Why Bad Beliefs Don't Die

[Jeff Corey]

I have given examples of flawed Randi tests in other threads. see for example http://boards.straightdope.com/sdmb/showthread.php?s=&threadid=181436 where I list some of my thoughts about him. I give some criticisms that apply to his test methodology in general, some specific criticism of individual tests.

Thanks for the link, which included a point-by-point refutation of your weaK arguments by the very next poster.

 

[Peter Morris]

>Frankly, you haven't demonstrated any difference.

I certainly have. What part do you not understand? I'm asking this in all seriousness. My gut feeling is that you are being purposely obtuse, but I will give you the benefit of the doubt. Can you really not discern the difference between;

A> An agrument against the validity of dowsing
and
B> Using the ideomotor effect as a potential explanation of how some dowsers can delude themselves?

Ranedi is claiming falsely that dowsers delude themselves with the ideomotor effect. The claim they delude themselves is a lie, because they admit it their own hand moving the rod.

And, look at it, the reason he is claiming that they are deluding themselves is because he is using the claim as an argument against dowswing. He is saying : look they are deluding themselves, therefore dowsing cannot work.

So, you are unable to demonstrate any difference.

 

[Peter Morris]

>I have given examples of flawed Randi tests in other threads. see for example http://boards.straightdope.com/sdmb...threadid=181436 where I list some of my thoughts about him. I give some criticisms that apply to his test methodology in general, some specific criticism of individual tests.


Wow, I just read through that thread.
I'm stunned that you would point people to that thread as I think it shows your lack of critical thinking and inability to deal with the subject of Randi and testing. I wish I had checked it out sooner as it would have saved me the time of trying to have a rational discussion with you... rather than repeat the same discussions that you have had with others before.

Effectively you are a troll.

Unless you come up with a rational argument, I am done with you. If anyone needs to know what's on Peter's mind, just check out the threads on Straightdope.com.

Yeah, I notice you don't actually have any answer beyond rude name calling, which is the same answer I got from Randi fanatics on Straight dope. They couldn't find an answer either.

I have listed numerous reasons why in my opinion Randi's tests are not resonable or fair. You are entitled to disagree if you wish. By all means supply answers to my points, if you can.

"shows my lack of critical thinking" How so? I think critically about everything, I question Randi just as much as dowsers. Most people on these forums just accept blindly whatever Randi tells them, without questioning.

 

[Peter Morris]

Thanks for the link, which included a point-by-point refutation of your weaK arguments by the very next poster.

My arguments were strong, the next poster had no refutation beyond spewing forth a stream of vitriol. I have still not seen a sensible answer. All I have received in response is hatred.

 

[Jeff Corey]

My arguments were strong, the next poster had no refutation beyond spewing forth a stream of vitriol. I have still not seen a sensible answer. All I have received in response is hatred.

Did you ever consider that you might have a vastly overinflated opinion as to the strength of your arguments?










No, I didn't think so.

 

[Darat]

...snip...

I have listed numerous reasons why in my opinion Randi's tests are not resonable or fair. You are entitled to disagree if you wish.

...snip...

I should point out Peter that you still have not supplied evidence for the claim you keep repeatedly making.

Perhaps if you can't provide the evidence to support your claim you should re-consider stating that claim?

A suggestion I have is that you re-state it as "I have listed numerous reasons why in my opinion some of Randi's test are not reasonable or fair.

To summarise:

I must assume that your claim that "... I consider Randi's tests to be deeply flawed too...." cannot not be substantiated and was an incorrect and misleading statement.

And you cannot support your claim that under the conditions I stated i.e. "truly blinded" dowsing has been proven to work.

 

[Peter Morris]

Thank you for your response and link. This is a review piece - do you have any links to the successful “truly blinded” tests?

I'll leave you to seek them out for yourself.

My point in this thread has been to show that the 'ideomotor' argument is flawed, I have repeatedly stated that I'm not advocating dowsing.

I try to read as little "between" the lines as possible in other people posts so please excuse the fact that because you didn't limit your claim I assumed you meant "all".

and do you know what happens when you assume?

Just to be certain I'm not misunderstanding you are now retracting your original claim of

"Actually, I consider Randi's tests to be deeply flawed too, but that's a seperate discussion."

And substituting a claim of:

"Some of Randi's tests are flawed. "

I'm not "retracting" anything, I stand by what I said, but to make it clearer:

There are flaws that Randi repeats time after time in many of his tests. These include the following, among others:

1) The test offered by Randi is frequently different from the claim that was made.

2) Most paranormal claims say that someone can score a little better than chance, but do so every time. For example, given a randon 1 in 10 chance, someone may claim to score 20% consistently over a long series of tests. Randi, however, requires a near-perfect score in a very short series, usually 8 hits out of 10 tries.

3) There is an adage that "nobody can be a judge in their own cause" What that means is that nobody can be trusted to conduct a test fairly when they have a personal stake in the results. Randi has an enormous stake in the tests,the JREF would be dead if he ever validated a paranormal effect. In order to be fair, the test would have to be designed and carried out by a third party without a personal interest.

I agree that this would be ridiculous however I have no idea why you are making this point, since it is has nothing to do with what I asked you, which was to explain the flaw in a test of a dowser that Randi carried out in which the dowser failed.

You agree its ridiculous, then do it anyway?

Look, I have cited specific examples of flawed Randi tests and stated why I think they were wrong.

But the example I select aren't good enough, you feel entitled to select examples for me. You chose one specific test, you challenge me to show flaw with that specific test, and if I'm unable to show error in that one test, well then you are going to ignore all the example that I chose.

I have seen flaws in a number of Randi tests, and I have given examples of such. That doesn't mean I can show a flaw in any test you choose.

That is the same type of flawed logic that Randi displays.

The claim : I can show faults in a number of Randi's tests, here's a list of specific examples.

The challenge: You have to show flaw in this particular test that wasn't on your list.

Your challenge isn't a fair test of my claim, and that's the same trick that Randi pulls frequently.

Do you see?

Again I don't know what point you are trying to make as this does not address the question I asked.

So in summary, you have agreed that your claim that "... I consider Randi's tests to be deeply flawed too...." cannot not be substantiated

No, I agree no such thing

and you still have not provided links to tests that support your claim that under the conditions I stated i.e. "truly blinded" dowsing has been proven to work.

I never made that claim.You have misremembered what I said. My exact words were : "Actually, there are plenty of tests that seem to verify dowsing, but Randi dismisses all of them bar his own."

I have shown you the existence of some tests that seem to verify dowsing. I never made the claim that it had been "proved" and the "truely blinded" part is your own addition.

I chose my words carefully, and I have shown what I said to be correct.

 

[Jeff Corey]

2) Most paranormal claims say that someone can score a little better than chance, but do so every time. For example, given a randon 1 in 10 chance, someone may claim to score 20% consistently over a long series of tests. Randi, however, requires a near-perfect score in a very short series, usually 8 hits out of 10 tries.

If the probability of getting it right was .01 and someone got 8 out of 10 correct, that would be quite impressive. The probability(p value) of that occurring by chance would be about .0000004.
So I think that would be a fair test.

 

[Peter Morris]

Did you ever consider that you might have a vastly overinflated opinion as to the strength of your arguments?

I judge it by the lack of coherrent counter-argument.
If you have an answer to my comments, please state it.

 

[Peter Morris]

If the probability of getting it right was .01 and someone got 8 out of 10 correct, that would be quite impressive. The probability(p value) of that occurring by chance would be about .0000004.
So I think that would be a fair test.

But suppose someone claimed that they could get 2 out of 10, and maintain that score over a long series.

Asking them to get 8 out of 10 on a single test would be an unfair test.

That is the problem.

 

[Mr Manifesto]

Say, speaking of bad beliefs which don't die...

 

[gnome]

But suppose someone claimed that they could get 2 out of 10, and maintain that score over a long series.

Asking them to get 8 out of 10 on a single test would be an unfair test.

That is the problem.

It is quite appropriate if the dowser claimed an 80% success rate. It's not fair to go back afterwards and say, "Well I didn't make 80% but I did better than chance..."

If your claim is to do better than chance consistently, then make that your claim in the first place. There are plenty of ways to test that.

 

[Peter Morris]

It is quite appropriate if the dowser claimed an 80% success rate. It's not fair to go back afterwards and say, "Well I didn't make 80% but I did better than chance..."

If your claim is to do better than chance consistently, then make that your claim in the first place. There are plenty of ways to test that.

True, there are many ways to test that. I think that the test should be " better than chance, and do it every time"

Unfortunately, Randi doesn't accept that. You have to agree a specific score in advance, and if you miss that score, you fail. (I've never seen Randi agree a target score of less than 80%)

Getting better than chance isn't an issue in Randi's tests. There was a dowsing test run by Randi with expected chance result 10%, actual result 22%, target score 80%. Randi's comment was to draw attention to it being well below the target, and ignore it being well above chance.

That is, in my mind a bad test. When somethin like that happens, Randi should extend the test to see if they can maintain the 22%, or whether it was a fluke.

The odds against it happening by chance are about 107-1, so a fluke is quite likely. I'm not advancing this as evidence of dowsing, just as evidence of Randi's flawed testing methods.

I think the tests ought to be "can the claimant do consistantly better than chance" rather than "can he reach a specific target"

 

[Jeff Corey]

Getting better than chance isn't an issue in Randi's tests. There was a dowsing test run by Randi with expected chance result 10%, actual result 22%, target score 80%. Randi's comment was to draw attention to it being well below the target, and ignore it being well above chance.

Cite the source for that or I'll be forced to conclude you are just making it all up.

 

[Peter Morris]

Cite the source for that or I'll be forced to conclude you are just making it all up.

http://www.skeptics.com.au/journal/divining.htm

Three seperate tests for three different claims:

Water - expected chance result 10%, actual result 22%
Brass -- expected chance result 10%, actual result 0%
Gold ---- expected chance result 10%, actual result 11%

The 22% result would occur by chance once every 107 times he runs similar tests.

In a well run test, given a similar result, the tester should make further tests to see if the subjects can maintain the score.

Finding a result far above chance expectation, Randi does two things. First, he crows about it being much less than the agreed target. Secondly, he massages the data, by adding the three results together, thus making it sound less impressive. You have to read the report very carefully to see the actual result was 22%.

This was three seperate tests for three seperate claims. Adding the results together is meaningless. The three results stand on their own.

Rather shabby of Randi, IMHO.

 

[Darat]

I'll leave you to seek them out for yourself.

My point in this thread has been to show that the 'ideomotor' argument is flawed, I have repeatedly stated that I'm not advocating dowsing.

That may have been your point but you also made the claim:

...snip...

Actually, there are plenty of tests that seem to verify dowsing, but Randi dismisses all of them bar his own.

...snip...

All I am doing asking you to support your claims, and since once again you are unable to then I have to conclude there is still no substance to the claim above.

...snip...

and do you know what happens when you assume?

....snip....

If you read what I wrote you will see I made the minimum of assumptions in my interpretation of your words, you had said:

...snip...

Actually, I consider Randi's tests to be deeply flawed too, but that's a seperate discussion.

...snip...

What you are now saying is that I should have assumed that you only meant to say "many of" or "some of"! And as I said I try not to make these types of assumptions.

...snip...

I'm not "retracting" anything, I stand by what I said, but to make it clearer:

There are flaws that Randi repeats time after time in many of his tests. These include the following, among others:

...snip...

I am glad you have understood my points and decided to change your claim.

You are no longer claiming "I consider Randi's tests to be deeply flawed too" you are now limiting your claim to not include all of Randi's test (which of course could all be flawed but you haven’t supplied evidence to support your claim).

...snip...
I never made that claim.You have misremembered what I said. My exact words were : "Actually, there are plenty of tests that seem to verify dowsing, but Randi dismisses all of them bar his own."

I have shown you the existence of some tests that seem to verify dowsing. I never made the claim that it had been "proved" and the "truely blinded" part is your own addition.

...snip...

I did not "misremember" what you said. I have several times quoted your exact words back to you.

The sequence of claims and questions was as follows:

...snip...

No, not really. I find many skeptics to be totally closed-minded. They just know that dowsing doesn't work, and accept any old argument that reinforces their belief, they often get very angry when I question the weak arguments they rely on.

...snip...

To which I replied:

...snip...

In the case of dowsing I don't rely on any weak "arguments" I rely on evidence, or rather lack of evidence.

Every report I've ever read in which a "dowser" is faced with a truely blinded situation their ability fades away.

...snip...

To which you replied:

...snip...

Actually, there are plenty of tests that seem to verify dowsing, but Randi dismisses all of them bar his own.

...snip...

How else could I interpret your response but that the tests you were referring to “truly blinded”? Otherwise your response would be sheer nonsense as it didn’t refer to my claim that you quoted when you stated the above.

Again in summary you still have not provided any evidence for your original claims therefore I still have to discount them. I do note however that you have retracted one claim and substituted another.

 

[Peter Morris]

All I am doing asking you to support your claims, and since once again you are unable to then I have to conclude there is still no substance to the claim above.

support given over and over. You have been unable to answer it.

If you read what I wrote you will see I made the minimum of assumptions in my interpretation of your words

No. You made the maximum.

I am glad you have understood my points and decided to change your claim.

No.

You are no longer claiming "I consider Randi's tests to be deeply flawed too"

Yes I am.

you are now limiting your claim to not include all of Randi's test

If you didn't make assumptions I wouldn't have needed to clarify this point. It's the same its always been.

which of course could all be flawed but you haven’t supplied evidence to support your claim).

I've statefd several times a lot of flaws in Randi's tests. You can disagree if you want, but to say that I haven't supplied evidence is a sheer lie. You have given no answer to my list, other than to deny it exists.
Just to tremind you, I have referred you to a long essay here http://boards.straightdope.com/sdmb...threadid=181436
I have stated three flaws in Randi's tests as follows:

1) The test offered by Randi is frequently different from the claim that was made.

2) Most paranormal claims say that someone can score a little better than chance, but do so every time. For example, given a randon 1 in 10 chance, someone may claim to score 20% consistently over a long series of tests. Randi, however, requires a near-perfect score in a very short series, usually 8 hits out of 10 tries.

3) There is an adage that "nobody can be a judge in their own cause" What that means is that nobody can be trusted to conduct a test fairly when they have a personal stake in the results. Randi has an enormous stake in the tests,the JREF would be dead if he ever validated a paranormal effect. In order to be fair, the test would have to be designed and carried out by a third party without a personal interest.

These just scratch at the surface.

If you claim I have no evidence, then you must be able to refute these points, but you have no answer, other than to deny they exist.

I do note however that you have retracted one claim and substituted another.

No.

 

[gnome]

1) The test offered by Randi is frequently different from the claim that was made.

This is necessary, as the original claim made is usually not testable as is. The claim is refined to fit the constraints of a test where success by cheating or chance (or, indeed, anything but the use of the claimed power itself) is minimized or eliminated. This is a joint process with the final result agreed to by the applicant.

2) Most paranormal claims say that someone can score a little better than chance, but do so every time. For example, given a randon 1 in 10 chance, someone may claim to score 20% consistently over a long series of tests. Randi, however, requires a near-perfect score in a very short series, usually 8 hits out of 10 tries.

From JREF's library entry on dowsing:

Most dowsers claim 100% accuracy. Very few claim anything less than 90%.

To continue to test when a person does not meet the conditions of the test, just because they performed better than chance on that occasion, amounts to allowing the applicant to move the goalposts after the fact. This seems foolish with a million dollars on the line. If their claim is to perform marginally better than chance, but consistently, that is testable and there is no reason to believe Randi would not devise a test designed to demonstrate that to satisfaction. If you can find me an applicant who claimed this, and Randi insisted on superb accuracy, please cite it.

3) There is an adage that "nobody can be a judge in their own cause" What that means is that nobody can be trusted to conduct a test fairly when they have a personal stake in the results. Randi has an enormous stake in the tests,the JREF would be dead if he ever validated a paranormal effect. In order to be fair, the test would have to be designed and carried out by a third party without a personal interest.

The test is designed by both parties--and is designed so that no judging is involved. Why this is possible is the subject of many other threads, but if you really want to press that we can get into it.

 

[crocodile deathroll]

Simple, people fight harder and are less afraid of there mortality if they believe they will be rewarded for their military action in the after life. In Norse mythology if they do no die in battle and die of disease they believe they are doomed to live an eternity in a horrible cold world of Niflheim
All the couragous warriers believe they go to this place even if it means acting upon on what on today's terms we call "ethnic cleansing", by slaughtering all those who do not share their bellicose culture by doubting the existence of Valhalla. Hence they also have the survival advantage for their bad beliefs as those that are not willing to fight are all going to be killed anyway.

CDR

 

[Vitnir]

Why don't you take this discussion to the forum "The Million Dollar Challenge"?

So far I have seen one reply on the article on www.csicop.org

Noone else have an opinion on what the article was about?