To say two things are correlated is not to say that one or the other variable is causal =>
_______________________________________________________________
"Correlation does not imply causation is a phrase in science and statistics that emphasizes that a correlation between two variables does not necessarily imply that one causes the other."
http://en.wikipedia.org/wiki/Correlation_does_not_imply_causation
_______________________________________________________________
CutePuppy: How about illegal immigration and crime rate? There has been unarguably more illegal immigration while the crime rate ostensibly dropped. Conclusions?
Same as with the pot laws => 1. There's been a positive correlation between the two since the mid-1990s. 2. You cannot say there is a causal relationship. 3. Further investigation is needed to draw any conlusions.
Do you agree?
_______________________________________________________________
Now, I'll be glad to discuss other points you raised, such as causes of falling crime and what drug policy should be. But first, I want to get cleared up what should be a simple point => When I say there a correlation, I am not making the case for causation. I am making a case against the claim legalizing pot would cause crime to rise significantly. With CA's violent crime rate falling by half since 1996, it seems a highly dubious proposition.
1. There's been a positive correlation between the two [immigration and falling crime] since the mid-1990s.
1. Yes, that's essentially what the logical fallacy Cum hoc ergo propter hoc describes.
In other words, if you can't prove causation, your entire argument falls apart, i.e., correlation without causation is coincidental, meaningless and irrelevant, as I have shown in many examples in my previous post.
2. Your argument doesn't even show correlation, because there is nothing in the data set of crime statistics that has a variable that says "marijuana." It's as simple as that. Straight line from point a to point b has no correlation to straight line from point x to point y.
Otherwise, I can make thousands of statements that "correlate" to the crime data which have no relationship to the crime data, except in my imagination - see few examples in my first post.
Really, check the statistics books/blogs on this, because you are confused on what correlation means, let alone why causation is essential for your premise to be even marginally arguable.
And, just for yourself, check again the technology change argument (though I don't claim causation) which has been much more relevant and important in crime statistics than whether the MJ laws have been relaxed (which is destroying the first argument that the "War on Drugs" has failed) or not - they are simply not present in the data (i.e., not a variable in the data, which it can be correlated with).
Maybe this resource on validity and variables can help: Research Methods