dichotomy Posted April 8, 2008 Posted April 8, 2008 This website has dog attacks (only for dogs of clearly identified breed by a professional) for the US and Canada 1982-2006, excluding dogs trained to attack or guard. Interesting how the number for pit bull terriers compares to that of other pit bulls, and that for rottweilers to rottweiler mixes, to any other dog. dog attacks by breed If I’m not mistaken the stats don’t have the entire population numbers of these breeds. I’d think these numbers would further magnify the real difference between breeds. E.g. if there are 100,000 Chihuahua’s with 1 death, and 100 pitbulls with 200 deaths, then Huston we have a problem. But still, the stats interesting and revealing.
Mr Skeptic Posted April 8, 2008 Posted April 8, 2008 If I’m not mistaken the stats don’t have the entire population numbers of these breeds. I’d think these numbers would further magnify the real difference between breeds. E.g. if there are 100,000 Chihuahua’s with 1 death, and 100 pitbulls with 200 deaths, then Huston we have a problem. But still, the stats interesting and revealing. Yes, I was looking for those too. As was John Cuthber. Best we found was numbers for the top ten breeds, and I didn't post it cause he already did. It seems that there should be something at the American Kennel Club. But its statistics are for breed popularity rankings, and doesn't seem to include pit bulls?
SkepticLance Posted April 8, 2008 Posted April 8, 2008 To MrSkeptic and dichotomy My earlier reference mentioned the percentage of pit bulls http://www.edba.org.au/courier.html For the time period they selected, 42% of all fatal dog attacks on humans in the USA were from pit bulls, and only 1% of the USA dog population were pit bulls. Fairly revealing, don't you think? To iNow, there is no psychoanalysis in my comments. I do not believe in it. Instead I refer only to the fact that you repeatedly deny something that is patently real. Whatever it is inside your head that induces you to do this is definitely beyond my comprehension, and I have no interest in any kind of psychoanalysis. And for the UK data you request, John has said he searched for it and could not find it. It may not exist. if the UK authorities never got round to collating that data, it is not available, no matter how often you ask for it. We do not need it. There is enough data elsewhere to show that pit bulls are the worst dog type in terms of killing people.
iNow Posted April 8, 2008 Author Posted April 8, 2008 42% of all fatal dog attacks on humans in the USA were from pit bulls, and only 1% of the USA dog population were pit bulls. Fairly revealing, don't you think? No. Not when your total population size is 3 dogs. And for the UK data you request, John has said he searched for it and could not find it. It may not exist. if the UK authorities never got round to collating that data, it is not available, no matter how often you ask for it. Then he should not have made the claim in the first place. Simple really. I am NOT being unreasonable. I am asking people to show enough respect for others as not post personal opinion as fact, but to instead support the claims they make.
SkepticLance Posted April 8, 2008 Posted April 8, 2008 iNow said "No. Not when your total population size is 3 dogs." Wrong on a number of accounts. The number 3 is human deaths from pit bull attacks per year, and comes from a 20 year population of 66, which is high enough to be statistically significant.
iNow Posted April 8, 2008 Author Posted April 8, 2008 iNow said "No. Not when your total population size is 3 dogs." Wrong on a number of accounts. The number 3 is human deaths from pit bull attacks per year, and comes from a 20 year population of 66, which is high enough to be statistically significant. No. I contend it's not. Why don't you prove to me and all of us all that it is statistically significant? Are you using a Pearson's r, or perhaps some sort of t-test, maybe a chi-squared? You want to convince me? Use data, not statements which are unfounded, unsupported, and fully conjectured. Further, 3 deaths per year and 66 deaths per 20 years is exactly the same number, so I'm not really sure why you've chosen such a comment to support your point that I was wrong in my suggestion. Six to one, half a dozen to another. John, You stated: "What is being said is that there are several more effective ways to achieve the goals for which you are arguing (safety of others), and that the ban will not achieve that end." No other means has been put forward and the ban in the UK has proved effective. Your point is without merit until you can support it with empirical evidence. In order to support your comment, you must: a) Show how many deaths per annum in the UK were caused by dogs prior to the ban on pit bulls b) Show how many deaths per annum in the UK were caused by dogs after the ban on pit bulls c) demonstrate that the difference is statistically significant. SkepticLance, You stated: iNow said "No. Not when your total population size is 3 dogs." Wrong on a number of accounts. The number 3 is human deaths from pit bull attacks per year, and comes from a 20 year population of 66, which is high enough to be statistically significant. Your point is without merit until you can support it by showing the calculations you used, the numbers used in those calculations, and the source of those numbers. If either of you are unable or unwilling to support your assertions in this way, then your approach has failed and your comments are baseless. You get an F in this class. Your basic premise is that despite the fact that owners are the responsible parties in the fatalities you've been trumpeting, you are incapable of restricting or punishing the behavior of those negligent owners. So, you instead wish to take a route which will not achieve the end you seek, which will not accomplish the stated goal for which you're arguing, but which will appease those who are afraid and convinced that the propaganda is true. Instead of treating the source of the problem, you are making a weak case for treating the symptom, and you are arguing in favor of a ban of one type of dog without having proven that this will have any statistically significant impact on society, or the small handful of deaths you are using in support of your pro-ban position. I implore you both to support your assertions above, or to retract them. Further, I ask that you both either a) make better arguments, b) acknowledge that you are wrong in your assertions (or incapable of supporting them with evidence), or c) walk away from this thread. I've tried to show you enough respect to support my position with logic, reason, and fact. I've offered mutliple lines of reasoning for my position. You have not countered any of my arguments, and, as demonstrated above in this post, you've failed to support your own.
SkepticLance Posted April 8, 2008 Posted April 8, 2008 iNow I am staggered by how far you will go to support an untenable position. First, you said the number was 3 dogs. It was not. It was 3 human deaths, and that was the tiniest part of the total statistics offered. To quote a number 3 like that is cherry picking of the worst type. I suggested that the true number was 66. This is true, but is also only part of the total. In fact, the 66 represented about a third of the total human deaths, which means the total number is approaching 200. 66 humans killed by pit bulls, compared to about half that killed by the next worst (rottweilers), from a total of almost 200. Those numbers are definitely large enough to be significant. According to the Endangered Dog Breeds Assn. in Australia, only 1% of dogs in America are pit bulls. This means that the deaths caused by that dog type run at 30 to 50 times that of the average, even allowing for expected error. Are you going to say that this is not significant? iNow, you really are avoiding reality.
iNow Posted April 8, 2008 Author Posted April 8, 2008 iNow I am staggered by how far you will go to support an untenable position. <...> that is cherry picking of the worst type. <...> Those numbers are definitely large enough to be significant. <...> Are you going to say that this is not significant? <...> iNow, you really are avoiding reality. Now not only have you resorted to making this personal with me (instead of focussing on facts and information), you are also apparently using a subjective definition of the word "significant," but trying to pass it off as a statistical one. http://en.wikipedia.org/wiki/Statistical_significance The significance level of a test is a traditional frequentist statistical hypothesis testing concept. In simple cases, it is defined as the probability of making a decision to reject the null hypothesis when the null hypothesis is actually true (a decision known as a Type I error, or "false positive determination"). The decision is often made using the p-value: if the p-value is less than the significance level, then the null hypothesis is rejected. The smaller the p-value, the more significant the result is said to be. In more complicated, but practically important cases, the significance level of a test is a probability such that the probablility of making a decision to reject the null hypothesis when the null hypothesis is actually true is no more than the stated probability. This allows for those applications where the probability of deciding to reject may be much smaller than the significance level for some sets of assumptions encompassed within the null hypothesis. http://www.graphpad.com/articles/interpret/principles/stat_sig.htm First, define a threshold P value before you do the experiment. Ideally, you should set this value based on the relative consequences of missing a true difference or falsely finding a difference. In practice, the threshold value (called a) is almost always set to 0.05 (an arbitrary value that has been widely adopted). Next, define the null hypothesis. If you are comparing two means, the null hypothesis is that the two populations have the same mean. In most circumstances, the null hypothesis is the opposite of the experimental hypothesis that the means come from different populations. Now, perform the appropriate statistical test to compute the P value. If the P value is less than the threshold, state that you "reject the null hypothesis" and that the difference is "statistically significant". If the P value is greater than the threshold, state that you "do not reject the null hypothesis" and that the difference is "not statistically significant". You cannot conclude that the null hypothesis is true. All you can do is conclude that you don't have sufficient evidence to reject the null hypothesis. I state yet again: iNow said "No. Not when your total population size is 3 dogs." Wrong on a number of accounts. The number 3 is human deaths from pit bull attacks per year' date=' and comes from a 20 year population of 66, [b']which is high enough to be statistically significant.[/b] I said 3 deaths per year. You said 66 deaths in 20 years. It's EXACTLY the same number, except you are trying to inflate it's perceptual salience. Again, your point about significance is using some arbitrary subjective definition, yet you are trying to pass it off as statistical. This claim is completely without merit until you can support it by showing the calculations you used, the numbers used in those calculations, and the source of those numbers. If you are unable or unwilling to support your assertions in this way, then your approach has failed and your comments are baseless. Now, are you going to continue with the uninformed, pseudo intellectual, psychoanalysis of me and my posts, or are you going to address the questions put forth?
John Cuthber Posted April 8, 2008 Posted April 8, 2008 iNow, Once again I am going to point out that it was you who first made the statement that the ban won't work and you have yet to provide any evidence for it. If you think evidence is so important, why don't you provide some or retract that statement? BTW, "I said 3 deaths per year. You said 66 deaths in 20 years. It's EXACTLY the same number, except you are trying to inflate it's perceptual salience" OK I wil send you $3 and you can send me $66- it's EXACTLY the same number. If you don't know why a running average over 20 years is more likely to represent reallity then I wonder why you asked about statistical significance. Incidentally, you are now asking for statistical evidence for smething you already accepted "I conceded long ago that I won't argue with you that based on the numbers shared this type causes more deaths than others." May I ask what changed?
iNow Posted April 8, 2008 Author Posted April 8, 2008 iNow,Once again I am going to point out that it was you who first made the statement that the ban won't work and you have yet to provide any evidence for it. Fine. My point was intended to suggest that there are more effective alternatives. I've already outlined many of those alternatives, however, I will do my part to move this conversation forward. In my estimation, a ban would not work for reasons pointed out already: a) the problem is not specific to one type of dog b) the problem is in the way the dogs are cared for and raised c) the object is to prevent deaths by dog, so measures applicable to all dogs are more appropriate d) the ban will simply displace the problem, not solve it. It's treating symptoms, not causes. Now. It's your turn. Will you please show data which describes the number of deaths by dog per annum in the UK before the ban on pit bulls and the number of deaths by dog per annum in the UK after the ban on pit bulls? BTW,"I said 3 deaths per year. You said 66 deaths in 20 years. It's EXACTLY the same number, except you are trying to inflate it's perceptual salience" OK I wil send you $3 and you can send me $66- it's EXACTLY the same number. Really? Do you really think I am missing something here? My point stands, and is valid. Come on, now. Incidentally, you are now asking for statistical evidence for smething you already accepted "I conceded long ago that I won't argue with you that based on the numbers shared this type causes more deaths than others."May I ask what changed? Actually, no. What I want is for SkepticLance to support his claim that 3 deaths per year is a statistically significant effect. These are his words, not mine. He needs to show total number of pit bulls in the population and how 3 deaths relates to that total. Alternatively, he needs to show all forms of death, and relate the numbers of deaths from pit bull to that number. It's about perspective. He's simply asserted that 3 deaths per year is statistically significant (or, 66 deaths in 20 years, whatever). ...but I've resigned myself with the understanding that SkepticLance has no such data to offer in support his contention, and that his assertion is nothing more than unfounded conjecture. You, however, have a much easier task, and you still refuse to move this conversation forward by addressing it. Numbers prior and post ban. That is all. If the ban is "effective" as you've claimed, it will show easily in the data.
SkepticLance Posted April 8, 2008 Posted April 8, 2008 iNow When I did basic practical statistics as part of my degree course, our professor asserted than a population of 20 was sufficient to gain a statistically significant result, when using a T test. The population we are dealing with is 66. The number we are comparing it to is that of deaths caused by the next worst breed (Rottweilers) which is about half. The difference is statistically significant. Your action is taking 3 as the number was the cherry picking I was talking about. The number 3 did not even appear in the official statistics. You calculated 3 as the average per year over 20 years. It is NOT the proper number. In trying to use it, you are cherry picking in the worst possible way. iNow, I do not think you are stupid. In terms of pure IQ, you are probably smart enough. However, I think you are displaying a stubborn pride. Because you are not stupid, you already know you are wrong, but you continue to push an untenable argument purely out of this pride. As I said before, if you want to continue the argument on the basis of concerns about personal liberty, or some similar argument, I have no problem. That is your right to do so. But to try to distort the facts when you are on a science forum is not really acceptable.
John Cuthber Posted April 8, 2008 Posted April 8, 2008 My guess is that he meant that if pitbulls happen to kill 3 people in one year and the next "most dangerous" dog kills 2 then that's not statistically significant. OTOH, (and I haven't done the maths to check this) 66 deaths over 20 years compared to (I think it was) 140 deaths over 20 years may well be significant from a breed that's far less than 1/3 of the population. I think it is but I'd need to check. The null hypothesis is that some other breed is at least as likely to kill a human as a pitbull is. Now we don't have good data for that but if the figures are 42 % and 1% for deaths caused by pitbulls and dogs that are pitbulls then the expected figure for pitbulls would be roughly 1 or 2 deaths over 20 years (about 1% of the total of about 160 deaths caused by dogs (based on 42% of them being 66 deaths)) The actual observed number is 66. 3 in one year might be a fluke, but roughly 3 a year for 20 years is pretty unlikely. I contend that it's a probability of less than 1 in 20 so it's statstically significant at the p=95% level. If you really insist I can do the stats, they don't frighten me, but I'm prety sure it's a waste of time particularly since I don't have a stats package on this computer. A breed that's only 1% of the population really shouldn't cause 42% of the deaths. Saying that such a result is statistically significant is hardly an "unfounded conjecture", it's common sense. And, since it really was you who said a ban would be effective it falls to you to prove it not me. Posting that the problem isn't specific to one breed of dog doesn't cut it. The problem of one breed of dog being disproportionately represented in the dogs that kill figures is specific to that type of dog. Other dogs are also poorly raised and trained. I contend that the reason pitbulls are responsible for more deaths must, at least in part, be due to their nature. Other methods of dropping the toll may well also be apropriate but I think the pitbulls, (one relatively uncommon breed responsible for a lot of the deaths) count as "low hanging fruit". I think it's a reasonable place to start. Even if all we do is displace the problem, that might help. Unfortunately some people get a pitbull because it "looks hard". Without that incentive they might chose to do something else (tattoos perhaps) or they might try to find another "hard" dog. Even if they do, I think they are likely to end up with a dog that's less of a risk to the population in general than the pitbull they currently have. I tried to find pre and post ban death rates- the data don't seem to be available. I have said that before and it's clearly not a refusal to provide data. You really ought to look at the difference between "I wont" and "I can't". Even if the raw numbers were available they wouldn't tell the whole story. What about total numbers of dogs owned? Say the number of deaths has gone up by 10%- it looks like the ban failed but if the number of dogs has gone up by 20% then the number of attacks per dog has fallen. The UK stopped licensing dogs decades ago so there are no definitive stats for the number of dogs. IIRC crime against the person has fallen in recent years in the UK but crime against property has risen. That might well mean that more people are getting dogs or that more of them are getting big dogs. We simply don't know and I don't think we ever will. That doesn't detract from the fact that a ban works because a dog that's not there doesn't do any harm.
iNow Posted April 8, 2008 Author Posted April 8, 2008 Again with the uninformed, pseudo intellectual, psychoanalysis of me and my posts... When I did basic practical statistics as part of my degree course, our professor asserted than a population of 20 was sufficient to gain a statistically significant result, when using a T test. The population we are dealing with is 66. The number we are comparing it to is that of deaths caused by the next worst breed (Rottweilers) which is about half. The difference is statistically significant. The point you are consistently missing is that your population is far greater than 3... far greater than 66... I really don't care what number you wish to toss about, as long as you are consistent on the time frame involved. What your professor was describing was within population significance, and how you can show "within group" effect significance when using a population of only 20. You are, however, misapplying the knowledge your professor tried to share with you. Your population is not 66, at least not for the claims you are trying to make here in this thread. The claims you’ve been making imply that your data applies to ALL pit bulls, so THAT is your population. Your population is all pit bulls, and your within group effects do not extrapolate AT ALL to this larger population. Not to pit bulls, not to all dogs, not to anything outside of the population you've selected... 66 dogs in 20 years. That is where your t-test (which you have not even performed) would apply. You've biased your selection, you have a very limited population sample, and you are suggesting that somehow these 66 dogs involved in killings are a representative cross section of the entire pit bull population. This is where your approach fails. You're only viewing those who have attacked and killed, and not the entire dog type, yet you're suggesting that your within group data (of the group which has killed) somehow applies to the greater population of all pit bulls. If you wish to show some statistically significant trend within that population of 66 dogs, then you are free to do so. However, your attempts to use this biased and non-representative sample to describe the entire pit bull population is what I am calling into question. How about this time you don’t respond again with your uninformed, pseudo intellectual, psychoanalysis of me and my posts. I tried to find pre and post ban death rates- the data don't seem to be available. I have said that before and it's clearly not a refusal to provide data. You really ought to look at the difference between "I wont" and "I can't". This distintion is of little relevance. Both apply equally to the issue I've been trying to bring to your attention. You concede that you can't provide numbers, yet at the same time you are trying to suggest that fatalities caused by all dogs are down as a result of the ban. You can't have it both ways. Of course if there are no pit bulls then there are no pit bulls to cause death. I'm not completely retarded, as you keep implying. The issue is that the ban's purpose is to decrease attacks, and yet you've freely admitted that you have no numbers about how many attacks are occurring. With the facts as stated above, you have zero ground on which to stand, and null support for your suggestion that the "ban has proven effective."
SkepticLance Posted April 8, 2008 Posted April 8, 2008 To iNow Of course the population is not 66. However, I was countering your assertion that there was no statistical significance since the number we were dealing with was 3, and that is a far more silly statement. The actual number of pit bulls would be of the order of 500,000. I am going by memory, so that may not be 100% accurate. I read a while back that the US dog population was something like 50 million. Since pit bulls are 1%, we are left with 500,000, plus or minus a smaller number since I cannot really be sure that I am remembering it right. But that is the correct order of magnitude. However, the numbers we are dealing with are numbers of human deaths caused by dog attack. Not the number of pit bulls in the US. So the statistical population is, as reported by John, about 140. Of that, 66 are deaths caused by pit bulls. Statistically very significant. iNow, I am certain in my own mind that you are aware of reality. You are arguing a position that you must realise is untenable. I have to assume it is only pride that prevents you from admitting the truth. Why don't you show your scientific credentials are actually good, by admitting that the figures show that pit bulls, as a type, kill humans at a rate way out of proportion to their numbers, compared to other breeds?
ParanoiA Posted April 8, 2008 Posted April 8, 2008 SkepticLance, enough with the condescending defeatist rhetoric. His position is quite tenable.
SkepticLance Posted April 8, 2008 Posted April 8, 2008 If iNow wants to use a good argument, such as the right to choice - ie. the liberty argument, his position is tenable. If he want to deny that pit bulls kill more people than other types of dog, his position is clearly untenable - it is unscientific to deny the data.
iNow Posted April 9, 2008 Author Posted April 9, 2008 And, since it really was you who said a ban would be [in]effective it falls to you to prove it not me. What I stated was a single point within a post which was offered to summarize the thread thus far. I was recapping the issues demonstrated by posters already in this thread, not presenting a new assertion. This was ONE of the points in my summary, and I said EXACTLY this: What is being said is that there are several more effective ways to achieve the goals for which you are arguing (safety of others), and that the ban will not achieve that end. I was not arguing that the ban would be ineffective, I was summarizing how this point had already been made... right here... in this very thread. You then responded: No other means has been put forward and the ban in the UK has proved effective. The first point was patently false, as other alternatives had been put forward. I demonstrated this clearly in my next response. The second point was a new assertion, a new claim being made in this thread... a new claim which YOU made. I asked you to support that claim, and you have conceded that you cannot. It's time to move past that issue, as I accept your concession that your claim cannot be supported. If he [iNow] want to deny that pit bulls kill more people than other types of dog, his position is clearly untenable - it is unscientific to deny the data. I have not made any such denial though, so my position remains tenable and scientific. What exactly is your point if not simply to misrepresent mine?
SkepticLance Posted April 9, 2008 Posted April 9, 2008 iNow I re-quote your statement in post 311 "The deaths of those 66 people had zero to do with type of dog, and everything to do with the negligence of the person in their charge." You have repeatedly suggested that pit bulls were not a greater danger than other dogs. The quibble about statistics can only be to push that message. Yet the data is clear cut. Tell me you agree that pit bulls, by their genes, are more prone to killing humans than other dog breeds and we can end this silly argument. I am not suggesting all pit bulls. We both know that the danger comes from a small minority. However, that is true for all breeds of dog, and the pit bull minority that is potentially lethal to humans is a larger minority than other dog breeds. Tell me you agree, and we can stop wittering at each other.
iNow Posted April 9, 2008 Author Posted April 9, 2008 iNow I re-quote your statement in post 311 "The deaths of those 66 people had zero to do with type of dog, and everything to do with the negligence of the person in their charge." You have repeatedly suggested that pit bulls were not a greater danger than other dogs. <sigh> I said the problem was with the owner, not the dog, and more specifically, not the type of dog. Further, I have not suggested what you said I've suggested. I have been suggesting that a ban is an absurd approach in an effort to resolve such a minor problem. I have been suggesting that it is a problem, and that better solutions exist, that a ban is a extreme precautionary measure not likely to achieve the desired end. I've have been suggesting that your position lacks data in support, and that the data you do have is questionable as pertains to dog type identification. I've suggested that you and John have fallen prey to the propaganda, and that your approach ignores the fact that aggressive attacks by pit bull are the exception, not the rule. I really don't know how much clearer I can make this for you. Tell me you agree that pit bulls, by their genes, are more prone to killing humans than other dog breeds and we can end this silly argument. Tell you I agree? Are you serious? I'm not a liar, and I don't agree with you, so I won't be telling you any such thing any time soon. Frankly, there isn't enough data to address your request here. Any conclusions made are rooted in unsupported premises. Right now, your contention that "pit bulls, by their genes, are more prone to killing humans" is a COMPLETELY UNSUPPORTED CONJECTURE, you have no data in support of such a position, and you are flailing about using nothing more than rhetoric. So, I don't know if "pit bulls, by their genes, are more prone to killing humans." The funny thing is, neither do you. So, WTF? Tell me you agree, and we can stop wittering at each other. You've got some real audacity to suggest that I should back down from my argument, when I'm the one making logical and consistent claims. Your own approach is nothing more than repetition of the same point which itself has been shown misframed, questionably supported, and nothing more than rhetorical handwaving. "Tell me you agree." You're pretty funny. It's not exactly like you've got me locked in a torture cell with electrodes strapped to my testicles. For the love of Thor, man, remove your cranium from your colon. You're making a weak argument, with one single point which does not extrapolate to the action against an entire type of dog which you have been trying to defend. Your request for me to tell you "I agree" reminds me a bit of the question regarding whether or not Hillary should ask Barack to be her VP.
SkepticLance Posted April 9, 2008 Posted April 9, 2008 iNow I am having a hell of a task nailing down exactly what your position is. You have made so many comments indicating strange ideas. The data clearly shows many more human deaths from pit bulls than any other breed, and as human deaths per million dogs, it is massively more, with the possible exception of Rottweilers, which is another story. The UK government was so convinced of this, they banned the breed. In NZ, it has been discussed in parliament with a ban being placed on the agenda, so sure were the parliamentarians. Australia I know less about, but I know that such a ban has been discussed by various authorities. To suggest that this massive disparity towards harmful attacks by pit bulls compared to other types of dog is caused by owners rather than dogs, seriously lacks credibility. All types of dog breed have some good owners and some bad. If bad owners were the problem, we would see it with other breeds also. Pit bulls were bred as killers of other dogs. And they are known as common killers of other dogs even on the street. The data also shows they kill people at a rate much higher than other breeds. If you believe all these killings and maimings of both other dogs and of humans is only because of bad training or bad handling by owners, then you have a very strange view. The data is not propaganda. Governments do not fall prey to propaganda, since they have very expert advisers helping them to avoid such, and the UK, NZ and Australian governments all clearly believe that pit bulls are especially dangerous. Remove the veil from your eyes and face reality. Pit bulls are a special danger because of their INHERENT tendency to violence. They were bred for it, which puts it into their genes. I accept that only a small minority are a real hazard to human life, but a small minority is enough to kill 66 people in 20 years. I am not so concerned about arguing for a ban on the breed. I think it would probably be the correct thing, but I accept that the decision to do so is subjective. I am more concerned about the fact that you repeatedly deny reality, and push a view that is plain wrong. It is your denial of the facts about the breed that especially keeps me arguing.
UsaForever Posted April 9, 2008 Posted April 9, 2008 The dogs have been here since the Late 1880's right at 100 yrs and no problems until recent. Does that kinda show its the way the people have become and how life has changed in our world? I would say these dogs are as much of our History as Apple Pie. The dog for its self is of great pride of being "American" To stand and fight or protect till death. I Love my dogs just as much as i love my family and if one of my family hurt another person i would expect them to be hurt just the same. In the day of dog fighting they would kill any dog and its offspring that was people aggressive and breed more selective to remove that trait. We are now at a time where dog fighting is illegal since "1976" and still the numbers of people who owned these dogs were of the type that breed for fighting and selective breeding for that type. Some people of today have these dogs for what they did in the past and stood for but they have not the knowledge or the care for the dog. Just the idea of something mean and tough that they may think is what they are. The dogs have been here since the Late 1880's right at 100 yrs and no problems until recent. Does that kinda show its the way the people have become and how life has changed? I could see a Ban of any dog that hurt people without cause but these same people who have breed a bad dog would also breed a different dog to do the samething. I truley belive its the type of people who breed these dogs for there type of life. I do apologize for seeming wild in my words but its late since reading all the posts since my last comment
iNow Posted April 9, 2008 Author Posted April 9, 2008 I guess it’s time for a new approach… again. So, with that said… Let’s spot check these numbers, as there are a lot of them being thrown around. It’s really quite important to put the facts into perspective. According to the National Safety Council: Numbers of Deaths Due to Injury, United States, in the year 2000 there were 26 deaths caused by attacks from dogs of all breeds. Out of those 26 deaths, 8 of those were caused by pit bulls. This means that pit bulls caused 30% of deaths by dog in the year 2000. While there is no completely accurate consensus on dog populations, the common numbers shared in those which are available to us show that pit bulls comprise an estimated 30 – 40% of the of the entire dog population. This percentage is drawn from a collection of various shelter and rescue statistical data. This percentage, where pit bulls account for 30-40% of all dogs, suggests that the percentage of the entire dog population composed of pit bulls is well aligned with the data on deaths by dog type referenced above. Pit bulls accounted for 30% of deaths by dogs of all types, and pit bulls make up 30-40% of the dog population. There are an estimated 53,000,000 dogs in the US. While I did reference above that pit bulls compose 30 – 40% of the entire dog population based on statistical data from shelter and rescue sources, let’s go extremely conservative and work with the assumption that they compose only 10% of that 53M population. I think this is more than fair, and should not pose any issue to those who support the pro-ban position. Agreed? This approach, estimating pit bulls to be only 10% of the entire dog population, suggests that there are approximately 5.3 million pit bulls in our society. 5,300,000 pit bulls… on the low end. In 2000, 13 pit bulls were involved in 8 fatal attacks. When you do the math, you discover that this represents roughly ONE dog out of every 407,000 (if we use the 13 number) or ONE dog out of every 662,000 (if we use the 8 number). Breaking this down into percentages, we find that, as a percentage of the pit bull population, the measure demonstrated previously in this thread to be the most appropriate when discussing a ban which is specific one dog type, and using a conservative estimate that pit bulls compose only 10% of the entire dog population… On the high end, we see that only 0.000245% of all pit bulls are involved in attacks. On the low end, we see that only 0.000151% of all pit bulls are involved in attacks. That means that, as a percentage of the population of this dog type, using conservative and (very likely) under representative values of the pit bull population, there are three zeros to the right of the decimal place and before the first non-zero number in front of the percentage sign. The percentage is in the 10-thousandths place… Zero point zero zero zero two four five… percent… On the high end. Now, despite the seeming “nano” quality of the issue, we have people who argue vehemently for a total ban, using rhetoric like “brutal,” “vicious,” “savage,” and “murderers.” It is estimated that 5,000,000 dogs per year are killed in shelters. As described above, pit bulls compose roughly 30-50% of the entire shelter population. Since there is such fear, propaganda, irrationality, and dissemination of complete misinformatoin about this dog type, pit bulls are also less likely to be considered for placement than any other breed. It does not seem unreasonable to propose that at least 25% of those dogs in shelters are euthanized (killed) due to an inability to find them a home. This means that we humans kill, on average, 1,500,000 pit bulls every year. Humans kill, as a low estimate, one point five MILLION pit bulls every year. On the high end, pit bulls kill 8 humans every year, and 3 humans every year on the low end. I think each of us should consider ourselves lucky that pit bulls don’t have such “advanced” and “mature” societies where they organize public stonings… erm… crucifixions… erm… racial profiling… erm… I mean… bans on animal type…, otherwise, we humans would be in a world of hurt. I am just being silly with this particular point, but you have to admit, we sure do seem to see a very clear asymmetry on this issue… Back to the serious points in this post, the only way to summarize all of this is: For every one pit bull which is involved in a killing, there are literally hundreds of thousands which are not. Let’s not throw the baby out with the bathwater. The numbers clearly don’t suggest that this is a phenomenon specific to one type of dog, nor do they suggest that violence is somehow inherent in pit bull genetics. . According to the CDC and National Safety Council, in 1992, the breed most involved in fatal attacks was the Rottweiler, not the pit bull… yet another weakness in the pro-pit bull-ban position. As has been repeatedly brought up in this thread, the most accurate explanation is that humans were at fault in those small handful of unfortunate deaths which did occur, and that humans are the source of the problem. The suggestion that “since we can't ban stupid people, we have to ban the dangerous dogs” is not the approach a mature and responsible society should take, especially since the numbers above show the dogs are not inherently dangerous. Treat the cause of the problem, not the symptom. Oh… and we could always talk about liberties and arbitrary attempts to remove them, but I first wanted to counter the nonsense about the precentages which has been shared in this thread, using actual numbers myself, before I decided to try making my first “tenable” argument. Enjoy.
Mr Skeptic Posted April 9, 2008 Posted April 9, 2008 To iNow: could you share the source(s) for your estimate of the pit bull population? The data I've seen so far (but I think it is for purebreds only) seems to imply that pit bulls are a miniscule portion of the population.
John Cuthber Posted April 9, 2008 Posted April 9, 2008 "You concede that you can't provide numbers, yet at the same time you are trying to suggest that fatalities caused by all dogs are down as a result of the ban. You can't have it both ways. Of course if there are no pit bulls then there are no pit bulls to cause death. I'm not completely retarded, as you keep implying." Yes, I do keep saying that the dogs that are not there don't kill anyone. I have been saying that for some time. That's what I meant when I said a ban is effective- it's a truism. On the other hand you chose, for some reason to interpret is as meaning that over a fairly long period during which a whole lot of other things happened, the overall number of deaths would go down. YOU said that- not me. The fact that I don't have numbers to support your interpretation of what I said isn't my responsibility. On the other hand, while you igonred the fact that you had raised the issue of the effectiveness of a ban in the first place, you insisted that I was in some way responsible for finding the evidence about it. You asserted that a ban wouldn't work. I asked you to prove it and you came up with some ideas that I refute. Once more, I ask you to support your assertion with evidence, or withdraw it. Speaking of evidence I see this has just popped up "While there is no completely accurate consensus on dog populations, the common numbers shared in those which are available to us show that pit bulls comprise an estimated 30 – 40% of the of the entire dog population. This percentage is drawn from a collection of various shelter and rescue statistical data. " It seems to be an assertion without citing a single reference. It also runs into the problem that, if as I contend, these dogs are less safe than other breeds, then they will be disproportionately represented in animal shelters, pounds and such. A lot has been said about pitbulls getting bad press. If they are then that too would mean more of them being dumped by their owners (and, just to establish some common ground here, I should point out that I think such behaviour is reprehensible). That would mean that a lot more of them would show up in the place where you found 30% or so. That suggests that the population has rather less than 30% pitbulls and the odds go against them again. Your data supply begs the question. BTW, in the summing up you seem to have overlooked a bit "For every one pit bull which is involved in a killing, there are literally hundreds of thousands which are not. " However, for every dog that kills there are literally millions of other dogs that don't. It's the fact that one number is roughly 10 (or 50 or whatever) times bigger than the other that matters here. If the question were about forbidding dogs then the fact that pitbulls kill more than their fair share wouldn't matter. If the question is about banning pitbulls then the fact that they are disproportionately represented among killer dogs is an essential part of the case. Another way of sumarising it would be "If you get a pitbull you are something like 50 times more likely to have a death on your hands than if you get another breed." By not including that fact in your summary you are ignoring the data. Hardly to be expected on a scientific site.
iNow Posted April 9, 2008 Author Posted April 9, 2008 could you share the source(s) for your estimate of the pit bull population? Sure. As I mentioned, I took the average of several shelters and rescue operations who shared statistics, and I brought together information from about 30-35 of the most reputable ones I could find who seperated the data in this way. I used the following search criteria, but no longer have the specific links I chose when aggregating the informatoin: http://www.google.com/search?hl=en&q=dog+shelter+population+statistics Please note, in all of my calculations that I used a number less than 1/3 the average amount so as to ensure my sample did not inflate, and in fact under represented, the true populatoin numbers. Another way of sumarising it would be "If you get a pitbull you are something like 50 times more likely to have a death on your hands than if you get another breed." Can you show the math of this conlcusion, since only one in roughly every 600,000 pit bulls is likely to be involved in an attack? You know what? Scratch that. Never mind. It's not even relevant. You continue to misframe the issue. You have yet to prove that risk of violence is breed specific, and the numbers I've shared STRONGLY argue against that point. You are suggesting that correlation means causation, but it does not. Risk of harm to others has nothing to do with type of dog. Risk of harm to others is contingent only on the actions of the person in charge of the animal. Your continued suggestion that risk of death is dog type specific is rooted in logical fallacy and misframing of the issue. You cannot prove causation using only correlation, as you clearly know. You have not demonstrated that risk to others is dog type specific, you have only shown a correlation. Based on the tiny ... actually "nano"... percentage likelihood that a pit bull will be involved in an attack, the data strongly supports my contention that risk of attack is an owner specific phenomenon, and implies that you are approaching this issue in an academically dishonest fashion.
Recommended Posts