezta
I just got what you meant about the population of pit bulls not being accurately represented due to people not registering them. This is unfortunate, but without an accurate gauge on just how many unregistered dogs there are out there working with registered dogs is probably the only accurate way that data can be compiled. Also, what about the people who misrepresent their pit bulls as staffordshire terriers etc? On one hand this may take away from the total number of pit bulls, thus making their per capita attacks higher (assuming all dogs misclassified did not cause attacks), or alternately, if pit bulls are indeed responsible for more attacks per capita (and the misclassified dogs attacked some one), it could be harming the staffordshire terrier's per capita attack rate.