Cloverlief wrote: WildCat wrote: Cloverlief wrote:
WildCat wrote:The contention that more guns = more murders is complete, utter bullshit and easily shown to be false. Besides my Illinois example Mexico has one of the lowest rates of gun ownership on the planet - and one of the highest murder rates.
No the contention is that less guns, less murder by guns. And that guns one and only purpose is to kill. Let's eliminate that.
Already disproven by Mexico. Very few guns, very high murder rate.
You're just spouting bullshit, and easily debunked bullshit at that.
Did you forget this is a skeptic's forum?
You seem to have done so. You and Ed seem to have confused this for the asshole and fallacy forums.
And again for the idiots in the back (that would be you and Ed), Correlation and Causation. Look it up.
Weeeeellllllll ... yeah correlation NEQ causation. However many models are correlative and it depends what questions are being asked and how the models are constructed.
If there were a strong relationship between guns and suicide one would expect there to be a strong correlation particularly when you have repeated measures across nations, cultures and socio-economic conditions. You might do some sort of ANOVAR on the data presented with those variables included but, on the surface, I would expect some new variable (or composite variable) to explain the bulk of the variance.
We did stats in my first company and what we used to tell the new people was "you can't ignore what you know". Meaning that if you got a positive price elasticity (ie price goes up, sales volume goes up) you damn well better be sure of that finding because otherwise you'll be laughed out of the room (n.b. Contac took off after they increased the price back in the day. Price and "effectiveness" are somehow linked in consumers minds). Where was I ... oh yeah.
Saying Correlation NEQ causation is a bit like the ignorant saying "you can't yell fire in a crowded theater" as an example of how the first amendment is limited. If you say the latter to a lawyer he will laugh at you. If you say the former, without qualification, the applied statistician will look at you with a raised eyebrow and a look of infinite superiority. Consider yourself looked at.
In the present case, there is no strong relationship between suicide and guns. Neither positive nor negative. If there were, you would see it. To claim that the IS such a relationship, given the data presented, it seems to me that the onus would be on the claimant to suggest HOW such a relationship is being masked. Because, to torture another quote, you really can't prove a negative. Which is what you are asking for, no? You are saying "prove that there is no relationship".
Your trite example of ice cream and temperature or whatever it was is, in case you haven't noticed, a positive, strong relationship, not an essentially random one. If your example were relevant it would apply if there were a strong relationship between guns and suicide. But then you'd be arguing against yourself. Do you see that?
I've actually got a leg up on you lot because I have actually fooled around with this sort of cross country data for years (well, not continuously) and whatever relationships there may be between guns and homicide rates are weak. The banal finding (which is really too stupid to spend much time on) is that guns are implicated in gun deaths. <yawn>. The issue at hand is loss of life.