|
Although this thread does not function under the same strict guidelines as the USPMT, it is still a general practice on TL to provide a source with an explanation on why it is relevant and what purpose it adds to the discussion. Failure to do so will result in a mod action. |
On January 21 2014 11:32 Zaqwe wrote:
As humans we can only ever observe correlation. Did you know the link between lung cancer and smoking was proven with correlations? The data not just in that paper but all data available shows consistently that gun ownership reduces crime.
In the absence of any compelling information otherwise I feel I am on solid ground with my statements. I am very willing to see new information though.
You know how they were able to prove a causal link there? They found out
Get this, this is fucking amazing it'll blow your mind.
They were able to prove that it's not the tumors that make you smoke, but the smoking that makes the tumors
*Crowd goes FUCKING WILD*
Science bitches it works. In the absence of a secondary potential explanation, you can establish a causal link fairly easily, although getting a humongous amount of data is great.
+ Show Spoiler +Also they followed people for many decades which is what you need to establish causal links by gathering a lot of evidence on ONE "specimen" (over multiple specimens). The same needs to be done with countries and states, otherwise you're comparing apples and oranges and you're not even close to a causal link.
|
If correlation doesn't prove causation then how do they know one caused the other at all?
Strong correlations that can't be falsified do in fact prove causation. Without a time machine there's no way to observe causation directly. Only correlation.
|
On January 21 2014 11:40 Zaqwe wrote:
If correlation doesn't prove causation then how do they know one caused the other at all?
You can make fair assumptions, given sufficient data and whatnot.
Strong correlations that can't be falsified do in fact prove causation.
When you have a strong correlation where you don't know which variable affects which, of if a third unknown variable is responsible for the correlation, then you don't have grounds to declare a causal link. If they believe that they had grounds for it, they would have done it in the article. They didn't because they know that's not hoe it works.
Plus when the correlation is established not within one context but using many people or countries with different contexts, it's even less close to establishing a causal link.
In this case, you don't know if gun control affects criminality OR if criminality triggers the authorities to implement gun control. So by assuming that it goes one way or the other, you make a leap of faith that no credible researchers have made. So you're some guy on a forum who's comfortable making huge declarations that's not supported by science.
To the quote "correlation does not imply causation", Zaqwe says "no, I do".
|
They didn't conclude any causation because that wasn't the goal of the paper. The paper was simply debunking the claim that reduced gun ownership would reduce crime.
The negative correlation between gun ownership and crime rate is robust, repeated in many locations, remains over time, and hasn't been falsified despite many well funded groups who would like to do so. It's safe to conclude there is a causative link.
|
On January 21 2014 11:49 Zaqwe wrote:
They didn't conclude any causation because that wasn't the goal of the paper. The paper was simply debunking the claim that reduced gun ownership would reduce crime.
The negative correlation between gun ownership and crime rate is robust, repeated in many locations, remains over time, and hasn't been falsified despite many well funded groups who would like to do so. It's safe to conclude there is a causative link.
They didn't conclude on any causation link because they know that their data is weak... they say it themselves. They literally say no solid conclusion can be drawn because of the weakness of the available data. If it was safe to conclude that, they would have said it. Nobody in their right mind would conclude on a causal link with scraps of data coming from various sources. They would be laughed out of academia.
The argument that "it wasn't the goal of the paper" is ridiculous. The strongest argument they could conclude on is "there is no apparent correlation" and you flip the thing around and you say there's a correlation the other way around that's so damn strong I can make up the "direction" of the causal link out of thin air and the existence of a causal link from the sheer weight of the data in this study, which the authors describe as lackluster.
So not only is the quantity of data is made evident by the authors, the direction of the causal link is your gift to science. You just came up with it.
|
I don't make my statements based on one single paper. The evidence is overwhelming that relaxing gun control laws results in lower crime rates. That paper is one drop in the bucket.
![[image loading]](http://img191.imageshack.us/img191/64/gquc.th.png) ![[image loading]](http://img842.imageshack.us/img842/905/ifju.th.png) ![[image loading]](http://img850.imageshack.us/img850/7069/5qx9.th.png)
* Since the outset of the Florida right-to-carry law, the Florida murder rate has averaged 36% lower than it was before the law took effect, while the U.S. murder rate has averaged 15% lower.
* Since the outset of the Texas right-to-carry law, the Texas murder rate has averaged 30% lower than it was before the law took effect, while the U.S. murder rate has averaged 28% lower.
* Since the outset of the Michigan right-to-carry law, the Michigan murder rate has averaged 4% lower than it was before the law took effect, while the U.S. murder rate has averaged 2% lower.
http://www.justfacts.com/guncontrol.asp#right-to-carry
|
Those laws happen to strike at the exact same time criminality was dropping across the board in US and Canada and you're still willing to establish a causal link between state policy and criminality in those states.
Cheers.
|
Ok I feel like I must say something here. In my experience as a biochemist, and in my discussions with my colleagues, I believe that the culture and environment an individual is brought up in is MUCH more influential to his behavior than the genetic variations among individuals. In other words, if you want to place the cause of why an individual did something, it might be more conclusive if you looked at his social upbringing and socioeconomic background rather than analyzing specific stretches of his DNA.
It's not that I do not trust all genetic research. But we must be careful when we try to explain human behavior (such as crime) through variations in our DNA. These differences account for physical differences, such as how much of a certain protein we produce, its shape, its efficiency, etc, but to say a specific race is more genetically prone to "do something"?
hm...
There are certain mechanisms and pathways we understand very well. A causes B causes C which then produces D. But anyone arguing for many alleles interacting with each other in specific ways to produce a specific human behavior has his work cut out for him.
Just my thoughts, hope it makes sense.
|
On January 21 2014 12:03 Djzapz wrote:
Those laws happen to strike at the exact same time criminality was dropping across the board in US and Canada and you're still willing to establish a causal link between state policy and criminality in those states.
Cheers.
Note the crime rate in those states dropped faster than the rest of the nation.
|
On January 21 2014 12:06 Zaqwe wrote:Show nested quote +On January 21 2014 12:03 Djzapz wrote:
Those laws happen to strike at the exact same time criminality was dropping across the board in US and Canada and you're still willing to establish a causal link between state policy and criminality in those states.
Cheers. Note the crime rate in those states dropped faster than the rest of the nation.
Yeah especially in Texas where criminality dropped faster before the law was enacted. Today, Florida's average criminality is much higher than the US average and Michigan's thing doesn't seem to have had a very measurable impact in comparison to the national average and very much is in the margin of error.
It's also not surprising that as criminality went down across the board, it would go down faster where there was a lot of it. See: Texas, your own graph.
And again I'm saying this as a guy who doesn't believe gun control to be effective.
|
On January 21 2014 12:04 Titusmaster6 wrote:
Ok I feel like I must say something here. In my experience as a biochemist, and in my discussions with my colleagues, I believe that the culture and environment an individual is brought up in is MUCH more influential to his behavior than the genetic variations among individuals. In other words, if you want to place the cause of why an individual did something, it might be more conclusive if you looked at his social upbringing and socioeconomic background rather than analyzing specific stretches of his DNA.
It's not that I do not trust all genetic research. But we must be careful when we try to explain human behavior (such as crime) through variations in our DNA. These differences account for physical differences, such as how much of a certain protein we produce, its shape, its efficiency, etc, but to say a specific race is more genetically prone to "do something"?
hm...
There are certain mechanisms and pathways we understand very well. A causes B causes C which then produces D. But anyone arguing for many alleles interacting with each other in specific ways to produce a specific human behavior has his work cut of for him.
Just my thoughts, hope it makes sense.
Well now we are just retreading old territory. I welcome any evidence you have which supports your beliefs. All the evidence I have seen shows that even though environment plays some role, genetics play a far larger one.
"[T]his graph is taken from a study of more than 11,000 people. You can see that, while increasing SES lowers the risk of incarceration only a little bit, increasing IQ lowers the risk sharply."
![[image loading]](http://img19.imageshack.us/img19/4935/3xuf.th.png)
http://www.sciencedirect.com/science/article/pii/S0191886911000912
|
^ measures of IQ and SES are confounded variables
|
On January 21 2014 12:15 sam!zdat wrote:
^ measures of IQ and SES are confounded variables
Elaborate.
|
"IQ" is not an innate thing that exists independently of everything, "IQ" is just an operationally defined concept = "whatever this IQ test measures" and part of what it measures is SES
|
IQ is mostly heritable (i.e. genetic). And the data you are replying to shows that "while increasing SES lowers the risk of incarceration only a little bit, increasing IQ lowers the risk sharply."
|
that's just a thing you believe because it suits you.
the point is, try to be very skeptical about what charts based on correlating things with what are, essentially, ad-hoc operationalized measures can tell you about the world. that's all. statistics are just statistics, they don't tell you anything about the way that power is structured in a society, certainly not in any direct way
|
You're being way too cryptic.
What, if anything, is actually wrong with that graph?
|
I'm not being cryptic! there's nothing wrong with the graph, it's just some dumb graph. what's wrong is the way that YOU are THINKING about the graph, which is that you think that IQ is a real thing and that something as simple as that can explain anything at all about the real world, which (as should be quite obvious with a moment's critical reflection) it can't, really, precisely because it is so simple and the world is (unfortunately) not.
tldr watch dat reification yo
edit: in other words, whenever you are thinking to yourself - see! it is so simple, here is a simple graph which explains The Thing - that is when you should be most afraid about what you do not know. which is why the more evidence from the internet someone can garner to support their claim, you should believe them LESS and all this stuff people to toss around about "proofs" and "evidence" and stuff in forums like this is quite literally counterproductive
edit: anyway, get back to whatever you're arguing about, I'm not sure what it is, but whatever it is, that graph is really not going to help
edit: @below yes like this for example, the problem is that the real world is like one big fucking confounded variable :O
|
As far as I can tell, the correlation between high gun ownership and low crime rates is just an expression of the fact that in cities, where crime tends to be higher, there is less gun ownership per capita then in rural areas, where crime tends to be lower. This indicates to me that crime is function of population density, rather than gun ownership.
|
On January 21 2014 12:51 sam!zdat wrote:
I'm not being cryptic! there's nothing wrong with the graph, it's just some dumb graph. what's wrong is the way that YOU are THINKING about the graph, which is that you think that IQ is a real thing and that something as simple as that can explain anything at all about the real world, which (as should be quite obvious with a moment's critical reflection) it can't, really, precisely because it is so simple and the world is (unfortunately) not.
tldr watch dat reification yo
edit: in other words, whenever you are thinking to yourself - see! it is so simple, here is a simple graph which explains The Thing - that is when you should be most afraid about what you do not know. which is why the more evidence from the internet someone can garner to support their claim, you should believe them LESS and all this stuff people to toss around about "proofs" and "evidence" and stuff in forums like this is quite literally counterproductive
edit: anyway, get back to whatever you're arguing about, I'm not sure what it is, but whatever it is, that graph is really not going to help
edit: @below yes like this for example, the problem is that the real world is like one big fucking confounded variable :O
You realise Stephen Jay Gould has been discredited, right?
"IQ/g is best single predictor, mental or non-mental. IQ/g usually predicts major life outcomes better than does any other single predictor in broad samples of individuals."
Gottfredson, L. S. (2006). Social consequences of group differences in cognitive ability (Consequencias sociais das diferencas de grupo em habilidade cognitiva). In C. E. Flores-Mendoza & R. Colom (Eds.), Introducau a psicologia das diferencas individuais (pp. 433-456). Porto Allegre, Brazil: ArtMed Publishers.
|
|
|
|
|
|
EPT![[image loading]](http://img191.imageshack.us/img191/64/gquc.png)
![[image loading]](http://img842.imageshack.us/img842/905/ifju.png)
![[image loading]](http://img850.imageshack.us/img850/7069/5qx9.png)
![[image loading]](http://img19.imageshack.us/img19/4935/3xuf.png)
