Killing people is the best solution.

@OCCOM BILL,

OCCOM BILL wrote:

But demonstration of cause-and-effect is NOT necessary to establish correlation. Correlation can exist independent of cause and effect as demonstrated in your own idiotic pirate example.

I never claimed otherwise.

OCCOM BILL wrote:

Stop behaving like a child and admit the correlation exists and is demonstrated by the DOJ stats.

Again, your graph is simplistic to the point of absurdity. I do not, and will not, find that graph to be compelling in any way.

0 Replies

@DrewDad,

DrewDad wrote:

OCCOM BILL wrote:
I don't understand how you can pretend the obvious correlation isn't there. I don't understand how you can STILL be pretending this, while trying to make the argument that the correlation that is there doesn't prove causation. I don't understand why you think you can move the goal posts from "suggests causation" to "proves causation" without anyone noticing. You are a VERY dishonest man.

The fact that you think it is obvious does not require the rest of the world to agree.

You put murders from all states together, regardless of whether they have the death penalty. How can you not understand this glaring flaw? Your murder rates are per capita, but you use absolute numbers for executions.

I well understand the model is not perfect, and have already stated as much (more dishonesty on your part). Inclusions of States without the Death Penalty are unfortunate; but wouldn't necessarily affect the measure of change by the stopping and restarting of executions in the country as a whole anyway. While the multitude of other factors cannot be accounted for (which is why NO ONE has claimed "proof of causation'), the correlation either exists or it doesn't and would do so regardless in a vacuum... so this flaw isn't really all that important. If indeed there is no deterrent relationship between the murder rate and executions; it would be equally undetectable in an All States or just the sample of DP states. Conversely, if a relationship exists (if we could eliminate the other factors); it would exist whether the non-DP states were included or not.

The fact the chart doesn't reduce the execution numbers to per capita is largely immaterial: The effect (or lack thereof) on the graph would be roughly the same anyway (altering only as population increase rates deviated), even while it reflected the decrease in executions per capita. This too is an incidental flaw, which in a vacuum would not prevent the correlation from being measured. Outside of the vacuum; neither of these flaws is as compelling as a multitude of other factors which collectively make proving causation from this correlation impossible. The problem you have; is that you can't get it through your thick skull that NO ONE is trying to claim otherwise. That argument is DrewDad Vs. his shadow.

DrewDad wrote:

Furthermore, you've done nothing to attempt to factor out confounding variables. Have you looked at the rest of the crime rate? Does the murder rate correlate to that? Have you looked at unemployment, income or changing demographics? How about divorce rates? The effect of hate-crime laws? Gun control?

All compelling reasons against "Causation", not "Correlation". I would add the end of Alcohol prohibition and the beginning of the War on drugs (not to mention 3 strikes laws) as compelling alternative explanations as well (against causation). However; none of these things disprove "Correlation". Learn the definition of correlation and admit your obvious error.

DrewDad wrote:

Your graph is simplistic to the point of absurdity; you seem to be the only person who finds it compelling.

Your anti-Death Penalty stance lends you a very sympathetic audience here on A2K, DrewDad. Don't attempt to use that as an ad populum argument, however, because not only would that be a fallacy in itself; the majority of your countrymen disagree with you.

1 Reply

@OCCOM BILL,

Bill, if you compute the correlation, and show your work, then I will admit it exists.

It will even admit now that sometimes one line goes up and the other line goes down. And vice versa.

Yippee.

Without the math, we can't know if this is more or less than coincidence.

Without knowing that, we don't know if the numbers correlate or not.

2 Replies

@DrewDad,

DrewDad wrote:

Bill, if you compute the correlation, and show your work, then I will admit it exists.

It will even admit now that sometimes one line goes up and the other line goes down. And vice versa.

Yippee.

Without the math, we can't know if this is more or less than coincidence.

Without knowing that, we don't know if the numbers correlate or not.

I recreated the exact same graph using the DOJ numbers myself, years ago, when I had this discussion with Joe. They are spot on. If we still had the search feature A2K used to have; I would have pointed you to it already. I am not going to repeat the exercise just so you can finally admit the truth and then remind me that it doesn't change your greater argument anyway, which it wouldn't. It would be decent of you accept my word, as I am certainly no liar. Alternately, anyone could A. Find the discussion between Joe and I. B. Crunch the numbers themselves or C. Accept the truth for what it is, because it isn't going to change any minds regarding causation anyway.

2 Replies

@DrewDad,

DrewDad wrote:

Without the math, we can't know if this is more or less than coincidence.

Even with the math; you won't be able to know if this is more or less coincidence. Neither will I be able to know.

0 Replies

@OCCOM BILL,

In the post you are quoting he's not disputing the accuracy of the numbers but saying that you haven't calculated the correlation coefficient.

1 Reply

@OCCOM BILL,

I never thought that you invented the numbers.

My contention is that the method is so flawed that any correlation you find would be meaningless. That's the entire point of the pirates graph.

A side complaint that I have is that you post a graph, and insist that a visual inspection demonstrates a correlation; I've seen too many misleading graphs to take such assertions at face value.

0 Replies

@OCCOM BILL,

OCCOM BILL wrote:

DrewDad wrote:

The found a correlation, investigated, and found an actual causal relationship.
No, they didn't. The case against smoking is built almost entirely on correlation. The causal relationship is assumed because the correlation is so strong and pronounced… but there is no direct evidence, so your contention is a total sham.

This is false Bill, and has been for a while. In 1996 causation was established between benzo[a]pyrene (found in tobacco smoke) and damage to the p53 gene. Smokers with lung cancer show the exact same damage to the p53 gene as the experiment produced and the tobacco industry dropped their opposition to the claim that smoking causes cancer that same year.

http://www.ncbi.nlm.nih.gov/pubmed/8832894

1 Reply

@Robert Gentel,

Robert Gentel wrote:

OCCOM BILL wrote:

DrewDad wrote:

The found a correlation, investigated, and found an actual causal relationship.
No, they didn't. The case against smoking is built almost entirely on correlation. The causal relationship is assumed because the correlation is so strong and pronounced… but there is no direct evidence, so your contention is a total sham.

This is false Bill, and has been for a while. In 1996 causation was established between benzo[a]pyrene (found in tobacco smoke) and damage to the p53 gene. Smokers with lung cancer show the exact same damage to the p53 gene as the experiment produced and the tobacco industry dropped their opposition to the claim that smoking causes cancer that same year.

http://www.ncbi.nlm.nih.gov/pubmed/8832894

At which point; they had more proof. This doesn't change the simple fact that the correlation between smoking and lung cancer had long implied causation since well before the discovery, does it Robert? Doesn't it, if fact, confirm that the dogmatic believe of the flawed assumption is and always has been wrong? No one here has disputed the fact that correlation doesn't prove causation; but it most certainly can and does, frequently, imply causation to various degrees of probability.

1 Reply

@OCCOM BILL,

OCCOM BILL wrote:

This doesn't change the simple fact that the correlation between smoking and lung cancer had long implied causation since well before the discovery, does it Robert?

I didn't say anything about that Bill. For how much you toss around intellectual honesty as a rhetorical device you should show a modicum of it yourself.

Let's recap:

DrewDad wrote:

The found a correlation, investigated, and found an actual causal relationship.

OCCOM BILL wrote:

No, they didn't. The case against smoking is built almost entirely on correlation. The causal relationship is assumed because the correlation is so strong and pronounced… but there is no direct evidence, so your contention is a total sham.

Yes they did, Bill and you are moving the goal posts into other parts of your argument instead of admitting that despite the strength of your convictions you were wrong.

Terms like "intellectual honesty" have real meaning, and aren't just a rhetorical bludgeon. A fundamental part of intellectual honesty is admitting where you are wrong.

1 Reply

@Robert Gentel,

Robert Gentel wrote:

In the post you are quoting he's not disputing the accuracy of the numbers but saying that you haven't calculated the correlation coefficient.

No, I haven't, and I don't see the purpose in trying; because I don't believe myself that the strength of the correlation can be accurately measured in any useful way. I posited that it exists, can be seen in comparing the numbers and that I don't believe this to be completely coincidence. It would require no agreement with my conclusion to acknowledge that the observable correlation exists. You are joining him in moving the goal posts if you are asking for anything more than evidence that the correlation is observable. If the numbers are accurate (and they are), and they reflect an inverted correlation over the period of moratorium on Executions (they do), then you have to concede the correlation exists... whether or not you think it has any bearing on causation or not. DrewDad has refused to do even that; what say you?

2 Replies

@OCCOM BILL,

OCCOM BILL wrote:

No, I haven't, and I don't see the purpose in trying; because I don't believe myself that the strength of the correlation can be accurately measured in any useful way.

Fair enough, but you should try saying this to DrewDad instead of pretending he's questioning the accuracy of the numbers themselves and saying that they can be fact checked for accuracy.

It's pretty obvious that your interpretation is what he's disputing and not the numbers themselves so why do you keep trying to push the argument into the question of the veracity of the numbers themselves? What's the point of making a big deal about the numbers coming from the DOJ when it's your interpretation of the numbers that was criticized?

Quote:

It would require no agreement with my conclusion to acknowledge that the observable correlation exists. You are joining him in moving the goal posts if you are asking for anything more than evidence that the correlation is observable.

This is a logomachy that centers around the colloquial use of "correlation" and the statistical term meaning correlation coefficient. Yeah, I think you can eyeball colloquially defined "correlation". No, I don't think you can eyeball the correlation coefficient.

1 Reply

@Robert Gentel,

Robert Gentel wrote:

OCCOM BILL wrote:
This doesn't change the simple fact that the correlation between smoking and lung cancer had long implied causation since well before the discovery, does it Robert?

I didn't say anything about that Bill. For how much you toss around intellectual honesty as a rhetorical device you should show a modicum of it yourself.

Let's recap:

DrewDad wrote:
The found a correlation, investigated, and found an actual causal relationship.

OCCOM BILL wrote:
No, they didn't. The case against smoking is built almost entirely on correlation. The causal relationship is assumed because the correlation is so strong and pronounced… but there is no direct evidence, so your contention is a total sham.

Yes they did, Bill and you are moving the goal posts into other parts of your argument instead of admitting that despite the strength of your convictions you were wrong.

Terms like "intellectual honesty" have real meaning, and aren't just a rhetorical bludgeon. A fundamental part of intellectual honesty is admitting where you are wrong.

Then where is yours, Robert? You know perfectly well the discussion you've entered featured DrewDad denying that correlation can imply causation; and that the only reason the inconsequential error you pointed out was brought up was in an example to attempt to disprove the misguided notion that the phrase is some kind of absolute law of science. Or do you? Please tell me: Do you think the correlation between smoking and lung cancer had long implied causation since well before the discovery or don't you?

Btw: Accusing me of intellectual dishonesty for an accidental, inconsequential error, is uncalled for. Your bias is showing.

1 Reply

@OCCOM BILL,

Bill, you're missing the point that even if a correlation exists, it is meaningless.

Since it is meaningless either way, then I don't need to get into an argument about whether or not it's there. I don't care if it is there or not, because it doesn't change the outcome.

0 Replies

@OCCOM BILL,

OCCOM BILL wrote:

Then where is yours, Robert?

Here you go again, using it as a rhetorical device.

Quote:

You know perfectly well the discussion you've entered featured DrewDad denying that correlation can imply causation; and that the only reason the inconsequential error you pointed out was brought up was in an example to attempt to disprove the misguided notion that the phrase is some kind of absolute law of science. Or do you?

Bill, I came in and read from the end backwards. I came across you being a blow hard about a simple fact that you were wrong about and I corrected it. What was intellectually dishonest about that? It seems to mean whoever disagrees with you when you casually toss it around like this.

Quote:

Please tell me: Do you think the correlation between smoking and lung cancer had long implied causation since well before the discovery or don't you?

I don't know. The causation has been fairly well established ever since I've researched it so I never had to think about whether or not the correlation implied causation.

And I think this whole "implied" argument is a bit of a silly one too with a lot hinging on the weight one puts on the implication so I haven't been following it.

Quote:

Btw: Accusing me of intellectual dishonesty for an accidental, inconsequential error, is uncalled for.

I didn't call you intellectually dishonest for your error, I called you intellectually dishonest for trying to rope me into other arguments instead of acknowledging that you'd been wrong when you called DrewDad's claim a "total sham".

Quote:

Your bias is showing.

Perhaps. I'm not sure what "bias" that would be but that's certainly a possibility somehow. But an alternate theory I'd like to propose is that you use terms like "intellectual dishonesty" and "bias" as rhetorical devices and are a bit quick on the trigger with them when disagreed with.

1 Reply

@Robert Gentel,

Robert Gentel wrote:

OCCOM BILL wrote:
No, I haven't, and I don't see the purpose in trying; because I don't believe myself that the strength of the correlation can be accurately measured in any useful way.

Fair enough, but you should try saying this to DrewDad instead of pretending he's questioning the accuracy of the numbers themselves and saying that they can be fact checked for accuracy.

It's pretty obvious that your interpretation is what he's disputing and not the numbers themselves so why do you keep trying to push the argument into the question of the veracity of the numbers themselves? What's the point of making a big deal about the numbers coming from the DOJ when it's your interpretation of the numbers that was criticized?

A. It wasn't that obvious to me. When you asked a clear question; I answered it.
B. If he wants to dispute my interpretation, of the correlation; he first needs to admit the correlation exists, no? I have freely admitted all along that my interpretation of the correlation is only my opinion and that it was perfectly reasonable to consider it coincidence. DrewDad's ridiculous insistence that no correlation exists has prevented us from moving on to discussing what it does or doesn't mean. You are now sitting in judgment of me, because of his refusal to accept the obvious.
C. The DOJ numbers are the most relevant response to challenges to the illustrated correlation. Had DrewDad restricted himself to disputing the potential for causation; I would agree pushing those numbers would be pointless. Not so, when he's disputing correlation.

Robert Gentel wrote:

Quote:
It would require no agreement with my conclusion to acknowledge that the observable correlation exists. You are joining him in moving the goal posts if you are asking for anything more than evidence that the correlation is observable.

This is a logomachy that centers around the colloquial use of "correlation" and the statistical term meaning correlation coefficient. Yeah, I think you can eyeball colloquially defined "correlation". No, I don't think you can eyeball the correlation coefficient.

And which do you think is typical on a message board? Is anyone going to mistake A2K's random poster's opinions on the Death penalty as a strict scientific discussion? What percentage of us do you think are even qualified for such a discussion? I further don't see how any scientist could fail to see the obvious flaw in the assumption, "Correlation doesn't imply causation" anyway. So much of our medical assumptions rely on precisely that.

1 Reply

@OCCOM BILL,

OCCOM BILL wrote:

I further don't see how any scientist could fail to see the obvious flaw in the assumption, "Correlation doesn't imply causation" anyway.

Oh. My. God. Now I'm gonna have to start calling you "OccaomFyre" or "FoxBill."

It is not an assumption. It is a truism. And scientists are the ones that came up with it.

The logical flaw is when you negate it and say "correlation does imply causation." Cum hoc ergo propter hoc. With this, because of this.

0 Replies

@Robert Gentel,

Robert Gentel wrote:

OCCOM BILL wrote:
Then where is yours, Robert?

Here you go again, using it as a rhetorical device.

Quote:
You know perfectly well the discussion you've entered featured DrewDad denying that correlation can imply causation; and that the only reason the inconsequential error you pointed out was brought up was in an example to attempt to disprove the misguided notion that the phrase is some kind of absolute law of science. Or do you?

Bill, I came in and read from the end backwards. I came across you being a blow hard about a simple fact that you were wrong about and I corrected it. What was intellectually dishonest about that? It seems to mean whoever disagrees with you when you casually toss it around like this.

Quote:
Please tell me: Do you think the correlation between smoking and lung cancer had long implied causation since well before the discovery or don't you?

I don't know. The causation has been fairly well established ever since I've researched it so I never had to think about whether or not the correlation implied causation.

And I think this whole "implied" argument is a bit of a silly one too with a lot hinging on the weight one puts on the implication so I haven't been following it.

Quote:
Btw: Accusing me of intellectual dishonesty for an accidental, inconsequential error, is uncalled for.

I didn't call you intellectually dishonest for your error, I called you intellectually dishonest for trying to rope me into other arguments instead of acknowledging that you'd been wrong when you called DrewDad's claim a "total sham".

Quote:
Your bias is showing.

Perhaps. I'm not sure what "bias" that would be but that's certainly a possibility somehow. But an alternate theory I'd like to propose is that you use terms like "intellectual dishonesty" and "bias" as rhetorical devices and are a bit quick on the trigger with them when disagreed with.

I see. I didn't realize I failed to say thank you for providing the gene link, or admit that I was unaware of it. How terribly dishonest of me.

You apparently picked up the conversation long after I became frustrated by DrewDad's incessant bobbing and weaving about essentially meaningless bullshit. Your bias is what's preventing you from stating the obvious that yes:
1. The correlation is obvious (even if meaningless).
2. Correlation can imply causation (even if it doesn't here).

This bias is annoying as hell, because you are considered an authority around here, so when you stop in to opine "DrewDad is right" about one detail, without commenting on the argument around it; it can be reasonably interpreted as an endorsement of the argument instead of the detail. I know you agree that causation isn't proven by those numbers; but no one said it was. I'd appreciate an honest answer to 1. and 2. above.

2 Replies

@OCCOM BILL,

OCCOM BILL wrote:

This bias is annoying as hell, because you are considered an authority around here, so when you stop in to opine "DrewDad is right" about one detail, without commenting on the argument around it; it can be reasonably interpreted as an endorsement of the argument instead of the detail.

Dear lord you are sensitive, I can't correct your facts without joining your whole long winded argument? I can think of something better to do with my time and if that means I'm "biased" then I guess I'm biased, biased against banging my head against a wall for no reason.

0 Replies

@OCCOM BILL,

OCCOM BILL wrote:

1. The correlation is obvious (even if meaningless).

If it is meaningless, then why are you so insistent about having it acknowleged? Let it go already.

OCCOM BILL wrote:

2. Correlation can imply causation (even if it doesn't here).

Correlation implies a relationship. Correlation implies that there may even be a very interesting relationship.

Absence of correlation implies absence of causation.

The fact remains that you posted a meaningless graph, and then attempted to imbue it with some sort of meaning by stating you didn't think it was coincidence. And then tried to bludgeon down any opposition. I'm beginning to think this "intellectual dishonesty" schtick is nothing more than projection on your part.

1 Reply

Killing people is the best solution.

Related Topics

Quick Links

My Account

able2know