The only variable that changed was prevalence of guns. The socio-economic and demographic makeup of the town didn't change, so there is no need to manually control for it. This was a rare situation where only the target variable changed in a very short period of time. All other variables remained the same. It should be a perfect situation for you to prove your point.
You haven't got a natural experiment so don't kid yourself. You've merely got convenient raw data. Similar convenient raw data can be used by the anti-gun lobby. Personally I find such unscientific approaches a proper nuisance, showing a contempt for the empirical process. We know that, except for the discredited work by Lott, the 'more guns=less crime' hypothesis is rejected time after time.
There are numerous variables required to understand crime. Basic criminology for you! Guns, however, are one. You've allowed personal bias, rather than the evidence, to dictate your position. This has led to error in your outlook and a stance inconsistent with empirical conclusions
Really? What is my "personal outlook?" You and your "studies," on both sides of the argument, start with a faulty premise, then proceed to "study" by excluding any data that does not fit the premise and...VIOLA!!!! Because you've excluded all the "faulty" data you think you've proven your premise. http://www.saf.org/LawReviews/McDowallAvailability.htm "Considering the variables one at a time, the results provide no evidence of a causal relationship between gun density and the total robbery rate. ...together these results indicate that gun density in a city is neither a cause nor an effect of the overall rate of robbery. On the other hand, however, gun density and the ratio of gun-to-nongun robberies have relatively large effects on each other. These effects hold in [Page 142] both the cross-lagged and cross-contemporaneous models, and in both models all effects are positive. This implies that gun density and weapons choice in robbery form a mutually reinforcing cycle, with increases in one variable leading to increases in the other. As gun density in a city rises, robberies are more likely to be committed with guns. In turn, as more of a city's robberies are committed with guns, the level of gun density in the city rises." In other words. If people are going to commit crimes and have a gun, they will use the gun but the presence of the gun has no impact on whether they commit the crime. If you want to make the argument for "gun control" of any sort this is the wrong one to make. It is as easily discounted as the pro-gun "guns prevent crime" argument.
The anti-gun lobby actually doesn't have such data available. Instead they are forced to use data muddled with numerous other changed variables. They are then forced to artificially control for the various other variables. Your previous arguments have shown that you will argue a point in the face of any amount of evidence, and absolutely refuse to ever consider changing your position, so I'll accept your vague claim that some mysterious and unnamed evidence supports you to be your acceptance of defeat. Till next time. . .
Yes it matters! I read an empirical study that said all anti gunners love it up the 6. Show me a study stating differently.
Already said: making comment based on personal perceptions, rather than valid literature review method. Happy to help you correct that though. Consider the paper you referenced (a vast improvement on your previous posts, so congrats!). Can you spot any methodological limitations? Are you convinced by the lack of control variables? Does the author really show robustness, in particualarly with regards omitted variable bias? I look forward to your response. A proper conversation on this topic, where there is properly conducted objective consideration of scholarly output, is far too uncommon on this sub-forum This is a nonsensical claim. There is no exclusion of 'faulty data' (unless of course you mean in terms of minimsing empirical bias). Instead there is a review of the literature and conclusion based on appropriate research methods. That the majority of research cannot reject the 'more guns=more crime' hypothesis is just factual. You'd know that if you had perused the literature. However, I don't want to put you off. Looking forward to your methodological reply
You've actually just said 'raw data is useful, but only when it agrees with me'. Note that you cannot refer to one single quasi-natural experiment published in the available scholarly research. Why do you think that is the case?
Thank you for the reference, I read the paper, didn't understand the process the authors used but I think I get the point. The summary helped: Okay, yer makes yer choices and yer lives with them.
The interesting aspect for me is that, whilst there is analysis that could be used to support the deterrence argument (typically theoretical mind you), it isn't employed because of two reasons. First, it involves review of the available analysis (a big no no for a lot of the pro-gunners). Second- and much more important as it isn't a sub-forum specific point- that evidence necessarily also describes the optimality of regulation. Something that is religiously avoided!
The shoddy statistical analysis is what rejects your hypothesis. Studies that fail to renormalize for cultural and societal effects, discount behavioral influences which are much greater than your six sigma allowable tollerance for errors. It is a poor hypothesis backed by poor analysis.
An interesting theoretical article is provided by Donohue and Levitt (1998, Guns, Violence, and the Efficiency of Illegal Markets, American Economic Review, Vol. 88, pp 463-467). Heres the abstract: In this paper, the authors analyze the determinants of the efficiency with which illegal markets allocate scarce resources. They develop a stylized model in which players compete for a fixed prize, with the winner determined by fighting ability. Efficiency in this context is determined by the amount of resources spent on fighting. Two factors affecting efficiency emerge from the model: lethality and predictability. Perhaps surprisingly, the use of more lethal mechanisms for resolving disputes does not have a clear impact on the social costs of violence. The intuition underlying this result is that, as the costs of losing a fight rise, the willingness to fight falls. The authors show that holding other factors constant, the resources spent on fighting are lowest when the cost of losing is either very low or very high (e.g., nuclear deterrence), but over a wide range of lethality levels, the overall social costs of fighting are fairly stable. A little long winded, but they sum it up nicely by paraphrasing the standard cliché: guns dont kill people, the unpredictability of guns kills people. Deterrence doesnt operate because there isnt an ex ante predictability of a winner. Those referring to deterrence would therefore have to refer to one man, one tactical nuke. I look forward to the change in tone from them!
Just shows how extreme you have to go with regards deterrence theory. OF course even then you're in murky waters. See, for example, what happens with MAD when a first strike capability is engineered
They do have a bearing for this thread! The paper potentially illustrates the absurdity of the deterrence arguments
A gun it the face is one hell of a deterrent. In some areas of the US (Guess you have not lived here) it is really almost necessary. You have no clue about some areas and what the potential is for home invasion is. Oh wait I am on ignore.
I'll say it again,tactical nukes have NO bearing on the right to bear arms...period What part of that is so hard to understand?
Can you show any evidence of deterrence effects at all, or do you just have "its been said" tabloidism?
