JREF Science

One aspect of scientific endeavors is that the specifc data, methods, and analyses are made available to all who are interested.

The results of the Challenge clearly don't apply.

I can understand the legal matters, which is why we only get tidbits of vague happenings, but to be taken seriously by the entire scientific community (as opposed to a small subset), wouldn't JREF's case be better served if they had some type of peer reviewed journal their findings were published in??

I believe that is why a lot of people have a hard time taken the JREF proceedings as anything other than mildly interesting and entertaining.

[Unas]

Originally posted by Whodini
One aspect of scientific endeavors is that the specifc data, methods, and analyses are made available to all who are interested.

The results of the Challenge clearly don't apply.What specific information regarding the Challenge isn't available?

[Soapy Sam]

Whodini, the Forum is not JREF. So far as I understand, JREF is , basically , James Randi and a few helpers.
Your idea may or may not have merit, but I suggest this is not the place to propose it. You should contact JREF directly.

[LucyR]

How on earth did this thread receive five stars?

I'm almost certain those five stars appeared before a single reply had been posted!

Presumably the originator of the thread is the guilty party.

Surely there should be a law against this sort of hubris.

[rwald]

I agree that publishing scientific data is a worthy goal (and voted yes in the poll), but would suggest that based on the frequency of the "first step" challenges which Randi fields, it may not be reasonable to request that all of them are published (besides, they're not even statistically significant, because Randi doesn't demand that high a level of significance for the first level of challenge). However, if anyone were to ever pass the first level, and proceed to the final challenge, I would hope those results would be published.

LucyR,


----
How on earth did this thread receive five stars?
----


I gave it a vote of 5 stars. If you disagree, feel free to vote as you wish.


----
Presumably the originator of the thread is the guilty party.
----


Yes. Guilty implies too much though, like I did something wrong or something. The voting mechanism is there to be used, and I use it to monitor the evolution of peoples' thoughts on the thread in question.

Most people do not vote, so I start off the voting and watch it from there.

Originally posted by Unas
What specific information regarding the Challenge isn't available?


Full disclosure on type of claim, type of procedure used to test it, the statistics involved (how many "hits" out of n the claimant got, how many were expected by chance, etc.), etc.

For example, I'd like to see even a simple pie chart of the type of claim tested by the JREF.

Then I'd like to see a histogram of the statistics involved (say the claimant scores) broken down by each type of claim, so one can see how, on average, claimants score.

Seeing the types of claims done per year could give insight on paranormal trends.

I'd also like to see theoretical discussions of why a test is done
the way it is.

I'd like to see the data-sets available for download so independent researchers can analyze them.

The main thing:

I'd like to see ALL of the research peer reviewed. The JREF, like any organization, has biases, and is subject to favoring the hits and not the misses, etc.

To be taken more seriously in a scientific manner, the JREF could stand having its results published in an independent peer-reviewed journal.

[rwald]

Oh, I get it now. You want to look at all the studies they've ever done, so you can perform a meta-analysis on them and be able to claim that paranormal activity exists...

Just clairifying things.

LucyR,


----
How on earth did this thread receive five stars?
----


I also have started many of my threads with a 1-star rating to test various theories.

My first theory is that people who disagree with my internet persona with vote a highly rated thread down. Those who agree with my internet persona will vote a low rated thread up.

What do you think?

[jj]

While I can see that it would be good, if possible, to submit the design, results, and evaluation of an JREF test to a scientific journal, I fear it would have difficulty being accepted.

Most scientific journals do not regard "dowsing doesn't work" as news.

On the other hand, such information ought to be published somewhere, I'm just not sure where.

Originally posted by rwald
Oh, I get it now. You want to look at all the studies they've ever done, so you can perform a meta-analysis on them and be able to claim that paranormal activity exists...
Just clairifying things.


Rwald.

I mentioned nothing about meta-analysis or any specific type of analysis that I'd perform. But a meta-analysis isn't inherently a bad thing. One could also perform a meta-analysis that shows that there is NO overall effect.

You must be using your mind reading or future prediction powers. Maybe you should take the test? :)

Having results peer-reviewed and published seems sensible for any scientific enterprise, don't you think? Doing this would only help JREF's mission, IMO.

[Tom Head]

Which journal do you have in mind?


Cheers,

[Interesting Ian]

Originally posted by Whodini



Full disclosure on type of claim, type of procedure used to test it, the statistics involved (how many "hits" out of n the claimant got, how many were expected by chance, etc.), etc.

For example, I'd like to see even a simple pie chart of the type of claim tested by the JREF.

Then I'd like to see a histogram of the statistics involved (say the claimant scores) broken down by each type of claim, so one can see how, on average, claimants score.

Seeing the types of claims done per year could give insight on paranormal trends.

I'd also like to see theoretical discussions of why a test is done
the way it is.

I'd like to see the data-sets available for download so independent researchers can analyze them.

The main thing:

I'd like to see ALL of the research peer reviewed. The JREF, like any organization, has biases, and is subject to favoring the hits and not the misses, etc.

To be taken more seriously in a scientific manner, the JREF could stand having its results published in an independent peer-reviewed journal.

Indeed, and unless such demands are acceded to, the failure of anyone to pass the challenge can scarcely be said to constitute any evidence against the appropriate "paranormal" phenomena which properly conducted parapsychological research tends to suggest exists.

PS I gave this thread 5 stars.

Edited to add: Just read the rest of the thread. Just wanted to say that I gave the thread 5 stars because you raise an interesting issue.

[Unas]

Originally posted by Whodini
Originally posted by Unas
What specific information regarding the Challenge isn't available?
Full disclosure on type of claim, type of procedure used to test it, the statistics involved (how many "hits" out of n the claimant got, how many were expected by chance, etc.), etc.For which specific challenge do you claim that this information is unavailable? How did you determine that the information is unavailable?

[Unas]

Originally posted by Interesting Ian
Indeed, and unless such demands are acceded to, the failure of anyone to pass the challenge can scarcely be said to constitute any evidence against the appropriate "paranormal" phenomena which properly conducted parapsychological research tends to suggest exists.The challenge cannot provide any evidence against any paranormal phenomenon, and that is not its purpose. The challenge exists to encourage those who claim that paranormal phenomena exist to demonstrate said phenomena under controlled conditions. If a claimant fails to demonstrate a specific phenomenon, then that means that the phenomenon could not be demonstrated under controlled conditions by that claimant.

For what specific challenge is the data that Whodini refers to unavailable? How have you determined that the data is unavailable?

[Pyrrho]

James Randi has said in the past that people can view information on past tests at the JREF offices.

The JREF isn't doing science. People make claims; the JREF tests their claims; people fail the tests at the preliminary, informal level; formal tests are not done, with few exceptions: the most recent and memorable being the girl who pretended she could see through opaque goggles. Most claims do not involve statistics -- dowsing, for example, is pretty much a binary test. A spectacular example of a dowsing failure exists on this board. The example of the man in Asia who claimed he could diagnose illnesses by looking at a person photograph required no statistical study, and was dealt with quite simply. The recent test of homeopathy was broadcast on television in the UK, and all the information is accessible. Many of the claims are just plain non-testable.

It's a thought, though. Randi might make some money if he wrote a book compiling all the wacky claims he's received over the years. I'm sure a lot of anti-Randi folks would snap it up and mine it for smoking guns.

[Pyrrho]

Originally posted by Interesting Ian
Indeed, and unless such demands are acceded to, the failure of anyone to pass the challenge can scarcely be said to constitute any evidence against the appropriate "paranormal" phenomena which properly conducted parapsychological research tends to suggest exists.


AFAIK, the JREF doesn't claim that paranormal phenomena do not exist, and it doesn't use test failures as evidence against paranormal phenomena.

Granted, the failure of all the claimants to prove their claims does cast doubt on the existence of the paranormal abilities they claimed to have.

If one cannot prove their claim to paranormal abilities, they have no business claiming to have them.

Unas,


----
For which specific challenge do you claim that this information is unavailable? How did you determine that the information is unavailable?
----


I mentioned full disclosure, so for all claims.

And the info. may be available if you spend the whole day at JREF asking politely. It is just that none of it is published in peer reviewed scientific journals.

----
The JREF isn't doing science. People make claims; the JREF tests their claims; people fail the tests at the preliminary, informal level; formal tests are not done, with few exceptions:
----


Pyrrho,

If science isn't done, and most of the tests are informal, why should anyone take the JREF challenge as anything other than interesting entertainment?

I agree that the challenge does something, but JREF's case would be magnified by 1000 times if they got their results published in a peer reviewed scientific journal.

Then we could say things like : 'Go look at The Journal of Science, Vol 3, p. 87-94, 2003', instead of 'Go look at the JREF website'.

LOL


----
Most claims do not involve statistics -- dowsing, for example, is pretty much a binary test.
----


That seems to involve statistics to me. If you present a dowser with n pairs of containers, one of each pair contains a gold nugget, then a priori you'd expect the dowser to get n/2 correct by chance alone. If the dowser gets more (how much more is determined by examining the probabilities from a Binomial distribution), then something could be going on. Then one would have to replicate the experiment again.
There are definitely statistical issues here: experimental design, methodology, analysis, p-values, results of dowsing trials from the past, etc.


----
It's a thought, though. Randi might make some money if he wrote a book compiling all the wacky claims he's received over the years.
----


That would be good, but again, it isn't a peer reviewed scientific journal publication, so people probably wouldn't view it as more than interesting entertainment.

[FutileJester]

Originally posted by Whodini
If science isn't done, and most of the tests are informal, why should anyone take the JREF challenge as anything other than interesting entertainment?

I believe that is all they are intended to be, with maybe the work 'educational' thrown in. I see the challenge as a public way to educate people about critical examination of claims. The challenge is informed and inspired by science, but it is not itself science. Nor does it intend to be.

I agree that the challenge does something, but JREF's case would be magnified by 1000 times if they got their results published in a peer reviewed scientific journal.

I agree with the sentiment, but there are big roadblocks. First is that the JREF mission is education, not science, and their meager resources are stretched doing just that. Peer-reviewed research is a lot of work; who's going to do it? And, as others have pointed out, what journal will want to publish thirty articles a year that amount to 'yep, dowsing still doesn't work'?

I would love to see details about as much of the testing as possible. I believe an online archive has been discussed before and may even be in progress. But even then, it won't be science.

Naturally, anything suggestive that comes up in the testing should inspire actual scientific studies. But the JREF is just not in a position to do these itself.

[Unas]

Originally posted by Whodini
I mentioned full disclosure, so for all claims.Evasion noted. Are there any specific JREF challenges for which the specific data, methods, and analyses are unavailable?
Originally posted by Whodini
And the info. may be available if you spend the whole day at JREF asking politely. I fail to see how that makes the information unavailable. Please explain.
Originally posted by Whodini
It is just that none of it is published in peer reviewed scientific journals. I fail to see how that makes the information unavailable. Please explain.

----
Naturally, anything suggestive that comes up in the testing should inspire actual scientific studies. But the JREF is just not in a position to do these itself.
----


So can people understand that when a believer says something like:

'JREF?! hehe, what a joke, it isn't even scientific, no peer review, no publishing in journals, why should I even consider it more than entertainment?'

, etc., that the believers have a pretty good point?

[Oso]

Originally posted by FutileJester
I believe that is all they are intended to be, with maybe the work 'educational' thrown in. I see the challenge as a public way to educate people about critical examination of claims. The challenge is informed and inspired by science, but it is not itself science. Nor does it intend to be. Absolutely correct. but I'm a supporter of the JREF Challange not as an attempt to build evidence against the paranormal or even to educate, I support the challenge because if there's any magic out there I want to see it! It'd be worth a million dollars, easy.

But so far it aint lookin' good.

[Skeptoid]

Fact is, scientific journals aren't in the business of peer reviewing and publishing negative results.

[jj]

Originally posted by Skeptoid
Fact is, scientific journals aren't in the business of peer reviewing and publishing negative results.

It does occasionally happen where there is a dispute.

However, most scientific journals do not dispute the obvious existance of evolution, or the obvious nonexistance of dowsing, PSI, etc.

[Skeptoid]

Who would even want to referee such a journal? Page after page of "Claimant could dowse at 100% level when he knew where the samples were placed ... dowsed at chance level when samples were placed with his back turned." ... "Claimant could read War and Peace upside-down when her father placed the blindfold ... couldn't name the title when Randi placed the blindfold." ....

One word -- Boooooring.

[FutileJester]

Originally posted by Whodini
So can people understand that when a believer says something like:

'JREF?! hehe, what a joke, it isn't even scientific, no peer review, no publishing in journals, why should I even consider it more than entertainment?'

, etc., that the believers have a pretty good point?

I would think that you would be among the first to say that science is not the beginning and end of every human endeavor. Do you consider everything that is not science to be a 'joke' and 'just entertainment'?

Bill Nye the Science Guy is not doing science. He is educating and entertaining. Is he therefore a 'joke'?

This is not the James Randi Science Foundation. It is the James Randi Educational Foundation. The purpose of the challenge is to educate people about how to think critically about extraordinary claims, and the challenge is a useful element of that education.

Saying 'it isn't good evidence' is a silly complaint when it is not being presented as evidence. There are, without a doubt, delusional and fraudulent people out there. The challenge has also, without a doubt, sometimes demonstrated the ways in which some people are deluded or fraudulent. Whether or not there are actual phenomena that will someday pass the challenge, it has served it's purpose in showing ways in which people have been fooled and ways to think about things to keep from getting fooled ourselves.

[Fidelio]

Originally posted by Whodini
...So can people understand that when a believer says something like:

'JREF?! hehe, what a joke, it isn't even scientific, no peer review, no publishing in journals, why should I even consider it more than entertainment?'

, etc., that the believers have a pretty good point?

I could be wrong about this but I think that's where the million $ comes in, to assuage the trauma and indignity of not being published in a scientific journal.

Originally posted by Skeptoid
Who would even want to referee such a journal? Page after page of "Claimant could dowse at 100% level when he knew where the samples were placed ... dowsed at chance level when samples were placed with his back turned." ... "Claimant could read War and Peace upside-down when her father placed the blindfold ... couldn't name the title when Randi placed the blindfold." ....
One word -- Boooooring.


No one said science is all exciting.

And frankly the results are exicting, especially if they are given more academic authority by being published in a peer reviewed scientific journal.

FutileJester,


----
Do you consider everything that is not science to be a 'joke' and 'just entertainment'?
----


Uh, no I don't.

My point was that if people are claiming the JREF challenge to be some kind of gold standard, then they are deluding themselves (like the dowsers, et al) because there is no free exchange of information and data to make it more 'sciencey' (tm)

;)


----
This is not the James Randi Science Foundation. It is the James Randi Educational Foundation. The purpose of the challenge is to educate people about how to think critically about extraordinary claims, and the challenge is a useful element of that education.
----


I agree with that.

[Interesting Ian]

Originally posted by jj


It does occasionally happen where there is a dispute.

However, most scientific journals do not dispute the obvious existance of evolution, or the obvious nonexistance of dowsing, PSI, etc.

Well that's interesting. So despite the collosal amount of evidence for anomolous cognitive abilities it is obvious they don't exist :rolleyes:

Might I suggest the JREF is lamentably failing in its mission to educate people if the attitude that you express is typical.

Edited to add:

BTW I'm not suggesting such phenomena definitely does exist, but to to claim that it is obvious that no "psi" exists is clearly preposterous. Go back to arguing with Win.

[Interesting Ian]

Originally posted by FutileJester
This is not the James Randi Science Foundation. It is the James Randi Educational Foundation. The purpose of the challenge is to educate people about how to think critically about extraordinary claims, and the challenge is a useful element of that education.



Educate people? Don't make me laugh! You mean of course "brainwash" people into an uncritical acceptance of the common modern western metaphysic.

[Unas]

Originally posted by Interesting Ian
So despite the collosal amount of evidence for anomolous cognitive abilities it is obvious they don't exist.What "collosal amount of evidence" are you referring to? Note that true evidence doesn't suddenly evaporate under controlled conditions.

[Interesting Ian]

Originally posted by Unas
What "collosal amount of evidence" are you referring to? Note that true evidence doesn't suddenly evaporate under controlled conditions.

Anecdotal and parapsychological research. Incidentally this "true evidence" gambit sounds suspiciously like the no true scotman fallacy (or whatever it's called). Having said that I know of no particular research suggesting it evaporates under controlled conditions. I suspect this is a lie put out by the so called "skeptics". The problem with anomalous cognitive abilities is that they cannot be consistantly replicated on each and every occasion, and in addition there is no readily acceptable overarching theory accounting for the phenomena.


BTW I've just been talking about the million dollar challenge in the paranormal forum today so go there if you or anyone else want to criticize my ideas on that subject.

[Unas]

Originally posted by Interesting Ian
Anecdotal and parapsychological research.Anecdotal evidence, of course, is by definition not obtained under controlled conditions, and cannot be considered conclusive evidence of anything.

What specific parapsychological research, performed under controlled conditions, provides conclusive evidence for the existence of "anomalous cognitive abilities"?
Originally posted by Interesting Ian
Incidentally this "true evidence" gambit sounds suspiciously like the no true scotman fallacy (or whatever it's called). Having said that I know of no particular research suggesting it evaporates under controlled conditions. I suspect this is a lie put out by the so called "skeptics".Your suspicions aren't relevant. Do you have some facts to back up your views?

[rwald]

The biggest continuing problem with Whodini's suggestion remains the question of which scientific journal would publish it. No matter how good this idea seems to be on the surface, if you can't find a journal which would publish 30 articles a year about how dousing doesn't work, then there's no reason to blame Randi. There's nothing he can do.

Originally posted by rwald
The biggest continuing problem with Whodini's suggestion remains the question of which scientific journal would publish it. No matter how good this idea seems to be on the surface, if you can't find a journal which would publish 30 articles a year about how dousing doesn't work, then there's no reason to blame Randi. There's nothing he can do.


Science?
Nature?
Statistical Science?
Journal of Parapsychology

Hell, even Skeptic or the Skeptical Enquirer

Or start up your own journal, policed by a panel of independent reviewers.

[rwald]

Science, Nature, et all, have better things to do than publish more examples of dousing not working. This isn't a viable option.

I'm not sure how many people here would consider the Journal of Parapsychology to be a trustworthy peer-reviewed journal...besides, do they want to publish evidence that their pet claim is false?

Skeptical Inquirer...this is actually a possibility. I mean, it may get redundent after a while, but it could be a recurring segment. This is something to think about.

And suggesting that Randi start his own journal...I don't think that's likely. If he publishes it on his own (online, in book form...), that would be reasonable, but requiring that he make a policed peer-revied journal? That's a bit much.

Conclusion: I can see some skeptical magazine publishing Randi's studies, but that's about it.

Of course, you would say that the above mentioned magazines aren't "real science," no better than self-publishing in a book...

----
Science, Nature, et all, have better things to do than publish more examples of dousing not working. This isn't a viable option.
----


How do you know what these journals will or wont do?


----
I'm not sure how many people here would consider the Journal of Parapsychology to be a trustworthy peer-reviewed journal...besides, do they want to publish evidence that their pet claim is false?
----


Who knows.

Skeptics could start up their own scientific peer-reviewed journals. That would be a big plus. Referencing the Skeptical Enquirer or a Martin Gardner book isn't exactly taken seriously by anyone except Skeptics.


----
, but requiring that [Randi] make a policed peer-revied journal? That's a bit much.
----


I'm not requiring anything, nor am I in any position to. They are simply suggestions.

[DrBenway]

I believe the JREF challenge is about science. The notion of establishing conditions which control for self-deception, fraud, or observational bias, when evaluating a demonstration of the paranormal, is a scientific notion.

Whether or not a journal exists to publish the JREF trials is moot. A summary of who was tested and how they were tested could be published in any format, anywhere. I've often seen a summary of specific preliminary trials in Randi's weekly commentaries.

Perhaps Whodini is asking for this information to be orginized in one place, so that he/she could easily review:
1. how many people have submitted applications for the JREF challenge
2. how many applications were rejected before any trials were performed
3. how many preliminary trials were done, who was involved, what was done
4. how many trials beyond the preliminary screens were done.

I think this is a splendid idea.

I also suspect that the above information won't take up much space. :D

[DrBenway]

Originally posted by Skeptoid
Fact is, scientific journals aren't in the business of peer reviewing and publishing negative results.

Journals do publish negative results, so long as the experimental methods in the study are sound and the question being addressed by the study is interesting.

Evaluating whether echinacea helps the common cold is interesting, even if the answer is "no."

Years ago, ESP was interesting. But several decades of negative results have made it less interesting. Today, only some clearly positive results will merit much attention from the scientific community.

[DrBenway]

Originally posted by Fidelio
I could be wrong about this but I think that's where the million $ comes in, to assuage the trauma and indignity of not being published in a scientific journal.
Hehe. A psychic is going to be "traumatized" by not getting published in a scientific journal?

Not to worry. If someone wins the million, you can bet that the scientists involved in the challenge experiment will publish a description of their observations. Those scientists will enjoy a rocket ride to fame and fortune, by virtue of being the first to introduce concrete proof of the paranormal to the scientific community. Their peers will review every tiny detail of the experiment, to rule out the usual suspects of self-deception, fraud, and bad statistics. Others will want to replicate the experiment, to verify that something real is there.

The world will change dramatically.

Until that day, what we're going to get is a trickle of folks who seem to think they've got something real, but who will be found out as self-deluded or dishonest at the front door.

Up to this point, the "published reports" of the JREF trials that I've seen look something like this: "We gave the girl a book to read with the blindfold on. Once we taped the edge of the blindfold to her face, she couldn't read anything."

This event can be padded out with a few more details, but it's not enough of anything to merit submission to "Nature."

However, a listing of the applications to the JREF with some details about what happened with each one (dates of any interviews with prospective challengers, who was present, what was done, etc.) --all in one place-- would be convenient, as Whodini mentioned.

[JAR]

Originally posted by Whodini
One aspect of scientific endeavors is that the specifc data, methods, and analyses are made available to all who are interested.

That's where you're wrong. If you read Randi's commentary, he tells how his tests for paranormal claims are done. In order to say what you've said, you'd have to have not bothered to read Randi's writings or you are just plain lying.

[JAR]

Originally posted by Whodini
wouldn't JREF's case be better served if they had some type of peer reviewed journal their findings were published in??



What peer reviewed journal would you be thinking of Whodini and why does JREF's findings need to be published there in order to be valid?

JAR,


----
That's where you're wrong.
----


I said:


"One aspect of scientific endeavors is that the specifc data, methods, and analyses are made available to all who are interested."


How can that be wrong? That is what scientific endeavors consist of.


----
If you read Randi's commentary, he tells how his tests for paranormal claims are done.
----


I understand that. But not for all of them, and not in detail, and there are no data sets I can download and analyze and make pretty graphs of. There is no journal I can cite. I just have to keep citeing a webpage and fellow academics laugh at it. There are no pvalues but rather 'I put the blindfold on her, and she couldn't do the test', LOL.


----
In order to say what you've said, you'd have to have not bothered to read Randi's writings or you are just plain lying.
----


Oh brother. Don't you too start on the ad-homs and character assasination already. At least wait until your 3rd or 4th post newbie.

Originally posted by JAR

What peer reviewed journal would you be thinking of Whodini and why does JREF's findings need to be published there in order to be valid?


JAR,

I don't know which journals, or perhaps a new one could be started up?

Certainly JREF's findings are valid on their own. But being peer reviewed in a scientific journal would make them much more powerful and they'd be taken more seriously by just about everyone.

[fishbob]

My point was that if people are claiming the JREF challenge to be some kind of gold standard, then they are deluding themselves (like the dowsers, et al) because there is no free exchange of information and data to make it more 'sciencey' (tm) The JREF challenge is exactly the Gold Standard - all a claimant has to do to get the gold is to do what they claim under controlled conditions, conditions they agree to beforehand. There is no data to exchange. The JREF challenge is not research, or science. It is a bet - Put Up or Shut Up. Very entertaining, and very educational.

Scientific research is subject to peer review so that others can repeat the process and see if they get the same results. Peer review is more than having somebody check your notes for typos.

[Interesting Ian]

Originally posted by DrBenway
Up to this point, the "published reports" of the JREF trials that I've seen look something like this: "We gave the girl a book to read with the blindfold on. Once we taped the edge of the blindfold to her face, she couldn't read anything."


How the f*ck can someone read a book with a blindfold on?? Is this the sort of stuff that Randi tests?? This is supposed to prove or give strong evidence that no anomalous phenomena coming under the purview of the "paranormal" exists whatsoever??

What a complete and total ****in* bunch of ****in* ********* you lot are.

Dickheads!

[Stimpson J. Cat]

Ian,

How the f*ck can someone read a book with a blindfold on??

They can't. That would require paranormal powers, which don't exist.

Is this the sort of stuff that Randi tests??

That's the kind of claims people make. He tests any claims that people are actually willing to have tested under controlled conditions.

This is supposed to prove or give strong evidence that no anomalous phenomena coming under the purview of the "paranormal" exists whatsoever??

No. That is not the purpose of the million dollar challenge. The purpose of the million dollar challenge is to reveal the people who claim to have demonstrable paranormal powers to be the frauds that they are. Whether or not paranormal phenomena exist doesn't change the fact that the paranormal claims being made by "psychics", dowsers, remote-viewers, faith-healers, homeopathy proponants, and so on, are pure BS.

What a complete and total ****in* bunch of ****in* ********* you lot are.

Dickheads!

Are you drunk again?

Dr. Stupid

[Unas]

Originally posted by Interesting Ian
How the f*ck can someone read a book with a blindfold on?? Is this the sort of stuff that Randi tests??The JREF tests the claims brought to it. If someone claims that they can read through a blindfold, and submits themselves to testing by JREF, then that is what will be tested.
Originally posted by Interesting Ian
This is supposed to prove or give strong evidence that no anomalous phenomena coming under the purview of the "paranormal" exists whatsoever??Nope. Each test does one thing: establish whether the subject's claimed paranormal abilities can be manifested under controlled conditions.
Originally posted by Interesting Ian
What a complete and total ****in* bunch of ****in* ********* you lot are. Dickheads! We're not the ones making assumptions about what JREF does based solely on our ignorance.

Fishbob,


----
The JREF challenge is exactly the Gold Standard
----


So your gold standard is your gold standard even when it is not peer reviewed and not all the information and data is disclosed to all?

LOL.

Originally posted by Interesting Ian


How the f*ck can someone read a book with a blindfold on?? Is this the sort of stuff that Randi tests?? This is supposed to prove or give strong evidence that no anomalous phenomena coming under the purview of the "paranormal" exists whatsoever??

What a complete and total ****in* bunch of ****in* ********* you lot are.

Dickheads!

What a great singer you are.

[Unas]

Originally posted by Whodini
So your gold standard is your gold standard even when it is not peer reviewed and not all the information and data is disclosed to all?You have yet to specify precisely what "information and data" is not disclosed.

[Interesting Ian]

Originally posted by Stimpson J. Cat


Are you drunk again?

Dr. Stupid [/B]

Nope, I just like to be honest and tell the truth.

[Valmorian]

Originally posted by Interesting Ian


Nope, I just like to be honest and tell the truth.

:rolleyes: That's right, if they test a paranormal claim you consider frivolous, they're <expletive deleted>? Why is that?

Hell, Ian, as far as I'm concerned, you're the biggest form of THAT particular tripe on this board, with the possible exception of that Troll Franko.

Ian... Someone comes to James Randi and claims they can read a book with a blindfold on. What exactly would you suggest they test the person for?

[Interesting Ian]

Despite the continuing insinuations that the failure of anyone to win the Randi challenge somehow renders the existence of certain appropriate "paranormal" phenomena somwhat unlikely, I am perfectly satisfied that it suggests nothing of the kind. Indeed I took the liberty of contacting Mr Randi approximately 2 years ago due to the continually claims by the skeptics that obviously no paranormal phenomena exists otherwise someone would have passed the challenge. It is evident from the following e-mail exchange with Randi how misplaced these sentiments are.

(My contributions are in indigo, Randi's contributions are in blue. The exchange has not been altered in any shape or form. I am "Mr Enigmatic")



Dear Mr Randi,

I would just like to clear up a couple of things about the challenge. In order to determine whether someone has actually passed the challenge, is there an objective, independent judging or adjudication made by people who are competent to judge whether the claim has been met? Further, do these people represent a fair cross section over the whole spectrum of belief,
including believers in certain psi phenomena, "skeptics", and those who have yet to decide, or is there a predominance of skeptics? More importantly, who picks these people and what criteria do they use in so choosing?

Finally, can I be assured that you yourself have absolutely no say or influence on whether the contestant has demonstrated the phenomenon in question?

Ian Wardell




Read the rules on the web page. And PLEASE don't ask me questions that are already answered on the web page. That's why we HAVE a web page. When you see a tag on a sweater in Macy's stating the laundry instructions, the fibre content, and the price, do you ask the clerk, "How much is this sweater, what's it made of, and how do I clean it?"


James Randi Educational Foundation
201 S.E. 12th Street (E. Davie Boulevard)
Fort Lauderdale FL 33316-1815
USA

phone: +1 954 467 1112
fax: +1 954 467 1660
Web Page: www.randi.org
e-mail: randi@randi.org




Dear Mr Randi,

I had already read the conditions/rules on the appropriate web page. The reason why I contacted you was because I was unclear as to their precise meaning. The only part which appears to be relevant to my original questions is the part which states "tests will be designed in such a way that no "judging" (inverted commas as in original) procedure is required. Results will be self-evident to any observer, in accordance with the rules
which will be agreed upon by all parties in advance of any formal testing procedure taking place".

I have my reservations whether results will *inevitably* be self-evident to all observers. Specifically, I suspect there may on occasions be a disagreement between the claimant and yourself. Given that there may be disagreement, *somebody* has to decide whether the contestant has demonstrated the phenomenon in question. Who are these people? Are they
completely impartial? If the claimant vehemently disagrees with their decision, what then? It is *imperative* that an independent body decides whether the claimant has passed/failed the
challenge, otherwise the challenge is simply meaningless.

I would also like to know what sort of challenge will be accepted by you? My concern here is that you will not agree to a comparatively mundane demonstration of a claimants "paranormal" powers. That is to say that I am
concerned that you will draw up the rules and conditions in such a way so as to make the claimants task that much harder, and hence, in effect, make the challenge impossible to win. What have you to say to this?


Ian Wardell



Message text written by "Mr Enigmatic"

>Given that there may be disagreement, *somebody* has to >decide whether the contestant has demonstrated the >phenomenon in question.

Never. If a claimant says he can guess at least 18 out of 20 playing cards correctly, and that has to be HIS decision, for example, the rules are drawn up this way: AND in advance, everyone signing off on them, agreeing.

1 correct = failure

2 correct = failure

3 correct = failure

4 correct = failure

5 correct = failure

6 correct = failure

7 correct = failure

8 correct = failure

9 correct = failure

10 correct = failure

11 correct = failure

12 correct = failure

13 correct = failure

14 correct = failure

15 correct = failure

16 correct = failure

17 correct = failure

18 correct =success

19 correct = success

20 correct = success


Now, suppose the claimant makes the 20 guesses, and gets 14 correct. Look at the table. That means a "failure." You follow?
No argument, no judging, no doubt. Understand? WHAT IS SO DIFFICULT TO UNDERSTAND?

>It is *imperative* that an independent body decides whether >the claimant has passed/failed the challenge, otherwise the >challenge is simply meaningless.

No, IT IS NOT! There IS NO DECISIUON TO BE MADE! The results are self-evident. Wrong is wrong, right is right!

>I would also like to know what sort of challenge will be accepted >by you?

Thousands of kinds. Specify. But first, read the rules. Yes, you said you did. This time, try to understand them.

>My concern here is that you will not agree to a comparatively >mundane demonstration of a claimants "paranormal" powers.


Hey, that's RIGHT! Why would I? If a man says he can play the violin, merely stroking the bow across the strings will not suffice. We would have an agreed-upon set of things he would have to do. CAN'T YOU UNDERSTAND THIS SIMPLE IDEA?

>That is to say that I am concerned that you will draw up the >rules and conditions in such a way so as to make the claimants >task that much harder, and hence, in effect, make the challenge >impossible to win.

READ THE DAMN RULES! I don't "draw up" ANYTHING! The CLAIMANT does! So long as an independant group -- of HIS choosing -- decides that the rules and conditions are adequate to prove the claim, WE DO IT!

>What have you to say to this?<

That's it. You seem incredibly slow to understand all this, "Enigmatic." You should change your name to "Dense."

James Randi

Randi




Dear Mr Randi,

The reason why I first contacted you was because so called "skeptics" continually refer to the failure to win the $1 million challenge to justify their "skepticism". My concern here is that this "skepticism" extends to their rejection, not just of the "wild" claims regarding the paranormal, but also of the existence of any phenomena falling within the purview of that particular subset of paranormal phenomena referred to as parapsychology. However, I cannot but conclude that the
failure of anyone to win the challenge has little if any implications for this phenomena, both because people taking the challenge generally attempt to demonstrate abilities/phenomena not encompassed by this subject matter, and more importantly because, they attempt to demonstrate phenomena/abilities way beyond what parapsychological studies suggest can be easily replicated in any case. I therefore ask you whether you would agree that the failure of anyone to meet the challenge has absolutely zero implications for the existence of that particular type of phenomena which parapsychologists study?

Now to address the contents of your last e-mail. The playing card example you cite conceivably may be deemed to be pretty clear cut on the question of whether a claimant has passed the
challenge, but I did state "I have my reservations whether results
will *inevitably* (emphasis as in original) be self-evident to all
observers". I wasn't suggesting that you would be unable to think of a challenge of the top of his head where the results were unambiguous!

I also find your assertion that it is the *claimant's* (emphasis as
in original) decision that he can guess correctly 18 cards out of 20
somewhat implausible. Leaving aside whether you are talking about "playing cards", and whether the claimant knows which 20 cards these are, why should he decide that he can guess 18? I mean think of the astronomical odds! Why should he not stipulate that he can guess just say 1 out of the 20 cards if it is purely *his* (the testees) decision? Besides which, you directly contradict yourself here when you affirm that you will not agree to a comparatively mundane demonstration of a claimants abilities. Mind you, you then contradict yourself *yet again* when you assert "READ THE DAMN RULES! I don't "draw up" ANYTHING! The CLAIMANT does"! If you do not draw up *ANYTHING*, and in fact the claimant does, then, in the card
example, there would be nothing to prevent the claimant from drawing up the rules in the following way:

0 = fail (correct/incorrect inapplicable).

1 correct = success

2 correct = success

3 correct = success

4 correct = success

5 correct = success

6 correct = success

7 correct = success

8 correct = success

9 correct = success

10 correct = success

11 correct = success

12 correct = success

13 correct = success

14 correct = success

15 correct = success

16 correct = success

17 correct = success

18 correct =success

19 correct = success

20 correct = success

I await your reply with interest.

Ian Wardell


END OF RANDI EXCHANGE

Originally posted by Unas
You have yet to specify precisely what "information and data" is not disclosed.


Thaiboxer,

Did I ever say that I would?

You might have to work sometime.

[DrBenway]

Originally posted by Interesting Ian
Now to address the contents of your last e-mail. The playing card example you cite conceivably may be deemed to be pretty clear cut on the question of whether a claimant has passed the challenge, but I did state "I have my reservations whether results will *inevitably* (emphasis as in original) be self-evident to all observers".

Ian, applicants make a variety of claims. Some claim to detect water using a dowsing rod. Some claim to see without using their eyes. Some claim to be able to sense a human "energy field" using their hands. Each claim will require a different approach to establish proof of that claim.

The person making the claim suggests how he or she will prove it up front. A dowser might say, "Hide some water in boxes on a table. I'll pick out which boxes have the water and which don't."

In a "guessing" situation like this, one has to rule out the possibility that the results couldn't be due to chance alone. Simple statistics help to sort out how many boxes their ought to be (or how many cards out of 20 guessed right, and so on).

If a person claims to read while blindfolded, you don't need statistics. You just have to see evidence the person can read, while ruling out the possibility of trickery.

It's difficult to provide a general structure for how claims are tested, because each claim may require a different approach. I think if you're wondering if the testing procedures are fair or valid, you'd have an easier time forming an opinion about this by reviewing particular examples. I've read about several examples in Randi's weekly commentaries, which you can find on this web site.


If you do not draw up *ANYTHING*, and in fact the claimant does, then, in the card example, there would be nothing to prevent the claimant from drawing up the rules in the following way:

0 = fail (correct/incorrect inapplicable).
1 correct = success
2 correct = success
3 correct = success
4 correct = success
5 correct = success
6 correct = success
7 correct = success
8 correct = success
9 correct = success
10 correct = success
11 correct = success
12 correct = success
13 correct = success
14 correct = success
15 correct = success
16 correct = success
17 correct = success
18 correct =success
19 correct = success
20 correct = success


But using this example, what would the person actually be claiming? That he can guess one out of 20 cards correctly? Anyone can do that, just given the odds of it happening by chance alone. There's nothing extraordinary about that.

Only people who posit extra-ordinary claims can apply for the challenge.

[FutileJester]

Originally posted by Interesting Ian
How the f*ck can someone read a book with a blindfold on?? Is this the sort of stuff that Randi tests??

So you don't think he should? Pretty amazing considering that you just claimed in another thread that the challenge could have no validity unless it tested all claims, no matter how ridiculous. Change your mind? Why am I quite sure that if Randi refused to test this girl, you would be complaining about that as well?

This is supposed to prove or give strong evidence that no anomalous phenomena coming under the purview of the "paranormal" exists whatsoever??

You really like this strawman. For the nth time, the challenge is not evidence of the non-existence of paranormal effects in any way. I don't hear anyone here claiming that. It is an entertaining educational tool.

[Unas]

Originally posted by Whodini
You might have to work sometime. It's your claim. If you are unwilling to back it up, I am not obligated to do the work for you.

[Interesting Ian]

Originally posted by DrBenway
But using this example, what would the person actually be claiming? That he can guess one out of 20 cards correctly? Anyone can do that, just given the odds of it happening by chance alone. There's nothing extraordinary about that.

Only people who posit extra-ordinary claims can apply for the challenge.


But Randi stated that it is purely up to the testee to decide what he can acheive. He stated:

READ THE DAMN RULES! I don't "draw up" ANYTHING! The CLAIMANT does! So long as an independant group -- of HIS choosing -- decides that the rules and conditions are adequate to prove the claim, WE DO IT!


So if Randi doesn't draw up anything, and in fact it is the testee who does this, why can't the testee decide that his challenge will consitute getting just 1 card out of 20?

Of course I agree this is ludicrous amd it is evident that Randi is talking out of his a*se here. So if it is not the testee who decides what he will need to do to pass the challenge, then who does decide??

[Unas]

Originally posted by Interesting Ian
But Randi stated that it is purely up to the testee to decide what he can acheive.False. See below.
Originally posted by Interesting Ian
So if it is not the testee who decides what he will need to do to pass the challenge, then who does decide?? As Randi noted, you didn't read the rules:
So long as an independant group -- of HIS choosing -- decides that the rules and conditions are adequate to prove the claim, WE DO IT!The testee doesn't get to decide whether the rules and conditions are adequate to prove his claim.

Words you will NEVER see a woo-woo post:

You're right. I hadn't thought of it that way.

[Keneke]

borb.

[Interesting Ian]

Originally posted by Unas
Originally posted by Interesting Ian
But Randi stated that it is purely up to the testee to decide what he can acheive.
--------------------------------------------------------------------------------

False. See below.



There is nothing below which vindicates your claim.


--------------------------------------------------------------------------------
Originally posted by Interesting Ian
So if it is not the testee who decides what he will need to do to pass the challenge, then who does decide??
--------------------------------------------------------------------------------

As Randi noted, you didn't read the rules:


--------------------------------------------------------------------------------
So long as an independant group -- of HIS choosing -- decides that the rules and conditions are adequate to prove the claim, WE DO IT!
--------------------------------------------------------------------------------

The testee doesn't get to decide whether the rules and conditions are adequate to prove his claim.




Nor did either Randi or I claim otherwise.

But Randi did claim that the testee chooses for himself what the challenge itself consists in. Randi says:

I don't "draw up" ANYTHING! The CLAIMANT does!

----
It's difficult to provide a general structure for how claims are tested, because each claim may require a different approach.
----


That's true, but a lot of the claims and the testing are VERY similar.

Originally posted by Unas
It's your claim. If you are unwilling to back it up, I am not obligated to do the work for you.


Thaiboxerken,

That's what I thought.

[Unas]

Ian:

You asked "if it is not the testee who decides what he will need to do to pass the challenge, then who does decide?"

That question has been answered, and it is not the testee that decides. Argue with reality, not me.

[Oso]

Originally posted by Interesting Ian
I await your reply with interest.

Ian Wardell


END OF RANDI EXCHANGE

That's an example worth following. I knew that Randi was a sharp guy.

[Interesting Ian]

Originally posted by Unas
Ian:

You asked "if it is not the testee who decides what he will need to do to pass the challenge, then who does decide?"

That question has been answered, and it is not the testee that decides. Argue with reality, not me.

Where was it answered? The only person who has answered is Randi, but you're saying he's wrong. I find it most interesting that you know more about his challenge than he does himself.

[Checkmite]

I was under the impression that the claimant makes his claim and explains exactly what will constitute "success", and the JREF would agree to those terms. Theoretically, the JREF may disagree with the claimant over what a "success" would be, whereupon the claimant proposes a revised definition of "success". To avoid any misunderstandings or disagreements such as the one Ian envisions, no test is performed until all parties fully agree on what "success" is. That is what Randi tried to explain - if the claimant says he can guess at least 18 out of 20 cards correctly, and the JREF agrees to test him, with success based specifically on those conditions (at least 18 out of 20) - then the test is performed, and anything less than 18 out of 20 is a failure. The claimant cannot contest this, because he failed to acheive his own definition of "success". All tests are designed to have such self-evident results.

Let's say a person claims to be able to read any book while blindfolded, for example. The JREF would ask him what he would consider "success". If the person proposes that the correct recital of a page or section of text from a random book would be "success", then the JREF tests him appropriately. If the claimant cannot correctly recite the page or section of text, he fails - by his own definition.

Now, a person could decide that "guessing 20 out of 100 words in a block of text" would be success - but there is nothing paranormal about being able to guess 20 words out of 100, the JREF would probably not agree to such silliness, and the test wouldn't be performed until the claimant proposed something more sensible. In like manner, a person can claim to be able to correctly guess 1 out of 20 cards; but here again, there is no "paranormal" quality being demonstrated, and the JREF would reject the application.

[fishbob]

Whodini sez: So your gold standard is your gold standard even when it is not peer reviewed and not all the information and data is disclosed to all? What data? There is no data until a claimant is successful. No claimants have been successful yet. So what is it you want published?

Ian sez: So if Randi doesn't draw up anything, and in fact it is the testee who does this, why can't the testee decide that his challenge will consitute getting just 1 card out of 20? The testee could certainly claim this, but nobody cares. 1/20 is not exceptional.

Ian also sez:Of course I agree this is ludicrous amd it is evident that Randi is talking out of his a*se here. So if it is not the testee who decides what he will need to do to pass the challenge, then who does decide?? It is pretty evident who is talking through his posterior here. Can you guys read?

[Interesting Ian]

Originally posted by Joshua Korosi
[B]I was under the impression that the claimant makes his claim and explains exactly what will constitute "success", and the JREF would agree to those terms. Theoretically, the JREF may disagree with the claimant over what a "success" would be, whereupon the claimant proposes a revised definition of "success".


Indeed, and JREF would continue to disagree until the claimant proposed a task that they felt was sufficiently unlikely to be fullfilled that they would feel safe in agreeing to it. In other words it is the JREF who dictate how likely or unlikely passing the challenge will be.

So much for the JREF challenge demonstating the non-existence of all paranormal phenomena! :rolleyes:

Fishbob,


----
What data? There is no data until a claimant is successful. No claimants have been successful yet. So what is it you want published?
----


There is plenty of data.

How many of each type of claim.

How many hits out of the number of trials.

The p-values.

Various graphs showing trends of applicant types and results over time.

Detailed write-ups of the experimental design and methodology.

[Stimpson J. Cat]

Ian,

Indeed, and JREF would continue to disagree until the claimant proposed a task that they felt was sufficiently unlikely to be fullfilled that they would feel safe in agreeing to it. In other words it is the JREF who dictate how likely or unlikely passing the challenge will be.

Obviously. Remember that the JREF does not believe any of these people have paranormal abilities. Clearly they aren't going to accept a challenge unless the likelihood of the claimant being able to succeed without paranormal abilities, is very small.

Since they don't believe that such paranormal abilities exist, obviously they are only going to accept challenges that they feel safe in agreeing to. I would think that would be obvious.

Tell me something. If somebody told you that they could accurately guess the flipping of a coin, and bet you $100 that they could, would you take the bet? I would. What would you accept as the threshold of them winning the bet? One successful guess? Two? Ten? I would demand enough successful guesses to rule out chance. And once I agree with them on what the required number is, no interpretation of the results is required. Either they guess enough correctly, or they don't. That is how the JREF challenge works. It is really simple.

So much for the JREF challenge demonstating the non-existence of all paranormal phenomena!

Once again, that is not what it claims to do. Anybody who says that it does, is talking out of their ass.

Dr. Stupid

[rwald]

Ian, for the nth time, the challenge isn't supposed to prove that no paranormal activity exists. It only proves that certain individuals who claim to have paranormal abilities do not, in fact, have those abilities.

I think the way the test is usually drawn up is that the challenger draws up the procedure, and then Randi or a statistician might suggest the specific statistic which needs to be exceeded. If someone said, "I can guess cards when blindfolded," then Randi would say, "How does 18/20 sound to you?" to which the challenger would say, "Fine." Or not. The challenger could say, "I want it to be 1/20." Then, Randi would say, "My statistican here says that there's a 33% chance of getting at least one right randomly. So you'll need a better average than that." I think that there's some percentage, like 1%, which the challenger must "beat" in order to win. I mean, you can't actually expect Randi's going to give out his million for something that's 33% likely? If it can happen either by random chance or by paranormal powers, Randi won't buy it. Only if it can happen by paranormal powers alone.

[Unas]

Originally posted by Interesting Ian
Originally posted by Unas
You asked "if it is not the testee who decides what he will need to do to pass the challenge, then who does decide?"

That question has been answered, and it is not the testee that decides. Argue with reality, not me.Where was it answered?sigh... In your own post (http://www.randi.org/vbulletin/showthread.php?s=&postid=362022#post362022), Ian:

So long as an independant group -- of HIS choosing -- decides that the rules and conditions are adequate to prove the claim, WE DO IT!
Originally posted by Interesting Ian
The only person who has answered is Randi, but you're saying he's wrong.I have made no such statement, Ian. You should try to avoid becoming another Franko.

[Unas]

Originally posted by Interesting Ian
Indeed, and JREF would continue to disagree until the claimant proposed a task that they felt was sufficiently unlikely to be fullfilled that they would feel safe in agreeing to it. In other words it is the JREF who dictate how likely or unlikely passing the challenge will be.Many claimants have already been tested by JREF.

Cite one instance in which the scenario you postulate actually occurred.

[Soapy Sam]

jj -"On the other hand, such information ought to be published somewhere, I'm just not sure where".

Perhaps we should do it. A scientific journal (apart from "popular " ones such as New Scientist) would not be interested. But the people we want to reach are not likely to read science journals anyway. One feature of several threads on this forum has been that most of us held ideas in our youth which we are embarrassed by later. Younger people rarely read science journals. They might benefit from an easy to find series of essays on why and how issues here have been argued in the past- and yes, that should include comment from the naysayers.

Andonyx has proposed in Paranormal that we should ourselves organise a debunking resource - a "woo-woo faq" database- a thread I for one intend to keep bumping for a week or so.

There are others, but perhaps we have a role to play? We've all seen how the same points keep coming up every few months.

OK, we don't all agree on the answers, but at least we might compile some of the better threads, to stop people reinventing the wheel. And why not include the results of the tests? Including Sylvia (No-show) Browne?

[DrBenway]

Ian,

In physics, an experiment is considered significant if its p-value is .01 percent. That means that only one in 10,000 similar experiments would be expected to produce the reported effect or a greater effect due to chance alone.

I'm not a statistics expert. I'm guessing that the JREF would use a similar p value level in any study involving repeated guesses that might be right simply by chance alone. Using the rule of thumb that "extraordinary hypotheses require extraordinary evidence," I would hate to think that we'd expect less proof of the paranormal than we demand of normal phenomenon.

It isn't Randi who sets the significance level.

So far, no applicant has ever made it past a preliminary screen. I don't think the preliminary screens set the significance level so high, as less effort is made to perfectly control all variables. Something like 1/50 is what I've seen someplace.

Whodini,

There isn't much *data* as of yet to publish (mostly just the sound of crickets chirping). If you read Randi's weekly commentaries, you'll encounter descriptions of what's happened with the few brave applicants who've tried to take on the challenge so far.

DrBenWay,


----
I'm not a statistics expert. I'm guessing that the JREF would use a similar p value level in any study involving repeated guesses that might be right simply by chance alone.
----


You mean a similar level of alpha. (.05, .01, etc)

I'd bet the JREF uses (and rightly so) a VERY small alpha level, because they want to make sure that someone doesn't win just by chance, but rather that it will be clear that the person won because of their paranormal skill, etc.


----
There isn't much *data* as of yet to publish (mostly just the sound of crickets chirping). If you read Randi's weekly commentaries, you'll encounter descriptions of what's happened with the few brave applicants who've tried to take on the challenge so far.
----


That is true. I'd like to see some synthesis of those descriptions of the experimental designs and methods. So for example, one could refer to the Dowsing page, and see a good design that one could replicate.

I understand that a problem with that is that all claims are different, but it could be kept general ('if a dowser claims to be able to detect k out of n, then set up n buckets', etc.) and at least that would a start.

[Skeptical Greg]

Originally posted by Whodini
One aspect of scientific endeavors is that the specifc data, methods, and analyses are made available to all who are interested.

The results of the Challenge clearly don't apply.

I can understand the legal matters, which is why we only get tidbits of vague happenings, but to be taken seriously by the entire scientific community (as opposed to a small subset), wouldn't JREF's case be better served if they had some type of peer reviewed journal their findings were published in??

I believe that is why a lot of people have a hard time taken the JREF proceedings as anything other than mildly interesting and entertaining.

Whodini,
As several people have alluded to, The JREF challenge does not claim to be scientific research.
There is no reason for the JREF to seek peer review.

The other point that bears repeating, is that peer review is usually sought, when some findings have been made.

Can you please give us an example, of a mainstream academic publication, that is known for publishing failed experiments.

Originally posted by Diogenes


Whodini,
As several people have alluded to, The JREF challenge does not claim to be scientific research.
There is no reason for the JREF to seek peer review.

The other point that bears repeating, is that peer review is usually sought, when some findings have been made.

Can you please give us an example, of a mainstream academic publication, that is known for publishing failed experiments.

It scares me when you make sense.

[Skeptical Greg]

Originally posted by sundog


It scares me when you make sense.

This is the new me...:D

[DrBenway]

Originally posted by Diogenes


Whodini,
As several people have alluded to, The JREF challenge does not claim to be scientific research.

If the evaluations of claims of the paranormal are not done scientifically, then I want my donation to the JREF returned right now! :p

Clearly, Randi is not a scientist. However, scientists *are* involved in tests of the challengers.


There is no reason for the JREF to seek peer review.

"Peer review," in its broad sense, means that methods employed are consistent with acceptable scientific procedures. As an example: the p-value of .01% is a common standard of significance in physics. Using such a standard in a JREF test, thus, has robust precedence. It meets communal (i.e., "peer review") standards of good science.


The other point that bears repeating, is that peer review is usually sought, when some findings have been made.

Right. Editorial peer review can only happen when there exists a paper to edit.


Can you please give us an example, of a mainstream academic publication, that is known for publishing failed experiments.

All of them, so long as the negative results are interesting (failing to report negative results contributes to the "file drawer effect).

[jj]

Originally posted by Interesting Ian


Well that's interesting. So despite the collosal amount of evidence for anomolous cognitive abilities it is obvious they don't exist :rolleyes:

Might I suggest the JREF is lamentably failing in its mission to educate people if the attitude that you express is typical.

Edited to add:

BTW I'm not suggesting such phenomena definitely does exist, but to to claim that it is obvious that no "psi" exists is clearly preposterous. Go back to arguing with Win.

Could you try to troll a bit harder? I am on several review boards, and I know what the review boards generally think.

And, if there was any evidence, let alone "collossal", you'd certainly see papers in a flash.

So take your straw men and burn them in somebody else's yard.

Diogenes,


----
As several people have alluded to, The JREF challenge does not claim to be scientific research.
There is no reason for the JREF to seek peer review.
----


I understand that point Diogenes. My point is that don't expect the bulk of non-skeptics to take the JREF challenge as seriously as the bulk of skeptics do.

Because of the non-science, I'd suspect most consider it interesting entertainment.

For what its worth, I consider it somewhere between. It is useful, but not really compelling.


----
Can you please give us an example, of a mainstream academic publication, that is known for publishing failed experiments.
----


First, as I restate later, there is probably no article in existence that is only interested in testing one hypothesis. Part of the advances in experimental design and software let us test many things at a time.

Also, there are probably no publications that "are known" for publishing failed experiments, but there are many examples where failed experiments (which is a very poor choice of words. By "failed experiments" you mean a p-value >.05 (statistically nonsignificant)) were published.

Here are a few related items:

An abstract of an article that talks about that:
http://archpedi.ama-assn.org/issues/v156n5/abs/poa10294.html

"Abstracts were more likely to be published as a full study if, using some criteria for statistical significance, they either reported that the outcomes of the treated group were superior to those of the control group or reported that a newer therapy was at least equivalent to an older treatment."


You can read some of these papers, by JAMA, on the problem:

http://www.ama-assn.org/public/peer/7_15_98/jpv71013.htm

"Our results indicate that statistically significant studies are published sooner than statistically nonsignificant studies, as reported for clinical studies. This finding cannot be attributed to differences in the sample sizes, funding sources, or health outcomes measured. Our finding of a publication delay can be partially explained by the reasons that investigators cited for not publishing. "


http://jama.ama-assn.org/issues/v287n21/ffull/joc10235.html

"Conclusions Among submitted manuscripts, we did not find a statistically significant difference in publication rates between those with positive vs negative results."

There are many many many published articles, of course, where parts of the results are statistically significant, and parts are statistically non-significant, like

A 3-Year Study of Water Relations of Urban Street Trees
Thomas H. Whitlow; Nina L. Bassuk; Deborah L. Reichert
Journal of Applied Ecology, Vol. 29, No. 2. (1992), pp. 436-450.

Reproductive Investment and Parasite Susceptibility in the Great Tit (no, not the kind you are thinking of Diogenes)
K. Allander
Functional Ecology, Vol. 11, No. 3. (Jun., 1997), pp. 358-364.

Viability is Positively Related to Degree of Ornamentation in Male Swallows
Anders Pape Moller
Proceedings: Biological Sciences, Vol. 243, No. 1307. (Feb. 22, 1991), pp. 145-148.

Solar Irradiance, Air Pollution and Temperature Changes in the Arctic
G. Stanhill
Philosophical Transactions: Physical Sciences and Engineering, Vol. 352, No. 1699, The Arctic and Environmental Change. (Aug. 15, 1995), pp. 247-258.

Synergistic Selection of Unpalatability in Plants (in Notes and Comments)
Juha Tuomi; Magnus Augner
Evolution, Vol. 47, No. 2. (Apr., 1993), pp. 668-672.

Longevity in Drosophila melanogaster and its Ebony Mutant in the Absence of Food
Donald Greiff
American Naturalist, Vol. 74, No. 753. (Jul. - Aug., 1940), pp. 363-376.

Rarer Need not Be Better if Commoner Is Worse: Frequency-Dependent Selection for Developmental Time at the Alcohol Dehydrogenase Locus of the Olive Fruit Fly, Bactrocera oleae
Nikos Cosmidis; Michael Loukas; Eleftherios Zouros
Evolution, Vol. 53, No. 2. (Apr., 1999), pp. 518-526.

Factors Affecting the Density and Distribution of Wild Dogs in the Kruger National Park
Michael G. L. Mills; Martyn L. Gorman
Conservation Biology, Vol. 11, No. 6. (Dec., 1997), pp. 1397-1406.

There are many many (hundreds and hundreds of thousands) of these because almost all the time one is not interested in JUST one thing in research, but rather many many things.

There are entire reams of literature on publication bias, volcano plots, etc.

A growing trend now is to report measures of effect size and confidence intervals, rather than using just using the p value.

[jj]

Originally posted by Diogenes


Whodini,
As several people have alluded to, The JREF challenge does not claim to be scientific research.
There is no reason for the JREF to seek peer review.

The other point that bears repeating, is that peer review is usually sought, when some findings have been made.

Can you please give us an example, of a mainstream academic publication, that is known for publishing failed experiments.
Well, I won't say failed experiment, but it's a negative result:

Kuo, S. and Johnston, J. D., ``A study of why cross channel prediction is not applicable to perceptual audio coding,'' 111 AES Convention, New York, NY, 21-24 September 2001.

Now, that is in fact a convention paper, not a journal paper, so it's not all the way there. It just happened to be at hand when you brought up the subject. The AES is a mainstream professional society, at least.

But some times there are negative results published. I'm not quite sure what a 'failed experiment' is, though. Is it one that did not provide a new gain, or is it something that showed that a hypothesis was wrong, or what?

[Interesting Ian]

Originally posted by jj
[B]

Could you try to troll a bit harder?

F*ck off a*sehole.




I am on several review boards, and I know what the review boards generally think.


What the f*ck is a review board??


And, if there was any evidence, let alone "collossal", you'd certainly see papers in a flash.



No you wouldn't, retard.

----
There is no reason for the JREF to seek peer review.
----


Diogenes,

BTW, so far, 13 out of 20 disagree with you.

[Oso]

Originally posted by Whodini
----
There is no reason for the JREF to seek peer review.
----

Diogenes,

BTW, so far, 13 out of 20 disagree with you. Hmm... I'd like to see that list.

[Interesting Ian]

Originally posted by Stimpson J. Cat
Ian,



quote:
--------------------------------------------------------------------------------
Indeed, and JREF would continue to disagree until the claimant proposed a task that they felt was sufficiently unlikely to be fullfilled that they would feel safe in agreeing to it. In other words it is the JREF who dictate how likely or unlikely passing the challenge will be.
--------------------------------------------------------------------------------



Obviously. Remember that the JREF does not believe any of these people have paranormal abilities. Clearly they aren't going to accept a challenge unless the likelihood of the claimant being able to succeed without paranormal abilities, is very small.


Yes I know that! :rolleyes:



Since they don't believe that such paranormal abilities exist, obviously they are only going to accept challenges that they feel safe in agreeing to. I would think that would be obvious.



Indeed, so why did Randi claim the decision on what constitutes the task the testee needs to perform is entirely up to the testee? It seems the converse is the case. Namely it is entirely up to the JREF.



Tell me something. If somebody told you that they could accurately guess the flipping of a coin, and bet you $100 that they could, would you take the bet? I would. What would you accept as the threshold of them winning the bet? One successful guess? Two? Ten? I would demand enough successful guesses to rule out chance. And once I agree with them on what the required number is, no interpretation of the results is required. Either they guess enough correctly, or they don't. That is how the JREF challenge works. It is really simple.



{sighs}

I was simply trying to establish who decides what the testee needs to acheive. If it is up to the JREF they can ensure that the probability is sufficiently low that the testee won't pass. Crucially this is so even if the testee actually possess quite stupendious paranormal powers, such as being able to guess on average 1 out of 9 times rather than 1 in 10 which one would expect by chance.



--------------------------------------------------------------------------------
So much for the JREF challenge demonstating the non-existence of all paranormal phenomena!
--------------------------------------------------------------------------------



Once again, that is not what it claims to do. Anybody who says that it does, is talking out of their ass.


People in this forum seem to be much more reasonable then the people I used to talk to in newsgroups. They were absolutely emphatic that the failure of anyone to pass the challenge was damning evidence against any paranormal phenomena whatsoever. Indeed the fact they were so absolute completely sure about it was the reason I contacted Randi in the first place. Absolutely complete and total unadulterated a*seholes.

Randi's challenge constitutes no evidence whatsoever against the putative phenomena which parapsychological research seems to suggest exists, nor does it constitute any evidence against those spontaneous unpredictable anecdotes of universally occuring anomalous phenomena coming under the purview of the "paranormal".

[Skeptical Greg]

Originally posted by Whodini
----
There is no reason for the JREF to seek peer review.
----


Diogenes,

BTW, so far, 13 out of 20 disagree with you.

Appeal to popularity...:rolleyes:

[Skeptical Greg]

Originally posted by jj

Well, I won't say failed experiment, but it's a negative result:

Kuo, S. and Johnston, J. D., ``A study of why cross channel prediction is not applicable to perceptual audio coding,'' 111 AES Convention, New York, NY, 21-24 September 2001.

Now, that is in fact a convention paper, not a journal paper, so it's not all the way there. It just happened to be at hand when you brought up the subject. The AES is a mainstream professional society, at least.

But some times there are negative results published. I'm not quite sure what a 'failed experiment' is, though. Is it one that did not provide a new gain, or is it something that showed that a hypothesis was wrong, or what?

Thanks for taking the time..

I did not question the possibility that negative results are published..

I asked:
Can you please give us an example, of a mainstream academic publication, that is known for publishing failed experiments.

This is important; because failures are all that JREF has to offer.

Since Whodini hasn't given us an example of a peer-reviwed journal, that would lend some credence,
to his poll, it is difficult to address the matter of whether or not JREF testing should be reviewed or not.

There is also the matter regarding the fact, that JEREF testing is not (and doesn't claim to be) scientific research.
( Is there a bad echo in here, or is it just me?)

[Skeptical Greg]

Originally posted by Oso
Hmm... I'd like to see that list.


He's referring to his poll, so far.

----
There is no reason for the JREF to seek peer review.
----


Appeal to wishful thinking and personal opinion.


Originally posted by Diogenes


Appeal to popularity...:rolleyes:


Appeal to EVIDENCE...

but it is appeal to authority when the evidence is against your opinion apparently.

[rwald]

Just for the record, it's not that scientific journals don't publish "failed experiments," it's that they're not going to publish an experiment showing that dousing isn't real. I concur with whoever said that they'll only publish "interesting" results. In some fields, any result is "interesting." In the paranormal, any good science journal will only find positive results "interesting."

----
This is important; because failures are all that JREF has to offer.
----


What about experimental design and methodology?

What about trends in applicant type?


----
Since Whodini hasn't given us an example of a peer-reviwed journal, that would lend some credence,
to his poll, it is difficult to address the matter of whether or not JREF testing should be reviewed or not.
----


As I mentioned at the beginning, creating a journal is also a possibility.

The parapsychological journals are peer reviewed Diogenes.


----
There is also the matter regarding the fact, that JEREF testing is not (and doesn't claim to be) scientific research.
----


And again (echo in here), I never claim that JREF does. I simply stated that you shouldn't expect the bulk of non-skeptics to take any of the JREF stuff as seriously as the bulk of skeptics do. In fact, I'd wager the bulk of non-skeptics consider it interesting entertainment.

Also, referencing JREF's results would be taken more seriously. One could say: "See Journal XYZ, 'Dowsing Trial # 721', Vol 8, # 4, Mar 2003", instead of "See the web page." or "Cuz Dio said so!"

LOL

[Skeptical Greg]

Originally posted by Whodini
Diogenes,

I understand that point Diogenes. My point is that don't expect the bulk of non-skeptics to take the JREF challenge as seriously as the bulk of skeptics do.






Thanks for taking the time to list some examples of negative results that were published, but I guess I didn't make myself clear in that regard....
You still haven't suggested a peer review journal that would be appropriate for reviewing JREF testing.

You would want someone besides JREF to review the stuff, wouldn't you?

And as far as:
don't expect the bulk of non-skeptics to take the JREF challenge as seriously as the bulk of skeptics do
Why would we? ( us skeptics )... That would be like saying:

"don't expect the bulk of Atheists to take God as seriously as the bulk of Christians do ..."

[Denise]

Whodini, for the record I don't think there is anything wrong with giving your thread 5 stars at the start.

Having said that, do you realize that JREF does not have a huge amount of money? Yes, they have the million, but they can't touch that. What you would propose would cost money for experts etc..

Now, if someone were to volunteer to write a paper and go down to Florida to review the records, I'm sure it wouldn't be a problem.

[Skeptical Greg]

Originally posted by Whodini
----
There is no reason for the JREF to seek peer review.
----


Appeal to wishful thinking and personal opinion.





Appeal to EVIDENCE...

but it is appeal to authority when the evidence is against your opinion apparently.

Yes Whodini, the results of your poll is evidence, of what some people believe.

Diogenes,


14 out of 21 now.

Diogenes,


----
You still haven't suggested a peer review journal that would be appropriate for reviewing JREF testing.
----


I suggested the idea. Because I did, that does not require me to list out all the details of the idea until completion.

I gave the suggestions, already, of: Skeptical Enquirer, Skeptic, Science, Nature, and various journals of parapsychology. I also suggested that some new journal could be created.

[Skeptical Greg]

Originally posted by Whodini
----
This is important; because failures are all that JREF has to offer.
----


What about experimental design and methodology?

What about trends in applicant type?
The experiments are designed by the applicants. Why aren't you so adamant that they submit to peer review?

----
Since Whodini hasn't given us an example of a peer-reviwed journal, that would lend some credence,
to his poll, it is difficult to address the matter of whether or not JREF testing should be reviewed or not.
----


As I mentioned at the beginning, creating a journal is also a possibility.

JREF should create a journal to review their own work?

The parapsychological journals are peer reviewed Diogenes.

Parapsychologists are JREF's peers?
----
There is also the matter regarding the fact, that JEREF testing is not (and doesn't claim to be) scientific research.
----


And again (echo in here), I never claim that JREF does. I simply stated that you shouldn't expect the bulk of non-skeptics to take any of the JREF stuff as seriously as the bulk of skeptics do. In fact, I'd wager the bulk of non-skeptics consider it interesting entertainment.

Answered earlier.. ( we seem to be cross posting a bit.. I'm gone for now..

Also, referencing JREF's results would be taken more seriously. One could say: "See Journal XYZ, 'Dowsing Trial # 721', Vol 8, # 4, Mar 2003", instead of "See the web page." or "Cuz Dio said so!"

LOL

Denise,

You're back!


----
Having said that, do you realize that JREF does not have a huge amount of money? Yes, they have the million, but they can't touch that. What you would propose would cost money for experts etc..
----


I don't know Denise, but good questions.

I just proposed the idea.

I'm sure getting donations from interested people and academia wouldn't be difficult.

[DrBenway]

Originally posted by Interesting Ian
Crucially this is so even if the testee actually possess quite stupendious paranormal powers, such as being able to guess on average 1 out of 9 times rather than 1 in 10 which one would expect by chance.
If the JREF uses the p<.01%, the experiment will meet the standard set for publication in one of the physical science journals. The JREF could pick a greater p-value, but the scientific world wouldn't take the experiment seriously.

The low p-value isn't there to harass the paranormal world. All studies in the physical sciences have to meet the same standard.

When the effect being studied is subtle, such as in the case you describe, where a psychic force might alter an expected 1/10 chance of a correct guess to a 1/9 chance, it is still possible to demonstrate that effect with a p<.01%. The way to do this is by increasing "n" or the number of observations of the phenomenon in question. If I weren't so completely rusty with my statistics, I could do a little math and tell you exactly how many trials you'd need to potentially meet the significance criteria. :)

Diogenes,


----
The experiments are designed by the applicants. Why aren't you so adamant that they submit to peer review?
----


Because it would be interesting to see? It would be interesting to catalogue the most common types of experimental designs for testing paranormal claims. True, a lot of them will differ in specifics, but general ideas are what are important.

It is valuable information.


----
JREF should create a journal to review their own work?
----


A journal could be created. Obviously it is not wise to have JREF create their own journal. But say some other entity creates a new journal for JREF results.


----
Parapsychologists are JREF's peers?
----


Well, it was what came to mind right away. I'm sure there are better choices.


----
( we seem to be cross posting a bit.. I'm gone for now..
----


As long as we aren't cross dressing.

[rwald]

Just for the record, I personally don't think I'd find the experimental designs particularly interesting. Most of them basically write themselves, if you think about the claim in a scientific manner.

What would be interesting, however, is a list basically describing what it is that is being claimed. That would be useful information.

[Skeptical Greg]

Originally posted by Whodini

I'm sure getting donations from interested people and academia wouldn't be difficult.

We are anticipating the results of your fund raising efforts..


One of the obvious things you seem to be overlooking Whodini, is that all of those non-skeptics,
who feel the JREF challenge is so unfair, and so un-winable, can't manage to put together their
own concerted effort, to demonstrate that it is.. (unfair, etc..)

As others have alluded to, if people like John Edwards and Sylvia Brown can make a decent living, think
what a real psychic could do.

[Unas]

Originally posted by Interesting Ian
Indeed, so why did Randi claim the decision on what constitutes the task the testee needs to perform is entirely up to the testee?What do the challenge rules say, Ian?
Originally posted by Interesting Ian
It seems the converse is the case.Why? Give an actual example of a preliminary test in which the applicant did not decide what task he would perform.
Originally posted by Interesting Ian
Namely it is entirely up to the JREF.Again: Cite an actual example in which this was the case.
Originally posted by Interesting Ian
I was simply trying to establish who decides what the testee needs to acheive.What do the challenge rules say, Ian?
Originally posted by Interesting Ian
Randi's challenge constitutes no evidence whatsoever against the putative phenomena which parapsychological research seems to suggest exists...It has never been represented as such by JREF.
Originally posted by Interesting Ian
...nor does it constitute any evidence against those spontaneous unpredictable anecdotes of universally occuring anomalous phenomena coming under the purview of the "paranormal". No collection of "spontaneous unpredictable anecdotes" can serve as scientific evidence for paranormal phenomena, either.

----
What would be interesting, however, is a list basically describing what it is that is being claimed. That would be useful information.
----


Hi Rwald,

I'd also be interested in that, as well as a graph of the types of claims, broken down by total % and year.

So one could see the trends.

[rwald]

It boils down to the JREF and the challenger coming to a consensus about what the specific procedure will be. If one side doesn't agree, they continue negotiating. There's no complexity to it.

----
We are anticipating the results of your fund raising efforts..
----


Diogenes,

You know me raising money or not raising money has nothing to do with the validity of my points raised.

Insert coin, try again.


----
One of the obvious things you seem to be overlooking Whodini, is that all of those non-skeptics
who feel the JREF challenge is so unfair, and so un-winable, can't manage to put together their
own concerted effort to demonstratethat it is so.
----


Ok...

which has nothing to do with people (skeptics and non-skeptics) feeling the need for a peer reviewed journal.

I've got to go work out, then do some t'ai chi sword. I'll write more later. :)

[DrBenway]

Originally posted by Denise
Whodini, for the record I don't think there is anything wrong with giving your thread 5 stars at the start.

Having said that, do you realize that JREF does not have a huge amount of money? Yes, they have the million, but they can't touch that. What you would propose would cost money for experts etc..

I imagine a solution to Whodini's basic question (as I think I understand it), which would be quite cheap:

1. Someone go through Randi's weekly commentaries and pick out the ones which specifically describe preliminary trials of JREF challengers (there haven't been that many).
2. Make a web page with a brief description of the claim being tested, and a link to the commentary with the information.

Then, visitors to this site can quickly find details about specific claims the JREF has evaluated, without having to wade through a lot of other stuff.

I don't think there's any need to dress up the trials that I've read about with fancy sounding technical terms. Some of the best contributions to scientific knowledge haven't needed ANOVAs, a specific editorial review board, etc. Sometimes the facts, like dinosaur bones, just speak for themselves.

[TechHead]

Originally posted by Unas
Are there any specific JREF challenges for which the specific data, methods, and analyses are unavailable?

There is at least one. In 1987, Randi tested a Mr. Dell Winders of Haines City, FL for dowsing ability. According to Mr. Winders, he correctly dowsed 8 of 12 times, which was statistically very significant for that particular test protocol.

According to Winders, Randi had denied the test ever took place. I once asked Randi about the test, and after a bit of discussion, Randi seemed to have a vague recollection of the event. But, he stated that Winders failed the test, even though he had no documentation to show that to be true.

Did the test really take place? Yup, I even have a partial videotape of the test being set up, with Randi & Winders discussing the protocol. Winders says it was filmed by a television station, on the beach (probably Ft. Lauderdale). Yet, Randi had no copy of the tape, I got my copy from Winders. The tape abruptly ends before the actual test begins, so it does not support either man's claim. So, here we have a test that really took place, that the claimant says he passed, yet Randi has no documentation of the test.

I know that Randi has files on other tests, and those files are available for perusal by anyone who wishes to visit JREF (I've been there). But are ALL the tests on file, available to the public? Is it possible that some tests do not show a lack of demonstration, i.e., the claimant was "successful"? Perhaps a study of those tests reveal, not true successes, but weaknesses in the test protocol, that was accounted for in subsequent tests. Even that information is valuable.

In my opinion, it is not necessary that the tests be published in a peer-reviewed journal. But, I think it is prudent to have all the tests that have been conducted, available without having to visit JREF. The web is ideal for this. At the very least, a searchable online database of test summaries, that give the participants, claims, protocol, and results. Claimants sign a contract allowing JREF to publish this information, and JREF should do so.

[Skeptical Greg]

Originally posted by Whodini
----
We are anticipating the results of your fund raising efforts..
----


Diogenes,

You know me raising money or not raising money has nothing to do with the validity of my points raised.

Insert coin, try again.
Says you...

----
One of the obvious things you seem to be overlooking Whodini, is that all of those non-skeptics
who feel the JREF challenge is so unfair, and so un-winable, can't manage to put together their
own concerted effort to demonstrate that it is unfair.
----


Ok...

which has nothing to do with people (skeptics and non-skeptics) feeling the need for a peer reviewed journal.

... one more time, after you've slept on it.. Why review something that is not being contested
in any kind of concerted effort by the opposing camp?



I've got to go work out, then do some t'ai chi sword. I'll write more later. :)

[Skeptical Greg]

Originally posted by TechHead


There is at least one. In 1987, Randi tested a Mr. Dell Winders of Haines City, FL for dowsing ability. According to Mr. Winders, he correctly dowsed 8 of 12 times, which was statistically very significant for that particular test protocol.

According to Winders, Randi had denied the test ever took place. I once asked Randi about the test, and after a bit of discussion, Randi seemed to have a vague recollection of the event. But, he stated that Winders failed the test, even though he had no documentation to show that to be true.

Did the test really take place? Yup, I even have a partial videotape of the test being set up, with Randi & Winders discussing the protocol. Winders says it was filmed by a television station, on the beach (probably Ft. Lauderdale). Yet, Randi had no copy of the tape, I got my copy from Winders. The tape abruptly ends before the actual test begins, so it does not support either man's claim. So, here we have a test that really took place, that the claimant says he passed, yet Randi has no documentation of the test.

I know that Randi has files on other tests, and those files are available for perusal by anyone who wishes to visit JREF (I've been there). But are ALL the tests on file, available to the public? Is it possible that some tests do not show a lack of demonstration, i.e., the claimant was "successful"? Perhaps a study of those tests reveal, not true successes, but weaknesses in the test protocol, that was accounted for in subsequent tests. Even that information is valuable.

In my opinion, it is not necessary that the tests be published in a peer-reviewed journal. But, I think it is prudent to have all the tests that have been conducted, available without having to visit JREF. The web is ideal for this. At the very least, a searchable online database of test summaries, that give the participants, claims, protocol, and results. Claimants sign a contract allowing JREF to publish this information, and JREF should do so.


So what is Mr. Dell Winders doing these days?

How did this undocumented test affect his abilities?

[Denise]

Originally posted by TechHead


At the very least, a searchable online database of test summaries, that give the participants, claims, protocol, and results. Claimants sign a contract allowing JREF to publish this information, and JREF should do so.

Are you willing to put up the money for this? Or are you willing to spend the time to do all this?

[TechHead]

Originally posted by Diogenes



So what is Mr. Dell Winders doing these days?

How did this undocumented test affect his abilities?

Mr. Winders had been selling high-tech dowsing devices, for about $2000 a pop. And using his supposed success in the Randi test as evidence that his gadgets work. Since Randi either has no record of the test, or refuses to divulge it, then there is no challenge to Winders' claim. In fact, the video tends to refute Randi's claim that the test never took place, lending more credibility to Winders, than Randi, as hard as it is for me to say that.

This is a case where Randi's test has helped to promote dowsing.

[TechHead]

Originally posted by Denise


Are you willing to put up the money for this? Or are you willing to spend the time to do all this?

JREF is the one in the business of testing people, and claim to be in the business of educating people. Posting test information would be incredibly educational. The incremental cost in writing up a summary of a test and posting it on the web, is close to zero compared to the test itself. Especially if they are properly documenting the test on paper, as they should be.

I admit that it would be a major undertaking to go back and create summaries of past tests. But I still believe it should be done, and JREF should at least make the effort to do so. I would gladly contribute to the effort, if such an effort were organized.

Personally, I do engineering analyses of quack dowsing devices, that cost 100's to 1000's of dollars (including one of Winders'!), and I do so entirely at my expense. I then write web-based reports on them, and make them freely available to the public.

[Denise]

Originally posted by TechHead


JREF is the one in the business of testing people, and claim to be in the business of educating people. Posting test information would be incredibly educational. The incremental cost in writing up a summary of a test and posting it on the web, is close to zero compared to the test itself. Especially if they are properly documenting the test on paper, as they should be.

I admit that it would be a major undertaking to go back and create summaries of past tests. But I still believe it should be done, and JREF should at least make the effort to do so. I would gladly contribute to the effort, if such an effort were organized.

Personally, I do engineering analyses of quack dowsing devices, that cost 100's to 1000's of dollars (including one of Winders'!), and I do so entirely at my expense. I then write web-based reports on them, and make them freely available to the public.

I assume you realize that this would involve evaluating at least 30 years of data. Why don't you email Randi about your willingness to support this idea. His email is Randi@randi.org.

[Denise]

Originally posted by TechHead


Mr. Winders had been selling high-tech dowsing devices, for about $2000 a pop. And using his supposed success in the Randi test as evidence that his gadgets work. Since Randi either has no record of the test, or refuses to divulge it, then there is no challenge to Winders' claim. In fact, the video tends to refute Randi's claim that the test never took place, lending more credibility to Winders, than Randi, as hard as it is for me to say that.

This is a case where Randi's test has helped to promote dowsing.

Mr. Winders makes a claim he is unable to substantiate. Because Randi does not have the data to show that Mr. Winders succeeded, does not by default mean that Mr. Winder's assertion is a correct one. That is critical thinking 101.

[rwald]

If one could pass critical thinking 101, one wouldn't be in the market for a dowsing device...

(I didn't use haikus! So there!)

Hi TechHead,


----
, I think it is prudent to have all the tests that have been conducted, available without having to visit JREF. The web is ideal for this. At the very least, a searchable online database of test summaries, that give the participants, claims, protocol, and results. Claimants sign a contract allowing JREF to publish this information, and JREF should do so.
----


I agree, and would add that it is not necessary to publish names at all, but just things like

a) date
b) location
c) type of claim
d) experimental design
e) methodology
f) results
g) # of times applicant has applied

, with the option of b,c,f, and g, being viewed as a % of the total in, say, a pie chart, and being viewed over time so one can see the trends.

As I've said, JREF is under NO obligation whatsoever to do it, but it would be interesting, and add more weight to their results.

Diogenes,


----
Why review something that is not being contested in any kind of concerted effort by the opposing camp?
----


Any kind of concerted effort? hehe, like that is some criteria for needing a journal?

People are interested in the information, period.

TechHead,


----
Posting test information would be incredibly educational. The incremental cost in writing up a summary of a test and posting it on the web, is close to zero compared to the test itself. Especially if they are properly documenting the test on paper, as they should be.
----


I agree TechHead. It would be a major pain to go back and write up all the results from the past, but... I'd at least hope they'd have some already typed up and not just sitting in files. :)

But, after designing some sort of computer form for easy entry, future entries would be effortlessly typed in, and database retrieval would be a snap.

Dangit Rwald, now I'm going to have to get used to your new avatar.

Just for that, I'm going to change my avatar too.
:)

DrBenWay,


----
I don't think there's any need to dress up the trials that I've read about with fancy sounding technical terms. Some of the best contributions to scientific knowledge haven't needed ANOVAs, a specific editorial review board, etc. Sometimes the facts, like dinosaur bones, just speak for themselves.
----


I agree with that DrBenWay. Some of the best analysis I've seen has been basic, but powerful, descriptive statistics.

[rwald]

Well, I figured, I seem to be arguing evolution more than quantum physics, so my avatar should reflect that. But I'll leave my title, at least until I think of something better.

I can't wait to see what you come up with...;)

[Denise]

Whodini, would you volunteer for such an activity? In other words would you take the time to set up the data base etc? I don't think Randi would be opposed, but I am not positive on that. We could get up a collection to send you down there for a week to do the work? What do you think? I for one would pledge 50 dollars towards the trip, and I am not rich.

Denise,


I have obligations where I am at now in life.

Right now I am finishing school.

Then I am flying to spend some time with my grandmother.

Next I probably will be moving to MD to work at a government job if they enjoyed my application.

I would, however, still accept your 50 dollars.

[Denise]

Originally posted by Whodini
Denise,


I have obligations where I am at now in life.

Right now I am finishing school.

Then I am flying to spend some time with my grandmother.

Next I probably will be moving to MD to work at a government job if they enjoyed my application.

I would, however, still accept your 50 dollars.

Ha ha! Sorry then!

Anyhow, why don't we try and find someone who would be willing to do this? It would be fascinating wouldn't it?

[Interesting Ian]

Originally posted by Denise
Whodini, for the record I don't think there is anything wrong with giving your thread 5 stars at the start.

Having said that, do you realize that JREF does not have a huge amount of money? Yes, they have the million, but they can't touch that. What you would propose would cost money for experts etc..

Now, if someone were to volunteer to write a paper and go down to Florida to review the records, I'm sure it wouldn't be a problem.

You're back! :D That showed people huh? Being deprived of your delectable presence for 3 days! ;) :)

[TechHead]

Originally posted by Denise


Mr. Winders makes a claim he is unable to substantiate. Because Randi does not have the data to show that Mr. Winders succeeded, does not by default mean that Mr. Winder's assertion is a correct one. That is critical thinking 101.

A-yup, and I'm not saying that Winders is correct. However, a few years ago when Winders was still busy selling dowsing gadgets, he discussed his Randi test on a forum. A number of people called Randi directly, and were told that he knew nothing of Winders, and that the test never took place. Once the video surfaced showing otherwise, Randi could no longer deny the event happened, and now simply says that Winders failed.

Again, neither man can substantiate the results of the test (or won't), but from a credibility standpoint, Winders has a leg up. Randi is pursuing a noble goal with JREF, and generally does a good job, but is not perfect, and should be more open with his methods and results.

[Skeptical Greg]

Originally posted by TechHead


Mr. Winders had been selling high-tech dowsing devices, for about $2000 a pop. And using his supposed success in the Randi test as evidence that his gadgets work. Since Randi either has no record of the test, or refuses to divulge it, then there is no challenge to Winders' claim. In fact, the video tends to refute Randi's claim that the test never took place, lending more credibility to Winders, than Randi, as hard as it is for me to say that.

This is a case where Randi's test has helped to promote dowsing.


If the device works, it is all the evidence Mr. Winders needs.
And he should consider the JREF mil a pittance and the challenge a waist of time.

[scotth]

Originally posted by Denise
Whodini, would you volunteer for such an activity? In other words would you take the time to set up the data base etc? I don't think Randi would be opposed, but I am not positive on that. We could get up a collection to send you down there for a week to do the work? What do you think? I for one would pledge 50 dollars towards the trip, and I am not rich.

Not addressed to me, but I'll answer anyway.

I build database driven web applications for a living.

This sounds like a rather trivial task to me.

They already have a database installed (probably MySQL) to run this message board. Everything is already in place to do this. Someone just needs to write it.

Adding the test results would be pretty easy after that.

[Oso]

Originally posted by scotth
I build database driven web applications for a living.

This sounds like a rather trivial task to me.I often have to import historical data into an existing databse.
In my experience it's never easy.They already have a database installed (probably MySQL) to run this message board. Everything is already in place to do this. Someone just needs to write it.The JREF would first have to design an appropriate database schema to store the data. Another non trivial task if it's gonna be useful.Adding the test results would be pretty easy after that. I'll grant you that one.

Well I'm off to work. It just so happens I'll be trying to extract some north sea historical well production data for import into an existing database. Aaaghhh!

[scotth]

Originally posted by Oso
I often have to import historical data into an existing databse.
In my experience it's never easy.[b]The JREF would first have to design an appropriate database schema to store the data. Another non trivial task if it's gonna be useful.[b]I'll grant you that one.

Well I'm off to work. It just so happens I'll be trying to extract some north sea historical well production data for import into an existing database. Aaaghhh!

The schema should be trivial. This is not an extremely complicated bunch of data to store.

I was pretty much considering that I would design the schema and the web application.

As far as I know, the historical data would probably have to be manually entered for the most part as it probably isn't in any organized electronic format.

[Oso]

Originally posted by scotth
I was pretty much considering that I would design the schema and the web application.You Da Man!

[scotth]

The question is: "Is this something that JREF really wants, and would use if given?"

There may be a reasons that posting all these results is undesirable. I don't know what they would be, but I could postulate some legal issues that might be of concern.

[Denise]

I don't know, Scott. Perhaps you could Pm Hal about it. The JREF does allow people to view the data, but I know they don't allow anyone to publish people's names etc. I don't know what Randi would say.

[TechHead]

I agree with scotth, setting up a database is rather trivial. And, I agree with Denise & others, getting all the past tests into that database will be difficult, and probably a very manual process.

But, going forward, it would be near zero cost to enter in new tests as they are completed. Going backwards, perhaps a small mass of volunteers could divvy up the tests and get them entered.

There should be no legal issues with this, as spelled out in Rule #3 of The Challenge.

Diogenes,

Now the EVIDENCE is 16 out of 23.

The %'s keep increasing there bud.

:D (Whodini) :rolleyes: (Diogenes)

[scotth]

Originally posted by Whodini
Diogenes,

Now the EVIDENCE is 16 out of 23.

The %'s keep increasing there bud.

:D (Whodini) :rolleyes: (Diogenes)

What evidence?

Edited to add:
Actually, evidence of what?

[Keneke]

I had to chime in on the critical thinking 101 issue. Randi is NOT the end all and be all to scientific thought. As a matter of fact, the real world is much stricter. We here in the halls of Missile Defense scoff and deride this sort of mystical crap, but we do it quietly and cliquishly. Randi is simply the most vocal debunker, and I feel that he is using his old crowd-working techniques to support his position. This also means that he takes the flak from True Believers, and should be given a medal because of it. However, you can always fool some people some of the time. Randi's not infallible. But until you come up with results, the government and the big businesses won't blink an eye. Sure, you can fool some people with a patent on a PMM. Sure, you can make money selling crystals. But I, for one, am glad there is an intelligentsia. When a person's persuasive power far outweighs the validity of their theories, as is the case with charlatans like John Edward, we must build an (aforementioned) Ivory Tower against it. True, this makes change difficult, but I would rather have that than a world of Grover's Corners.

As long as we're all going to be assumptive, I would very much like to be assumptive on scientific theory and materialism. It's a much better system than anything else. For example, remote detection (sonar, I think) works better than dowsing, just as proven methods work better than random chance.

[Skeptical Greg]

Originally posted by Whodini
Diogenes,

Now the EVIDENCE is 16 out of 23.

The %'s keep increasing there bud.

:D (Whodini) :rolleyes: (Diogenes)


A statitician, and you're excited about a poll with 23 entries?:rolleyes:

And lets review the poll question...:
Publish results in peer-reviwed journal?

This was based on your assumption that the JREF challenge is a scientific endeavor.. JREF Makes no such claim.

I don't think many people are very excited about your poll, Whodini. But then again, it woudn't surprise me if
you have a thesaurus that lists ' 23 ' as a synonym for ' many '...

[Skeptical Greg]

Originally posted by scotth


What evidence?

Edited to add:
Actually, evidence of what?



Whodini has chosen to use the word ' evidence ', as the desription of his poll results.


It is of course, irrefutable evidence, that 23 persons chose to participate.

[scotth]

Forgot about the poll.

I get the 16 of 23 now.

That shows that it would be popular to publish in a peer-reviewed journal. Not that it is feasible or appropriate.

----
A statitician, and you're excited about a poll with 23 entries?:rolleyes:
----


heh, see? :rolleyes: (Diogenes) was an appropriate descriptor.

What I am excited about is:

60.87% (70% if you count Maybe) of JREF board participants who voted think that JREF publishing results in peer-reviwed journal is a decent idea.

(double compared to the next highest choice)


----
This was based on your assumption that the JREF challenge is a scientific endeavor.. JREF Makes no such claim.
----


I didn't make that assumption either.


----
I don't think many people are very excited about your poll, Whodini.
----


Really? Votes say otherwise, and over 1000 viewers and a decent thread rating, and more importantly, interesting discussion on its feasibility.

C'mon now Dio. Deny some more. :)


----
But then again, it woudn't surprise me if
you have a thesaurus that lists ' 23 ' as a synonym for ' many '...
----


Illucid.

[jj]

Originally posted by Interesting Ian

F*ck off a*sehole.



What the f*ck is a review board??



No you wouldn't, retard.

I suggest you figure out what a review board is, for starters. From there, you can figure out what kind of people are on it, and so on.

Then you can stop embarassing yourself.

[jj]

Originally posted by Diogenes


Thanks for taking the time..

I did not question the possibility that negative results are published..

I asked:


This is important; because failures are all that JREF has to offer.

Since Whodini hasn't given us an example of a peer-reviwed journal, that would lend some credence,
to his poll, it is difficult to address the matter of whether or not JREF testing should be reviewed or not.

There is also the matter regarding the fact, that JEREF testing is not (and doesn't claim to be) scientific research.
( Is there a bad echo in here, or is it just me?)

If you look up-thread a bit, you'll notice that I have agreed strongly with the general version of your conclusion, I'm just limiting it a little bit. No need to thank me for looking up the paper, I'm one of the authors. (chuckle) It was easy to find.

JREF testing is to find out if a given tester's claims hold up under controlled circumstances, no more or less. Unlike what uninteresting Jan thinks, it does not conclude a-priori that experiments will fail, it simply arranges experiments that are extraodinarly likely to fail IF NO PARANORMAL ABILITY EXISTS, which is a big difference.

On the other hand, given even a very slight "shade" to a decision based on a paranormal ability, it's not hard to detect such things. There is a whole field of "signal detection" out there, and in fact some folks have tried to apply it to paranormal abilities. So far, nada.

[jj]

Originally posted by rwald
If one could pass critical thinking 101, one wouldn't be in the market for a dowsing device...

(I didn't use haikus! So there!)

Critical Thinking
Not because he had the two
He just wanted to.

[Skeptical Greg]

Originally posted by Whodini

Really? Votes say otherwise, and over 1000 viewers and a decent thread rating, and more importantly, interesting discussion on its feasibility.

C'mon now Dio. Deny some more. :)

I haven't denied anything. I expressed an opinion..
1,000 views and 23 votes? That seems to say something about how interesting people find it.


----
But then again, it woudn't surprise me if
you have a thesaurus that lists ' 23 ' as a synonym for ' many '...
----
Illucid.

Oh, right beside your dictionary that has ' illucid ' in it..

[gnome]

I would say the testing procedures need not meet the standards of peer review for this reason--it does not matter if a rigorous scientific standard is applied. What matters is if the claimant performed at the level he agreed to perform at.

While I am sure that scientific methods influence JREF's side of the test negotiations, as soon as the claimant agrees to the conditions, the scientific validity is no longer in question... just success or failure.

On the other hand, I would be interested (as would anyone here I bet) in more information about tests that have been done... how many of us are here because we found JREF's challenge fascinating and love hearing any stories we can about it?

So if someone manages to put this database together, I would certainly partake of it.

Diogenes,


----
I haven't denied anything. I expressed an opinion..
1,000 views and 23 votes? That seems to say something about how interesting people find it.
----


You obviously find it interesting.

Hypothesis proved.

[rwald]

So, we went from saying a sample of 23 is statistically meaningful to saying a sample of 1 is statistically meaningful...:rolleyes:

[scotth]

Originally posted by Whodini
Diogenes,


----
I haven't denied anything. I expressed an opinion..
1,000 views and 23 votes? That seems to say something about how interesting people find it.
----


You obviously find it interesting.

Hypothesis proved.

Where exactly was the proven hypothesis stated? I can't seem to find it.

[Skeptical Greg]

Originally posted by rwald
So, we went from saying a sample of 23 is statistically meaningful to saying a sample of 1 is statistically meaningful...:rolleyes:

;)

[Skeptical Greg]

Originally posted by Whodini
Diogenes,


----
I haven't denied anything. I expressed an opinion..
1,000 views and 23 votes? That seems to say something about how interesting people find it.
----


You obviously find it interesting.

Hypothesis proved.

I believe rwald has addressed this..

[Interesting Ian]

Originally posted by jj


I suggest you figure out what a review board is, for starters. From there, you can figure out what kind of people are on it, and so on.

Then you can stop embarassing yourself.

Why should I wish to do that? I have no interest. You claimed there is not a collosal amount of evidence for certain appropriate "paranormal" phenomena. Try to understand that the word evidnce is not shorthand for "scientific evidence". I use the word evidence in the normal sense of the term.

[Interesting Ian]

Originally posted by jj


JREF testing is to find out if a given tester's claims hold up under controlled circumstances, no more or less. Unlike what uninteresting Jan thinks, it does not conclude a-priori that experiments will fail, it simply arranges experiments that are extraodinarly likely to fail IF NO PARANORMAL ABILITY EXISTS, which is a big difference.



I'm sorry, could you clarify what I think?

The experiments may well be extraordinary likely to fail even if anomalous cognitive abilities exist. Let's suppose the probability of passing the challenge is 1 in 1 billion assuming no anomalous abilities or phenomena. Now if such paranormal phenomena exist the testee might be able to actually double his chances of winning (which any fair minded person would conclude is quite extraordinary). Nevertheless he would still only have 1 chance in 500 million in passing the challenge.

[scotth]

Originally posted by Interesting Ian


The experiments may well be extraordinary likely to fail even if anomalous cognitive abilities exist. Let's suppose the probability of passing the challenge is 1 in 1 billion assuming no anomalous abilities or phenomena. Now if such paranormal phenomena exist the testee might be able to actually double his chances of winning (which any fair minded person would conclude is quite extraordinary). Nevertheless he would still only have 1 chance in 500 million in passing the challenge.

Before I conclude on my own, I will put the question to you directly. Are you suggesting that the tests have been or are designed this way or merely pointing out that it is possible?

Edited to add:
I can't imagine why you would bother bringing it up if it wasn't an underhanded accusation.

[Skeptical Greg]

Originally posted by Interesting Ian


I'm sorry, could you clarify what I think?

The experiments may well be extraordinary likely to fail even if anomalous cognitive abilities exist. Let's suppose the probability of passing the challenge is 1 in 1 billion assuming no anomalous abilities or phenomena. Now if such paranormal phenomena exist the testee might be able to actually double his chances of winning (which any fair minded person would conclude is quite extraordinary). Nevertheless he would still only have 1 chance in 500 million in passing the challenge.


And you expect a prize of $1,000,000 to be given up for someone who can exhibit some phenomenon,
once in 500,0000 tries.

I could live with a job like that..

[Interesting Ian]

Originally posted by Diogenes



And you expect a prize of $1,000,000 to be given up for someone who can exhibit some phenomenon,
once in 500,000 tries.

I could live with a job like that..

500,000,000 not 500,000

[Stimpson J. Cat]

Ian,

The experiments may well be extraordinary likely to fail even if anomalous cognitive abilities exist. Let's suppose the probability of passing the challenge is 1 in 1 billion assuming no anomalous abilities or phenomena. Now if such paranormal phenomena exist the testee might be able to actually double his chances of winning (which any fair minded person would conclude is quite extraordinary). Nevertheless he would still only have 1 chance in 500 million in passing the challenge.

Once again, that is not the purpose of the challenge. Somebody who possesses such a weak paranormal ability is not going to be applying for the challenge. indeed, such a person would have no reason to believe he possesses such an ability at all.

I am really surprised you are taking this attitude towards the challenge. Your posts seem to clearly indicate that you agree that the various so-called psychics, faith-healers, remote-viewers, etc... who are out there claiming to have demonstrable paranormal powers, and charging people money for their services, are frauds. These are the types of people that the challenge is designed to reveal.

I would think that you would support such an endeavor. After all, these scumbags are not only defrauding people, they are also undermining the credibility of any honest investigation into paranormal phenomena.

The bottom line is that whether or not anomalous paranormal phenomena of some form exist, the people out there who are claiming to possess paranormal powers are, without exception, either liars, or seriously deluded. The purpose of the JREF challenge is to demonstrate this fact to the public. The liars refuse the challenge, because they know they will fail, and the delude go for the challenge, and then can't figure out why they failed.

If your only beef with the JREF challenge is that people try to claim it disproves the possibility of any sort of paranormal phenomena, then your beef is with those people, not the JREF. The JREF has made no such claim.

I personally have never seen anybody make such a claim. Perhaps you misinterpreted what they were saying? Maybe they were not referring to some sort of vague generic "paranormal phenomena", but instead to the specific types of paranormal claims that frauds and loonies are constantly making? If not, then they are wrong. But that isn't James Randi's fault.


Dr. Stupid

[Interesting Ian]

Originally posted by scotth


Before I conclude on my own, I will put the question to you directly. Are you suggesting that the tests have been or are designed this way or merely pointing out that it is possible?

Edited to add:
I can't imagine why you would bother bringing it up if it wasn't an underhanded accusation.

Not sure what you mean. The probability is a guesstimate. I guessed 1 in 1000 to pass the preliminary test and 1 in 1,000,000 to pass the proper challenge.

I know that skeptics claim the odds are much greater. As you know a couple of days ago I was told that the probability of passing the challenge was 1 in 50. I flat out didn't believe that and then someone said that 1 in 50 only applied to one particular challenge. then somebody else said the 1 in 50 was the probability of passing the preliminary testing.

So I've been misinformed time after time after time. Now in the past couple of days I read somewhere that no-one had actually ever passed the preliminary test. Therefore i strongly suspect that this 1 in 50 chance of passing the preliminary test is also a falsehood. I came up with some figures which seemed to me to be realistic.

[scotth]

Originally posted by Interesting Ian


Not sure what you mean. The probability is a guesstimate. I guessed 1 in 1000 to pass the preliminary test and 1 in 1,000,000 to pass the proper challenge.

I know that skeptics claim the odds are much greater. As you know a couple of days ago I was told that the probability of passing the challenge was 1 in 50. I flat out didn't believe that and then someone said that 1 in 50 only applied to one particular challenge. then somebody else said the 1 in 50 was the probability of passing the preliminary testing.

So I've been misinformed time after time after time. Now in the past couple of days I read somewhere that no-one had actually ever passed the preliminary test. Therefore i strongly suspect that this 1 in 50 chance of passing the preliminary test is also a falsehood. I came up with some figures which seemed to me to be realistic.

The suggestion was that the tests were or are designed such that even a person who has the ability claimed would STILL only have a 1:500,000,000 chance of passing.

[BillHoyt]

Originally posted by Interesting Ian
Try to understand that the word evidnce is not shorthand for "scientific evidence". I use the word evidence in the normal sense of the term.

Actually, Ian, it is shorthand for "scientific evidence." This is the REF forum, mostly populated by people trying to understand claims in the context of the rules / norms / processes of science. When JREFers say "evidence" they do mean "scientific evidence." Why? Because science has epistemological privilege.

Now I know we've been down that road before, and I know you usually want to follow such statements with assertions about my lack of insight or intelligence or simply my lack of making sense, but that is how it is.

"Peer review:" this is one of the recent layers of science that became necessary after centuries of monographs and the advent of journals. Peer review is the beginning of the process of intersubjective verification.

"Good scientific evidence" starts with a peer-reviewed paper. The research peers, usually blinded to the author's name, pore over the paper, and analyze the methods and the conclusions. They check that the history is accurately reported, that the methods really can be used to get good results for the study at hand. They check that the data show what is claimed. They check that the discussion and conclusions do not stray beyond what the data show. They "judge" the work before it is submitted for publication.

After publication, though, is when the real fun begins...

Cheers,

[DrBenway]

Originally posted by gnome
I would say the testing procedures need not meet the standards of peer review for this reason--it does not matter if a rigorous scientific standard is applied. What matters is if the claimant performed at the level he agreed to perform at.
I don't think you do Randi any favors when you argue that the JREF challenge is not scientific, or that the standards of proof for the JREF challenge are less that is required in ordinary scientific studies. In fact, I think you make the JREF a hypocrite when you use this argument.

The JREF insists that the standards of evidence used in science be applied to claims of the paranormal. The JREF faults the "woo-woos" for neglecting to use good scientific methods in establishing the truth of their claims. Then, according to your argument, the JREF blows off using the scientific method in assessing challengers for the $1,000,000 prize.

This is absurd.

Scientists (and statisticians, when needed) are involved in the process of evaluating applicants for the prize. The same standards used to establish *physical* effects are applied to hypothesized *paranormal* effects.

You don't need a white lab coat to test a hypothesis scientifically. Anyone who understands the principles of the scientific method and how one establishes statistical significance can do it.

A child can perform a test of a coin, to rule-out the hypothesis that it's a weighted coin which will come up heads more often than tails, by flipping the coin a certain number of times and recording whether it fell "heads" or "tails" each time. Using a bit of math to assess the probability of the result obtained, he can judge whether his coin is weighted or not.

This is science. The "peer review" part comes in, when the child describes exactly what he did, in such a way that another person could replicate his work. The "peers" are those able to understand and replicate the work.

A more specific sort of "peer review" is editorial peer review. This happens when a paper is sent to a journal for publication. In the case of the coin toss experiment, editorial peer review would be silly. Yet the coin toss experiment, if done right, is most certainly an example of scientific work.

It's that simple. It's the way do do things. No one is allowed to fudge their way around the scientific method, especially not the JREF.

The preliminary screens set the probability of the results being obtained by chance alone at a higher level (e.g., 1/50). If a claimant passes the screen, the subsequent test is designed with more attention to carefully controlled conditions, and sets the standard of results being obtained by chance alone at a lower level (e.g., 1/10,000, as is typical for scientific work in the physical sciences).

I haven't looked at the recent evaluation of homeopathy. I'm guessing the "by chance alone" test was set at the more liberal 1/20, as is typical of studies in the medical literature.

Why do doctors get to publish work using the 1/20 standard, while physicists must use the 1/10,000 standard? Well, that's a longer discussion. But you'll notice that physicists don't often appear in the press saying things like, "What we told you two years ago about such-and-such being good for you is actually a bunch of bull."

Like the coin toss study, a study of dowsing that failed to show evidence of paranormal phenomenon would lack merit for publication. Why? Too boring. But not getting published doesn't mean the study wasn't "scientific."

There's science; there's logic; and there's art. What else is there?

[gnome]

All granted, but my point is... supposing JREF proposes a scientific standard that could pass peer-review. The claimant counter-proposes something less rigorous, but JREF believes the claimant can't even perform that, and agrees. The claimant fails, as predicted.

Of what benefit is peer review in such a situation?

Edited to add: I see from your post you are not pushing peer-review. But what I am presenting is a test that is not scientifically rigorous but still useful to the JREF.

[jj]

Originally posted by Interesting Ian


. Let's suppose the probability of passing the challenge is 1 in 1 billion assuming no anomalous abilities or phenomena. Now if such paranormal phenomena exist the testee might be able to actually double his chances of winning (which any fair minded person would conclude is quite extraordinary). Nevertheless he would still only have 1 chance in 500 million in passing the challenge.

Um, Ian, really. Look into "signal detection theory".

Let's say that the chance to succeed randomly is 10^-9, which is approximately 2^-30.

That's the chance of getting 30 coin flips of heads with a fair coin.

Now, let's say that the coin has a .501 shot of landing heads. How many coin flips are required?

[Interesting Ian]

Originally posted by jj


Um, Ian, really. Look into "signal detection theory".

Let's say that the chance to succeed randomly is 10^-9, which is approximately 2^-30.

That's the chance of getting 30 coin flips of heads with a fair coin.

Now, let's say that the coin has a .501 shot of landing heads. How many coin flips are required?

How many coin flips are required for what?

Look it must be possible to conclude some sort of anomalous process at work, even if the statistical departure from chance results is comparatively small. Just make sure there is enough trials and be sure to eliminate any possible artefacts skewing the results.

[BillHoyt]

Let's suppose the probability of passing the challenge is 1 in 1 billion assuming no anomalous abilities or phenomena. Now if such paranormal phenomena exist the testee might be able to actually double his chances of winning (which any fair minded person would conclude is quite extraordinary). Nevertheless he would still only have 1 chance in 500 million in passing the challenge.

Ian,

You are missing the statistical point of the matter. If the probability is 1 in 1 billion, then the test must be designed to distinguish the results from chance. The expectation for 1 billion trials is 1 positive result. That means you must have many more than 1 billion trials so that the number of positive results can be high enough to clearly distinguish from chance.

Having done this, though, that is, having run through x billions of trials, the claimant's chances are no longer 1 in 500 million.

In analyzing this, you are confounding your null hypothesis (psi exists at twice chance levels) with its antithesis (psi doesn't exist).

Cheers,

[Interesting Ian]

Originally posted by BillHoyt
Try to understand that the word evidnce is not shorthand for "scientific evidence". I use the word evidence in the normal sense of the term.
--------------------------------------------------------------------------------



Actually, Ian, it is shorthand for "scientific evidence." This is the REF forum, mostly populated by people trying to understand claims in the context of the rules / norms / processes of science.



If the word "evidence" henceforth is going to be used shorthand for "scientific evidence", what other word are we going to use for the current definition of the word "evidence"?




When JREFers say "evidence" they do mean "scientific evidence." Why? Because science has epistemological privilege.



Science is concerned with mapping out the patterns of the perceptually experienced. It does not give us knowledge about the ethical dimension, how we ought to lead our lives, what the world is, what we are etc. Thus saying it is epistemologically privileged is seriously misleading.



Now I know we've been down that road before, and I know you usually want to follow such statements with assertions about my lack of insight or intelligence or simply my lack of making sense, but that is how it is.



Your unreasoned dogmatic approach has never impressed me and continues not to impress me.

[Interesting Ian]

Originally posted by BillHoyt


Ian,

You are missing the statistical point of the matter. If the probability is 1 in 1 billion, then the test must be designed to distinguish the results from chance. The expectation for 1 billion trials is 1 positive result. That means you must have many more than 1 billion trials so that the number of positive results can be high enough to clearly distinguish from chance.

Having done this, though, that is, having run through x billions of trials, the claimant's chances are no longer 1 in 500 million.

In analyzing this, you are confounding your null hypothesis (psi exists at twice chance levels) with its antithesis (psi doesn't exist).

Cheers,

No this is erroneous reasoning. There might be a very large number of trials required in order to determine whether the results are statistically significant. But it wouldn't be billions!

[BillHoyt]

Originally posted by Interesting Ian


No this is erroneous reasoning. There might be a very large number of trials required in order to determine whether the results are statistically significant. But it wouldn't be billions!

Ian,

If the p(x) = 1/1Billion, the expectation function for 1 billion trials is 1. Your hypothetical claimant can double that. That is, he can achieve 2 in 1 billion. But this is a Poisson distribution. Under the null hypothesis, getting 2 hits by chance alone is not significant at the .05 level, let alone the .01 level.

Cheers,

[Skeptical Greg]

Originally posted by Interesting Ian


No this is erroneous reasoning. There might be a very large number of trials required in order to determine whether the results are statistically significant. But it wouldn't be billions!

Nothing incorrect about your statement, but the example given was ' 1 in a billion '..

If the probablility is 1 in a billion, then there would be nothing anomalous about 999,999,999 trials without a hit.

But back to the JREF challenge.

It is not about generating a statistical fluke, it is about performing a skill that one claims they posses.

[BillHoyt]

Originally posted by Interesting Ian
If the word "evidence" henceforth is going to be used shorthand for "scientific evidence", what other word are we going to use for the current definition of the word "evidence"?
In this context, how about "useless"?
Science is concerned with mapping out the patterns of the perceptually experienced. It does not give us knowledge about the ethical dimension, how we ought to lead our lives, what the world is, what we are etc. Thus saying it is epistemologically privileged is seriously misleading.
I'm afraid this is misleading. Are we talking about ethics or values? No. We are talking about the physical world here. A claimant walks into JREF and says "I can do x." That is physical, no matter what your presumptions about the source of the ability to do x. It has now entered scientific turf, and science's epistemological privilege most assuredly holds.
Your unreasoned dogmatic approach has never impressed me and continues not to impress me.
And these ad homs, with which your posts are often riddled, fail to impress me. Let's stick with facts, evidence and logic here, eh?

Cheers,

[Interesting Ian]

Originally posted by Diogenes


Nothing incorrect about your statement, but the example given was ' 1 in a billion '..

If the probablility is 1 in a billion, then there would be nothing anomalous about 999,999,999 trials without a hit.



What do you mean by a "hit"? What constitutes a trial? I think maybe it might be useful to talk about a specific challenge. Let's suppose we wish to test for micro-psychokinesis. Let's test for it by attempting to affect the output of a RNG to obtain the greatest number of ones possible. Let's say we manage to acheive an output on average of 101 one's for every 100 zero's. Now what would a "hit" and a "trial" mean in the context of a particular challenge like this? Would each "one" outputted constitute a hit? Would a trial consist in attempting to alter the output of 100 digits? Enlighten me.

[BillHoyt]

Originally posted by Interesting Ian


What do you mean by a "hit"? What constitutes a trial? I think maybe it might be useful to talk about a specific challenge. Let's suppose we wish to test for micro-psychokinesis. Let's test for it by attempting to affect the output of a RNG to obtain the greatest number of ones possible. Let's say we manage to acheive an output on average of 101 one's for every 100 zero's. Now what would a "hit" and a "trial" mean in the context of a particular challenge like this? Would each "one" outputted constitute a hit? Would a trial consist in attempting to alter the output of 100 digits? Enlighten me.

In the case you described, we have minimally 201 trials, with 101 hits and 100 misses. That is not significant. But you said "average." So, how big is the actual N you are positing?

Cheers,

[Interesting Ian]

Originally posted by BillHoyt
What do you mean by a "hit"? What constitutes a trial? I think maybe it might be useful to talk about a specific challenge. Let's suppose we wish to test for micro-psychokinesis. Let's test for it by attempting to affect the output of a RNG to obtain the greatest number of ones possible. Let's say we manage to acheive an output on average of 101 one's for every 100 zero's. Now what would a "hit" and a "trial" mean in the context of a particular challenge like this? Would each "one" outputted constitute a hit? Would a trial consist in attempting to alter the output of 100 digits? Enlighten me.
--------------------------------------------------------------------------------



In the case you described, we have minimally 201 trials, with 101 hits and 100 misses. That is not significant. But you said "average." So, how big is the actual N you are positing?



What is N? The actual number of digits outputted? Well it would have to be large enough so that the overall probability of acheiving this average was sufficiently low that it would be deemed to have passed the challenge. So for 1 billionth of a chance it would be . . . er . . whatever it might be. Let's say X.

But of course for a given amount of trials (such as X), the testee might get statistically significant results, let's say over X trails he manages to get slightly less than 101 ones for 100 zero's, but nevertheless he gets a sufficent excess of ones over zero's, so that the probability of getting such a result is only 1 in 500,000,000 if no "anomalous" processes were at play. Nevertheless he would not win the challenge. This might be deemed to be a bit unfair.

[Interesting Ian]

Actually I've just thought of a way in which Randi could cheat. Unless the number of trials is specified beforehand, then at any particular point where the results are statistically significant, Randi could keep testing the testee in the hope that the overall result ceases to be statistically significant. Of course he runs the risk of the overall result becoming even more statistically signioficant. However from his perspective that doesn't matter as the million is lost whether the challenge has just to say been met, or whether it was comprehensively defeated.

[BillHoyt]

Originally posted by Interesting Ian
What is N? The actual number of digits outputted? Well it would have to be large enough so that the overall probability of acheiving this average was sufficiently low that it would be deemed to have passed the challenge. So for 1 billionth of a chance it would be . . . er . . whatever it might be. Let's say X.

N is the number in the sample size. If your units are the binary digits, it would be the total number of RNG bits tested. You don't necessarily need a billion such trials to get a significant result.

But of course for a given amount of trials (such as X), the testee might get statistically significant results, let's say over X trails he manages to get slightly less than 101 ones for 100 zero's, but nevertheless he gets a sufficent excess of ones over zero's, so that the probability of getting such a result is only 1 in 500,000,000 if no "anomalous" processes were at play. Nevertheless he would not win the challenge. This might be deemed to be a bit unfair.
All hypothesis testing is done to a significance level. Typically, researchers look for at least .05 chance. More stringently, .01 chance. That is, the results must be better than chance would give you in 1 in 20 trials (.05) or 1 in 100 trials (.01).

Now you are claiming "micropsychokinesis," meaning the effect is very small. So now the trial cannot be simply a single bit, can it? So the RNG must be run at least 1 billion times (according to your claimed psi effect.) We already have a technical problem: what is the RNG's pseudorandom sequence repeat length? Can we actually use it to generate the 1billion+ bits you suggest?

Cheers,

[BillHoyt]

Originally posted by Interesting Ian
Actually I've just thought of a way in which Randi could cheat. Unless the number of trials is specified beforehand, then at any particular point where the results are statistically significant, Randi could keep testing the testee in the hope that the overall result ceases to be statistically significant. Of course he runs the risk of the overall result becoming even more statistically signioficant. However from his perspective that doesn't matter as the million is lost whether the challenge has just to say been met, or whether it was comprehensively defeated.
Ian,

The number of trials must be set a priori because this affects the success criteria. The test conditions and the success criteria are all agreed to before hand.

This is also basic science. You can't determine the success criteria a posteriori, because you muck up the statistics.

Cheers,

[Skeptical Greg]

Originally posted by Interesting Ian


What constitutes a trial? I think maybe it might be useful to talk about a specific challenge. Let's suppose we wish to test for micro-psychokinesis.

Snip...

Enlighten me.

I'll let BH do the enlightening...

What constitutes a trial, is what JREF agrees to..

I don't think " micro-psychokinesis " is going to make it to the prelims.

If someone can 'really' do ' para ', they do not need JREF's million to get rich.

To paraphrase my earlier statement:

It's not about giving someone the opportunity to exhibit a statistical fluke, it is about having them
demonstrate a skill, that one claims they posses.

[Interesting Ian]

Originally posted by BillHoyt
[B]

N is the number in the sample size. If your units are the binary digits, it would be the total number of RNG bits tested. You don't necessarily need a billion such trials to get a significant result.


All hypothesis testing is done to a significance level. Typically, researchers look for at least .05 chance. More stringently, .01 chance. That is, the results must be better than chance would give you in 1 in 20 trials (.05) or 1 in 100 trials (.01).

Now you are claiming "micropsychokinesis," meaning the effect is very small. So now the trial cannot be simply a single bit, can it? So the RNG must be run at least 1 billion times (according to your claimed psi effect.) We already have a technical problem: what is the RNG's pseudorandom sequence repeat length? Can we actually use it to generate the 1billion+ bits you suggest?



Just a quick comment because haven't got time.

Obviously no-one can affect pseudo-random numbers! :eek: Do RNG's only generate pseudo-random sequences???

[DrBenway]

Originally posted by Interesting Ian

Obviously no-one can affect pseudo-random numbers! :eek: Do RNG's only generate pseudo-random sequences???

Correct.

I hope you are now having yourself a good laugh.

Originally posted by Interesting Ian


Just a quick comment because haven't got time.

Obviously no-one can affect pseudo-random numbers! :eek: Do RNG's only generate pseudo-random sequences???

You know, he's got a point there.

[BillHoyt]

Originally posted by DrBenway


Correct.

I hope you are now having yourself a good laugh.

I don't think he understands, DrBenway.

Ian,

RNGs are all pseudo-random number generators. There are three main classes, but all have repeat lengths. That is, they will appear to generate numbers randomly for X runs, but they will then repeat exactly the same sequence for the next X+1 to 2X runs.

Given this, do you want to restate your assertion about the ability to affect RNGs?

Cheers,

[BillHoyt]

I guess it's clear a number of us are having a good laugh anyway... :D

Cheers,

Originally posted by BillHoyt


Given this, do you want to restate your assertion about the ability to affect RNGs?

Cheers,

I think even Ian realizes that this would be a silly claim. He clearly means some hypothetical "real" random number generator.

But as far as I know this could only be done non-computationally, by tracking of some naturally random phenomenon like Brownian motion or quantum fluctuations, meaning that the successful psychic would actually be controlling these.

Of course (pre-empting Franko here) Franko would claim that Ian might as well claim he can move mountains, because there is no such thing as a random sequence.

I have a classic keyboard called a MemoryMoog that used one of the early digital white noise generation algorithms. You can actually hear it repeat every few seconds.

[Stimpson J. Cat]

Ian,

Just a quick comment because haven't got time.

Obviously no-one can affect pseudo-random numbers! Do RNG's only generate pseudo-random sequences???

They can, but there wouldn't be much point, since pseudo-random number generators are deterministic.

Most "paranormal" studies use circuits which produce quantum noise. They then digitize the output into a bit-stream.

Of course, as is the case with any physical system, there will be a certain amount of bias in the system (1 being more or less likely than 0). This bias may vary as a function of time, as well.

To compensate for this bias, they do a bitwise exclusive-or (XOR) operation on the bit-stream. This does not destroy the bias, but reduces it greatly. Because there is still alway going to be a bias, they do control studies to make sure that the bias is beneath a set threshold. This allows them to get reliable statistics for data sets up to a certain size.

Unfortunately, such tests never show anything. People then demand that meta-analysis be done. They group several studies together, and low and behold, they get a statistically significant result, not realizing that it is probably just due to the bias, which was not controlled for at a high enough level for the new, longer data set.

Of course, all of this ignores the fact that looking for statistical anomalies in random sequences is utterly useless as a test for the paranormal, since a positive result only indicates that some non-random influence was present, and it is not possible, even in principle, to control for every possible influence. In effect, the only hypothesis that can be falsified by such a test is that the bit-stream is a purely random process, with no outside influences of any kind. And that hypothesis is trivially false to begin with.

Dr. Stupid

[Interesting Ian]

Originally posted by DrBenway


Correct.

I hope you are now having yourself a good laugh.

Well I'm sure I've read research where people seemingly have been able to influence RNG's by an apparently anomolous cognitive process. Obviously it couldn't have been an anomalous cognitive ability though if it were only pseudo-random generated sequence. Micro-psychokinesis cannot acheive the impossible! Hmmm . . .I'm going to have to look into this.

Originally posted by Interesting Ian


Well I'm sure I've read research where people seemingly have been able to influence RNG's by an apparently anomolous cognitive process. Obviously it couldn't have been an anomalous cognitive ability though if it were only pseudo-random generated sequence. Micro-psychokinesis cannot acheive the impossible! Hmmm . . .I'm going to have to look into this.

Re-read Stimpy's post.

[Interesting Ian]

Originally posted by Stimpson J. Cat
[B]Ian,



They can, but there wouldn't be much point, since pseudo-random number generators are deterministic.

Most "paranormal" studies use circuits which produce quantum noise. They then digitize the output into a bit-stream.



So there are RNG which produce genuine random numbers?? Thought the "skeptics" might have been lying again.



Of course, as is the case with any physical system, there will be a certain amount of bias in the system (1 being more or less likely than 0). This bias may vary as a function of time, as well.

To compensate for this bias, they do a bitwise exclusive-or (XOR) operation on the bit-stream. This does not destroy the bias, but reduces it greatly. Because there is still alway going to be a bias, they do control studies to make sure that the bias is beneath a set threshold. This allows them to get reliable statistics for data sets up to a certain size.

Unfortunately, such tests never show anything.



So you claim, but then "skeptics" always lie.





People then demand that meta-analysis be done. They group several studies together, and low and behold, they get a statistically significant result, not realizing that it is probably just due to the bias, which was not controlled for at a high enough level for the new, longer data set.

Of course, all of this ignores the fact that looking for statistical anomalies in random sequences is utterly useless as a test for the paranormal, since a positive result only indicates that some non-random influence was present, and it is not possible, even in principle, to control for every possible influence. In effect, the only hypothesis that can be falsified by such a test is that the bit-stream is a purely random process, with no outside influences of any kind. And that hypothesis is trivially false to begin with.



Ok, I don't know much about parapsychological research but it would be extremely easy to eliminate any putative bias in the RNG. Simply got the subject first of all to try and influence the RNG to produce more one's than zero's, and then get him to try and influence the very same RNG in the very same setting to produce more zero's than ones! If successful on both occasions this at least eliminates any bias in the RNG.

{SIGHS} :rolleyes:

[Interesting Ian]

Originally posted by sundog


II
Well I'm sure I've read research where people seemingly have been able to influence RNG's by an apparently anomolous cognitive process. Obviously it couldn't have been an anomalous cognitive ability though if it were only pseudo-random generated sequence. Micro-psychokinesis cannot acheive the impossible! Hmmm . . .I'm going to have to look into this.

SD
Re-read Stimpy's post. [/B]

I f*cking didn't read his post till after I had sent mine!! :mad: :mad: :mad:

[BillHoyt]

Originally posted by Interesting Ian


Well I'm sure I've read research where people seemingly have been able to influence RNG's by an apparently anomolous cognitive process.
That's right. That is the claim of this research.
Obviously it couldn't have been an anomalous cognitive ability though if it were only pseudo-random generated sequence. Micro-psychokinesis cannot acheive the impossible! Hmmm . . .I'm going to have to look into this.
Ian, an RNG is a pseudo-random generator. That doesn't mean it isn't really an RNG. It is a real machine. The algorithm is pseudo; it mimics a random number generator.

Cheers,

[Interesting Ian]

Originally posted by BillHoyt
[B]
That's right. That is the claim of this research.

Ian, an RNG is a pseudo-random generator. That doesn't mean it isn't really an RNG. It is a real machine. The algorithm is pseudo;



There is no such thing as a pseudo algorithm.




it mimics a random number generator.



Gosh! Thanks for explaining that! I might not have realised!

:rolleyes:

Originally posted by BillHoyt

That's right. That is the claim of this research.

Ian, an RNG is a pseudo-random generator. That doesn't mean it isn't really an RNG.

Cheers,

Huh? Explain, please?

[Stimpson J. Cat]

Ian,

Unfortunately, such tests never show anything.
--------------------------------------------------------------------------------

So you claim, but then "skeptics" always lie.

Well, considering that the parapsychologist doing these tests say so too, I guess the fact that we skeptics are all a bunch of liars doesn't really matter. :rolleyes:

Ok, I don't know much about parapsychological research but it would be extremely easy to eliminate any putative bias in the RNG. Simply got the subject first of all to try and influence the RNG to produce more one's than zero's, and then get him to try and influence the very same RNG in the very same setting to produce more zero's than ones! If successful on both occasions this at least eliminates any bias in the RNG.

It isn't that simple. When the subject is trying to get more ones, and does, it is not sufficient to just say he got more ones. You have to assess the statistical significance. And in order to do that, you have to know something about the statistical distribution. The statistics are all done under the assumption of no bias. If there is a bias, then this assumption is not valid, and the statistical confidence cannot be determined. The situation is even more complicated, because (a) Even if there is no systematic bias towards 0s or 1s, a bias in the variability of the distribution can result in false positives. And (b) the XOR process that is used to reduce the systematic bias in the stream also has the effect of amplifying any correlations that are present in the data, but the statistics of course have to assume that there are no correlations between bits.

These issues are not specific to parapsychology, Ian. These are factors I have to deal with in my own research every day. Only in my case, it is statistical testing for evidence of nonlinear dynamical behavior in noisy biological data.

Dr. Stupid

[BillHoyt]

Originally posted by Interesting Ian


There is no such thing as a pseudo algorithm.



Gosh! Thanks for explaining that! I might not have realised!

:rolleyes:

Ian,

Where on earth is the disconnect? All RNGs mimic random number generators. They use algorithms call pseudorandom number generators. These algorithms appear to yield random number sequences, but, after a certain sequence length, the algorithm repeats the entire sequence. The sequence is deterministic, but appears, throughout the sequence length, to be random.


Cheers,

[BillHoyt]

Originally posted by sundog


Huh? Explain, please?

Just trying to figure out the disconnect with "pseudorandom".

Cheers,

[Interesting Ian]

Originally posted by BillHoyt
[B]

Ian,

Where on earth is the disconnect?



I wish you'd speak in English!




All RNGs mimic random number generators.



Since RNG stands for "random number generator", that translates as "All random number generators mimic random number generators". Very profound :rolleyes:

If you're trying to say that all RNG's merely generate pseudo random numbers then you're wrong.



They use algorithms call pseudorandom number generators. These algorithms appear to yield random number sequences, but, after a certain sequence length, the algorithm repeats the entire sequence. The sequence is deterministic, but appears, throughout the sequence length, to be random.



Yes yes yes, I understand all this! God knows what the purpose of your post was.

Hey, I've just had a thought. Everyone was ecstatic about your return. I was the only exception amongst all those people (well maybe apart from UCE as well). So given that I was the only one who didn't want you to return, and everybody else did, why am I the poor sod who has to converse with you all the time!? Come to think of it that's maybe partially explain my aversion to you. I bet if you droned on about pseudo random number generators to Denise she wouldn't have been so keen on you coming back! LOL

[scotth]

Concerning random number generators.

There are plug in cards and modules available for PC's that are (as best as anyone, including the designer) true random number generators.

Originally posted by scotth
Concerning random number generators.

There are plug in cards and modules available for PC's that are (as best as anyone, including the designer) true random number generators.

Really? How do they work?

[scotth]

Originally posted by sundog


Really? How do they work?

In general, they work by sampling some random phsyical process. Often it is some type of electronic noise.

All of them that I have seen also have a top limit on the number of random bits per second they can generate.

[Interesting Ian]

Originally posted by Stimpson J. Cat

Hey Stimpy! I must be psychic. I knew you would respond along the lines you did! :eek:



Stimpy
--------------------------------------------------------------------------------
Unfortunately, such tests never show anything.
--------------------------------------------------------------------------------
II
So you claim, but then "skeptics" always lie.
--------------------------------------------------------------------------------


Stimpy
Well, considering that the parapsychologist doing these tests say so too, I guess the fact that we skeptics are all a bunch of liars doesn't really matter.



The parapsychologist?? Could it really be? You've managed to locate a parapsychologist who hasn't had positive results for micro-psychokinesis?? :eek: Oh well, this clearly discredits the whole field of micro-psychokinesis then! :rolleyes:

Who's this parapsychologist then? Hang on, let me guess. Susan Blackmore??


II
--------------------------------------------------------------------------------
Ok, I don't know much about parapsychological research but it would be extremely easy to eliminate any putative bias in the RNG. Simply got the subject first of all to try and influence the RNG to produce more one's than zero's, and then get him to try and influence the very same RNG in the very same setting to produce more zero's than ones! If successful on both occasions this at least eliminates any bias in the RNG.
--------------------------------------------------------------------------------



It isn't that simple.



Of course not. it wouldn't be would it!?


When the subject is trying to get more ones, and does, it is not sufficient to just say he got more ones. You have to assess the statistical significance.


yea, the fact that whenever he tries to influence the RNG to produce more 1's and it complies, and the fact that whenever he tries to influence the RNG to produce more 0's and it complies is wholly irrelevant! :rolleyes:



And in order to do that, you have to know something about the statistical distribution.


I presume the distribution would be random.



The statistics are all done under the assumption of no bias. If there is a bias, then this assumption is not valid, and the statistical confidence cannot be determined.



Why should there be a bias? And if there is why can't it be eliminated?



The situation is even more complicated, because (a) Even if there is no systematic bias towards 0s or 1s, a bias in the variability of the distribution can result in false positives.



But heaven forbid never false negatives! At lease not if we're trying to investigate anomalous abilities.



And (b) the XOR process that is used to reduce the systematic bias in the stream also has the effect of amplifying any correlations that are present in the data, but the statistics of course have to assume that there are no correlations between bits.



Oh well, that's the mysterious XOR process for you! LOL

[scotth]

I have been in contact with some of the illuminati in Florida.

Creating an online database of the $1M challenges will be getting discussed soon.

There is some solid initial support. I'll keep everyone posted.

Edited to add:

I am told that this idea will be discussed during the "radio show" tomorrow night.

[RichardR]

Originally posted by Interesting Ian
Hey, I've just had a thought. Everyone was ecstatic about your return. I was the only exception amongst all those people (well maybe apart from UCE as well). So given that I was the only one who didn't want you to return, and everybody else did, why am I the poor sod who has to converse with you all the time!? Come to think of it that's maybe partially explain my aversion to you. I bet if you droned on about pseudo random number generators to Denise she wouldn't have been so keen on you coming back! LOL Thanks Ian, that gave me the best laugh I've had all day. :D

[RichardR]

Originally posted by scotth
I have been in contact with some of the illuminati in Florida.

Creating an online database of the $1M challenges will be getting discussed soon.

There is some solid initial support. I'll keep everyone posted. The Illuminati are doing it? Oh no, don't tell bigfig.

[Interesting Ian]

Originally posted by RichardR
Originally posted by Interesting Ian
Hey, I've just had a thought. Everyone was ecstatic about your return. I was the only exception amongst all those people (well maybe apart from UCE as well). So given that I was the only one who didn't want you to return, and everybody else did, why am I the poor sod who has to converse with you all the time!? Come to think of it that's maybe partially explain my aversion to you. I bet if you droned on about pseudo random number generators to Denise she wouldn't have been so keen on you coming back! LOL
--------------------------------------------------------------------------------

Thanks Ian, that gave me the best laugh I've had all day.


I aim to please :D

[FutileJester]

To clarify a bit:

No real-world RNG's are perfect. Pseudo-random generators that are implemented in software are the worst, since they are deterministic. Various types of hardware RNG's also exist, but all have some degree of bias. This is complicated by the fact that the bias may not be constant (for instance, a RNG based on the thermal noise in a resistor will be sensitive to temperature).

I believe this could be controlled for. For instance, you could use 'control' RNG's running at the same time under similar conditions. Say you use 5 RNG's, all of which are recorded for every trial. In each trial, one of the 5 RNG's is chosen at random to be influenced by the subject. Is this statistically valid, in the sense that a confidence could be assigned a priori? Stats experts, help me out here...

[BillHoyt]

Originally posted by FutileJester
To clarify a bit:

No real-world RNG's are perfect. Pseudo-random generators that are implemented in software are the worst, since they are deterministic.

Actually, Futile, the claims for RNGs are great, and the performance is decidedly worse than the best algorithms. If you look for the typical solutions to the problems of RNGs, you will find they are often coupled with PRNGs.

What is the real state of RNGs versus the manufacturer claims? We can take as just one touchstone the fact that the U.S. NIST (http://csrc.nist.gov/CryptoToolkit/tkrng.html) , who approve random number generation algorithms and equipment for cryptographic (and other) use have yet to approve an RNG. This is significant. Not one RNG.

Cheers,

[Stimpson J. Cat]

Ian,

Well, considering that the parapsychologist doing these tests say so too, I guess the fact that we skeptics are all a bunch of liars doesn't really matter.
--------------------------------------------------------------------------------

The parapsychologist?? Could it really be? You've managed to locate a parapsychologist who hasn't had positive results for micro-psychokinesis?? Oh well, this clearly discredits the whole field of micro-psychokinesis then!

I never claimed that it discredits anything. I was merely pointing out that you don't have to take "skeptics" word for the fact that these tests don't show statistically significant results. There own studies admit this fact.

Who's this parapsychologist then? Hang on, let me guess. Susan Blackmore??

no, I am talking about the people actually doing this research with these RNG's, who admit that the only way they can get "significant" results, is with meta-analysis.

It isn't that simple.
--------------------------------------------------------------------------------

Of course not. it wouldn't be would it!?

What does that mean? Do you have any experience with statistical data analysis?

When the subject is trying to get more ones, and does, it is not sufficient to just say he got more ones. You have to assess the statistical significance.
--------------------------------------------------------------------------------

yea, the fact that whenever he tries to influence the RNG to produce more 1's and it complies, and the fact that whenever he tries to influence the RNG to produce more 0's and it complies is wholly irrelevant!

You are saying that as though such decisive results have ever actually been found. If it was that clear cut, then no statistical analysis would be necessary. But in practice, it is not. When people say that the subject has been able to influence the output of the RNG, they are talking about a very small effect, where there is a slight statistical correlation between the output of the RNG, and the subject's intention. In these cases, statistical analysis is necessary to establish whether there has actually been any effect at all!

And in order to do that, you have to know something about the statistical distribution.
--------------------------------------------------------------------------------

I presume the distribution would be random.

Once again, this establishes your ignorance of the subject. What kind of random distribution? Poisson? Gaussian? Uniform? Uncorrelated? Exponentially correlated? Power-law correlated?

The statistics used assume an uncorrelated Poisson process. If the actual bit-stream is not an uncorrelated 50/50 Poisson process, then the statistical results are meaningless. In practice, the bit-stream is only approximately such a process. There are always biases, both in the relative probabilities, and in the correlations. If these biases have not been sufficiently controlled for, then the results are meaningless.

The statistics are all done under the assumption of no bias. If there is a bias, then this assumption is not valid, and the statistical confidence cannot be determined.
--------------------------------------------------------------------------------

Why should there be a bias? And if there is why can't it be eliminated?

There is always a bias, because there is always influence on any physical system by outside sources. It is not possible to completely shield against these influences. At best, one can do tests to make sure that the external bias is below a certain level. As I already explained, this puts a limit on how sensitive the statistical test can be. This is simply a fact of life that anybody doing experimental data analysis has to deal with.

The situation is even more complicated, because (a) Even if there is no systematic bias towards 0s or 1s, a bias in the variability of the distribution can result in false positives.
--------------------------------------------------------------------------------

But heaven forbid never false negatives! At lease not if we're trying to investigate anomalous abilities.

False negatives? All a negative result means is that you have failed to find what you are looking for. This is standard in science. For example, in particle physics, when people were attempting to falsify the hypothesis that Neutrinos have zero rest mass, the conclusions of those (failed) attempts were never "The neutrino has zero rest mass". The conclusion was always "We have established that the rest-mass of the neutrino is less than epsilon", where epsilon is the limit of how sensitive their analysis could be, based on the physical constraints of the experiment.

And (b) the XOR process that is used to reduce the systematic bias in the stream also has the effect of amplifying any correlations that are present in the data, but the statistics of course have to assume that there are no correlations between bits.
--------------------------------------------------------------------------------

Oh well, that's the mysterious XOR process for you! LOL

There is nothing mysterious about XOR, Ian. It is defined by the following logic table.


a b a XOR b
0 1 1
1 0 1
0 0 0
1 1 0


The point being that if you have a Poisson process whose relative probabilities are not 50/50, the operation of taking the XOR of each pair will produce a Poisson process with closer to 50/50 relative probabilities. Like I said, though, if the process is not uncorrelated, then the XOR operation also has the effect of increasing the correlations, as does any operation that gives a result that depends on consecutive inputs.


FutileJester,

I believe this could be controlled for. For instance, you could use 'control' RNG's running at the same time under similar conditions. Say you use 5 RNG's, all of which are recorded for every trial. In each trial, one of the 5 RNG's is chosen at random to be influenced by the subject. Is this statistically valid, in the sense that a confidence could be assigned a priori? Stats experts, help me out here...

That is one way to control for one type of bias, but it will not completely eliminate it, and it will not control for all types of bias. The best you can do is perform extensive testing to empirically determine what the extent of the bias is, and then design your test so that this level of bias will not skew the results.


Dr. Stupid

[Interesting Ian]

Originally posted by Stimpson J. Cat
[b]Ian,



I never claimed that it discredits anything. I was merely pointing out that you don't have to take "skeptics" word for the fact that these tests don't show statistically significant results. There own studies admit this fact.



no, I am talking about the people actually doing this research with these RNG's, who admit that the only way they can get "significant" results, is with meta-analysis.



What does that mean? Do you have any experience with statistical data analysis?



You are saying that as though such decisive results have ever actually been found. If it was that clear cut, then no statistical analysis would be necessary. But in practice, it is not. When people say that the subject has been able to influence the output of the RNG, they are talking about a very small effect, where there is a slight statistical correlation between the output of the RNG, and the subject's intention. In these cases, statistical analysis is necessary to establish whether there has actually been any effect at all!



Once again, this establishes your ignorance of the subject. What kind of random distribution? Poisson? Gaussian? Uniform? Uncorrelated? Exponentially correlated? Power-law correlated?

The statistics used assume an uncorrelated Poisson process. If the actual bit-stream is not an uncorrelated 50/50 Poisson process, then the statistical results are meaningless. In practice, the bit-stream is only approximately such a process. There are always biases, both in the relative probabilities, and in the correlations. If these biases have not been sufficiently controlled for, then the results are meaningless.



There is always a bias, because there is always influence on any physical system by outside sources. It is not possible to completely shield against these influences. At best, one can do tests to make sure that the external bias is below a certain level. As I already explained, this puts a limit on how sensitive the statistical test can be. This is simply a fact of life that anybody doing experimental data analysis has to deal with.



False negatives? All a negative result means is that you have failed to find what you are looking for. This is standard in science. For example, in particle physics, when people were attempting to falsify the hypothesis that Neutrinos have zero rest mass, the conclusions of those (failed) attempts were never "The neutrino has zero rest mass". The conclusion was always "We have established that the rest-mass of the neutrino is less than epsilon", where epsilon is the limit of how sensitive their analysis could be, based on the physical constraints of the experiment.



There is nothing mysterious about XOR, Ian. It is defined by the following logic table.


a b a XOR b
0 1 1
1 0 1
0 0 0
1 1 0


The point being that if you have a Poisson process whose relative probabilities are not 50/50, the operation of taking the XOR of each pair will produce a Poisson process with closer to 50/50 relative probabilities. Like I said, though, if the process is not uncorrelated, then the XOR operation also has the effect of increasing the correlations, as does any operation that gives a result that depends on consecutive inputs.




You're sooo lacking in a sense of humour :(

[jj]

Originally posted by Interesting Ian


You're sooo lacking in a sense of humour :(

What substance! What evidence! I am awed.

But it would be better were I not to say by what.

Ian, I was quite serious about looking into signal detection algorithms.

The question I asked is pretty easy to answer. If you aren't going to take a whack at it, you need to reconsider your allegations of rampant type 2 errors.

[Interesting Ian]

Originally posted by Stimpson J. Cat



quote:
--------------------------------------------------------------------------------
Well, considering that the parapsychologist doing these tests say so too, I guess the fact that we skeptics are all a bunch of liars doesn't really matter.
--------------------------------------------------------------------------------

The parapsychologist?? Could it really be? You've managed to locate a parapsychologist who hasn't had positive results for micro-psychokinesis?? Oh well, this clearly discredits the whole field of micro-psychokinesis then!
--------------------------------------------------------------------------------



I never claimed that it discredits anything. I was merely pointing out that you don't have to take "skeptics" word for the fact that these tests don't show statistically significant results. There own studies admit this fact.



I do have to take the "skeptics" word for it because I am not aware of any such studies. I am averse to doing this as I find that they consistantly misrepresent the results.



--------------------------------------------------------------------------------
Who's this parapsychologist then? Hang on, let me guess. Susan Blackmore??
--------------------------------------------------------------------------------



no, I am talking about the people actually doing this research with these RNG's, who admit that the only way they can get "significant" results, is with meta-analysis.


Yes, because it cannot be consistantly replicated and the results are variable. But of course meta-analysis is fine. The real problem is that we are lacking an overarching theory to accommodate this putative phenomenon. We have no hypothesis/theory to test which means we have to be less confident in the results than we otherwise would be, and hence whether this phenomenon even exists.


--------------------------------------------------------------------------------
It isn't that simple.
--------------------------------------------------------------------------------

Of course not. it wouldn't be would it!?
--------------------------------------------------------------------------------



What does that mean? Do you have any experience with statistical data analysis?


No absolutely none whatsoever. It's just that I knew you would say that.



--------------------------------------------------------------------------------
And in order to do that, you have to know something about the statistical distribution.
--------------------------------------------------------------------------------

I presume the distribution would be random.
--------------------------------------------------------------------------------



Once again, this establishes your ignorance of the subject. What kind of random distribution? Poisson? Gaussian? Uniform? Uncorrelated? Exponentially correlated? Power-law correlated?



There cannot possibly be different kinds. If there were different kinds it wouldn't be random now would it?? :rolleyes: If a RNG was outputting 0's and 1's there could not possibly be any pattern to their distribution as each digit outputted is wholly independent of any of the other digits outputted.



--------------------------------------------------------------------------------
The situation is even more complicated, because (a) Even if there is no systematic bias towards 0s or 1s, a bias in the variability of the distribution can result in false positives.
--------------------------------------------------------------------------------

But heaven forbid never false negatives! At lease not if we're trying to investigate anomalous abilities.
--------------------------------------------------------------------------------



False negatives?



Yup, I was just trying to inject a bit of levity :) If someone were trying to influence the output of an RNG, and they apparently failed to do so, the "skeptics" would never consider it's because of a possible bias in the machine cancelling out what would have been a positive effect.



--------------------------------------------------------------------------------
And (b) the XOR process that is used to reduce the systematic bias in the stream also has the effect of amplifying any correlations that are present in the data, but the statistics of course have to assume that there are no correlations between bits.
--------------------------------------------------------------------------------

Oh well, that's the mysterious XOR process for you! LOL
--------------------------------------------------------------------------------



There is nothing mysterious about XOR, Ian.



Oh dear me! :rolleyes: Are you real?




It is defined by the following logic table.


What's a "logic table"??



code:--------------------------------------------------------------------------------
a b a XOR b
0 1 1
1 0 1
0 0 0
1 1 0
--------------------------------------------------------------------------------


That conveys no meaning to me.





The point being that if you have a Poisson process



Ah yes! the mysterious poison process! Er . . is it? LMAO!




whose relative probabilities are not 50/50,



The relative probabilitites of what? {Takes wild stab in the dark} You mean the scenario where the output of 0's and 1's are nopt exactly equal to each other??



the operation of taking the XOR of each pair will produce a Poisson process with closer to 50/50 relative probabilities.


So you claim. But as I have absolutely zero idea of what you're talking about I can't comment. You're going to have to go into much greater detail I'm afraid.

[Interesting Ian]

Originally posted by jj

II
You're sooo lacking in a sense of humour
--------------------------------------------------------------------------------


jj
What substance! What evidence! I am awed.


Non-sequitur



But it would be better were I not to say by what.



What are you babbling on about??



Ian, I was quite serious about looking into signal detection algorithms.


Then go do so and stop typing meaningless posts.



The question I asked is pretty easy to answer. If you aren't going to take a whack at it, you need to reconsider your allegations of rampant type 2 errors.



What the f*ck are you talking about??? What the f*ck is a type 2 error??

Listen d*ckhead, either speak in plain English if you wish to communicate with me, or don't communicate with me at all. Do I make myself abundantly clear???

[BillHoyt]

Originally posted by Interesting Ian

What the f*ck are you talking about??? What the f*ck is a type 2 error??

Ian,

A bit of advice that will probably go unheeded. If you wish to criticize science it helps to understand science. If you wish to learn about things you don't understand, there are several of us here willing to help you. Our willingness decreases somewhat with responses like this.Our willingness also diminishes when you take us around the block trying to pose as somebody who understands the issues.

Type 1 error is to reject a true null hypothesis. Type II error is to accept a false null hypothesis.

Cheers,

[Stimpson J. Cat]

Ian,

I never claimed that it discredits anything. I was merely pointing out that you don't have to take "skeptics" word for the fact that these tests don't show statistically significant results. There own studies admit this fact.
--------------------------------------------------------------------------------

I do have to take the "skeptics" word for it because I am not aware of any such studies. I am averse to doing this as I find that they consistantly misrepresent the results.

Are you aware of any studies where statistically significant results using RNG's were reported, that were not based on meta-analysis? I certainly am not.

no, I am talking about the people actually doing this research with these RNG's, who admit that the only way they can get "significant" results, is with meta-analysis.
--------------------------------------------------------------------------------

Yes, because it cannot be consistantly replicated and the results are variable. But of course meta-analysis is fine.

I have already explained why meta-analysis is not fine. Do you have a response to my actual argument, or are you just going to ignore what I said, and continue to insist that it "fine", even when experts in that field explain to you why it isn't?

The real problem is that we are lacking an overarching theory to accommodate this putative phenomenon. We have no hypothesis/theory to test which means we have to be less confident in the results than we otherwise would be, and hence whether this phenomenon even exists.

Not less confident, no confidence. There is no falsifiable hypothesis for Psi phenomena. Without a falsifiable hypothesis, there can be no reliable supporting evidence, and without reliable supporting evidence, there is no logical reason to believe that such phenomena exist.

Note that I am not saying this proves that such phenomena don't exist. That is an untestable hypothesis. I am just saying that there is no rational reason to believe they do.

Once again, this establishes your ignorance of the subject. What kind of random distribution? Poisson? Gaussian? Uniform? Uncorrelated? Exponentially correlated? Power-law correlated?
--------------------------------------------------------------------------------

There cannot possibly be different kinds. If there were different kinds it wouldn't be random now would it?? If a RNG was outputting 0's and 1's there could not possibly be any pattern to their distribution as each digit outputted is wholly independent of any of the other digits outputted.

What you have just described is an uncorrelated Poisson process. There are other types of random processes. There are, in fact, many different types of random processes. Furthermore, no physical RNG device is a perfect uncorrelated Poisson process, hence my earlier statement that the null-hypothesis is trivially false. We already know for a fact that the RNG is not a perfect uncorrelated Poisson process. The only question is whether it deviates from such a process by more than can be accounted for by the biases and artifacts that we already know to be present.

False negatives?
--------------------------------------------------------------------------------

Yup, I was just trying to inject a bit of levity If someone were trying to influence the output of an RNG, and they apparently failed to do so, the "skeptics" would never consider it's because of a possible bias in the machine cancelling out what would have been a positive effect.

Sure they would. They would just consider it to be far more likely that the reason no effect was measured, is because there was no effect there to be measured. That is, after all, the most parsimonious explanation, given all the available information.

There is nothing mysterious about XOR, Ian.
--------------------------------------------------------------------------------

Oh dear me! Are you real?

I think so. :p

It is defined by the following logic table.
--------------------------------------------------------------------------------

What's a "logic table"??

Oh brother. :confused:

Tell me something, Ian. Why would somebody with absolutely no knowledge of how statistical testing works, or even a basic knowledge of formal logic, take it upon themselves to try to argue about the validity of the statistical testing being done by the JREF? You have made it abundantly clear that you are not qualified to comment on, much less criticize, the methods being used.

The point being that if you have a Poisson process
--------------------------------------------------------------------------------

Ah yes! the mysterious poison process! Er . . is it? LMAO!

The fact that you don't know what a Poisson process is doesn't bother me, but the fact that you seem to proud of your ignorance does. And the fact that you presume to say something like

There cannot possibly be different kinds. If there were different kinds it wouldn't be random now would it??

when you are apparently aware that you know nothing about random processes and statistics, says quite a bit about your rationality.

whose relative probabilities are not 50/50,
--------------------------------------------------------------------------------

The relative probabilitites of what? {Takes wild stab in the dark} You mean the scenario where the output of 0's and 1's are nopt exactly equal to each other??

Yes, the relative probability of getting a zero or a one.

the operation of taking the XOR of each pair will produce a Poisson process with closer to 50/50 relative probabilities.
--------------------------------------------------------------------------------

So you claim. But as I have absolutely zero idea of what you're talking about I can't comment. You're going to have to go into much greater detail I'm afraid.

I am not going to teach you basic formal logic and statistics, if that is what you mean. I have neither the time, nor the necessary facilities, to do so. If you really want to know, I suggest you find an introductory text on the subject.

If you don't really want to know, then I can only ask why you would presume to criticize people for using methods you don't understand, and why you would presume to tell experts in a field that you know nothing about, that they are wrong.

For example, you said this above

I do have to take the "skeptics" word for it because I am not aware of any such studies. I am averse to doing this as I find that they consistantly misrepresent the results.

How would you know? You clearly do not have the necessary understanding of statistical analysis to be able to understand the arguments put forth by either party, so why do you automagically assume that the skeptic is the one misrepresenting the facts? And if you really care about what the facts are, why don't you expend the effort necessary to get a minimal understanding of the subject, so that you can make such a determination?

Instead you seem to revel in your ignorance, and believe whatever you want to believe, content in the knowledge that nobody can possibly convince you that you are wrong, as long as you lack the necessary background to understand their arguments.


Dr. Stupid

[Interesting Ian]

Originally posted by BillHoyt
[B]

Ian,

A bit of advice that will probably go unheeded. If you wish to criticize science it helps to understand science.



Where have I criticized science? As far as I am aware I have no problems with science. Scientists maybe, but not science. Enlighten me as to what problems I have with science.



If you wish to learn about things you don't understand,



Indeed I do. However I am not aware of anything that I do not understand regarding that which I have raised in this thread.



there are several of us here willing to help you.



I do not require any help.



Our willingness decreases somewhat with responses like this.



You honestly think I give a flying f*ck?



Our willingness also diminishes when you take us around the block trying to pose as somebody who understands the issues.



If there are any issues that I don't understand then let me know what they might be. How do they refute anything I have said in this thread?



Type 1 error is to reject a true null hypothesis. Type II error is to accept a false null hypothesis.



None of the strings of words "type 1 error", "null hypothesis", type II error" convey any meaning. If you wish to communicate with me then I suggest you speak in English.

Otherwise shut the f*ck up.

[Interesting Ian]

Originally posted by Stimpson J. Cat
quote:
--------------------------------------------------------------------------------
I never claimed that it discredits anything. I was merely pointing out that you don't have to take "skeptics" word for the fact that these tests don't show statistically significant results. There own studies admit this fact.
--------------------------------------------------------------------------------

I do have to take the "skeptics" word for it because I am not aware of any such studies. I am averse to doing this as I find that they consistantly misrepresent the results.
--------------------------------------------------------------------------------



Are you aware of any studies where statistically significant results using RNG's were reported, that were not based on meta-analysis? I certainly am not.



No I do not. Indeed I have very little knowledge of parapsychological research, nor indeed am I particularly interested in it.



--------------------------------------------------------------------------------
no, I am talking about the people actually doing this research with these RNG's, who admit that the only way they can get "significant" results, is with meta-analysis.
--------------------------------------------------------------------------------

Yes, because it cannot be consistantly replicated and the results are variable. But of course meta-analysis is fine.
--------------------------------------------------------------------------------



I have already explained why meta-analysis is not fine.



Where?



Do you have a response to my actual argument,



What argument? An argument trying to establish what?



or are you just going to ignore what I said, and continue to insist that it "fine",



As far as I am aware meta-analysis is fine. This is not to suggest for one moment that it is ideal but, due to the variance in the results in parapsychological research, that is all we have. If you're claiming that therefore we cannot have scientific proof of micro-psychokinesis then {shrugs} fine. I feel the problem is more that of not having an underlying theoretical framework in which to accommodate the putative phenomena though.




even when experts in that field explain to you why it isn't?


Using your interpretation one could argue that nothing is fine since whatever method we employ is prone to error or might be susceptible to erroneous conclusions.



--------------------------------------------------------------------------------
The real problem is that we are lacking an overarching theory to accommodate this putative phenomenon. We have no hypothesis/theory to test which means we have to be less confident in the results than we otherwise would be, and hence whether this phenomenon even exists.
--------------------------------------------------------------------------------



Not less confident, no confidence. There is no falsifiable hypothesis for Psi phenomena.



It would be "falsified" if the preponderance of scientific research were to show results which were not of statistical significance. Of course as I have mentioned many times before, this would not make one iota of a difference to my personnal belief in certain appropriate "paranormal" phenomena. Scientific "falsification" does not invalidate anecdotal evidence and personnel experiences. It's as simple as that. Deal with it.



Without a falsifiable hypothesis, there can be no reliable supporting evidence,



Either there is supporting evidence or there is not. It is meaningless to prefix the word "reliable" in front of "supporting evidence". Of course what constitutes evidence may cease to be evidence once we have further appropriate data. Like other "skeptics" and so it seems scientists, you do not understand what the word evidence means. Evidence simply means that which renders a certain hypothesis as being more likely being true. Evidence for a hypothesis certainly need not entail that it is likely to be true.



and without reliable supporting evidence, there is no logical reason to believe that such phenomena exist.



The word "logical" in front of reason conveys nothings. The pertinent point is that there are excellent reasons to believe that certain appropriate "paranormal" phenomena exists. Even if it were "falsified".

I shall answer the rest of your post when I have time.

[Stimpson J. Cat]

Ian,

No I do not. Indeed I have very little knowledge of parapsychological research, nor indeed am I particularly interested in it.

The why did you claim that skeptics are consistently misrepresenting the results of such research? If you know so little about, and have so little interest in, parapsychological research, then what makes you think that skeptics are misrepresenting the results of such research?

I have already explained why meta-analysis is not fine.
--------------------------------------------------------------------------------

Where?

Does this ring a bell?

Of course, as is the case with any physical system, there will be a certain amount of bias in the system (1 being more or less likely than 0). This bias may vary as a function of time, as well.

To compensate for this bias, they do a bitwise exclusive-or (XOR) operation on the bit-stream. This does not destroy the bias, but reduces it greatly. Because there is still alway going to be a bias, they do control studies to make sure that the bias is beneath a set threshold. This allows them to get reliable statistics for data sets up to a certain size.

Unfortunately, such tests never show anything. People then demand that meta-analysis be done. They group several studies together, and low and behold, they get a statistically significant result, not realizing that it is probably just due to the bias, which was not controlled for at a high enough level for the new, longer data set.

I was explaining that meta-analysis cannot be applied to this type of data, because the bias is only controlled for to a certain level (the level required for the original tests), and in order for the meta-analysis to be valid, the bias would have to be controlled for to a much higher level. There are other problems with meta-analysis, too, but this alone is enough to invalidate it. This is also why meta-analysis is not used for hypothesis testing in any legitimate field of scientific research.

or are you just going to ignore what I said, and continue to insist that it "fine",
--------------------------------------------------------------------------------

As far as I am aware meta-analysis is fine. This is not to suggest for one moment that it is ideal but, due to the variance in the results in parapsychological research, that is all we have.

It doesn't have to be. The purpose of meta-analysis is to increase the effective sample-size. If this is what is necessary to get statistically significant results, then what these researchers should be doing is redesigning their tests to use larger sample sizes (and of course, controlling for the bias at the much higher level, required for the larger sample-size).

If you're claiming that therefore we cannot have scientific proof of micro-psychokinesis then {shrugs} fine.

If it exists, then we should be able to find evidence for it. It is just matter of designing the experiment properly. Of course, the very first step should be to construct a testable hypothesis...

I feel the problem is more that of not having an underlying theoretical framework in which to accommodate the putative phenomena though.

That is the entire problem. Without a testable theory, all of this so-called research is a complete waste of time.

even when experts in that field explain to you why it isn't?
--------------------------------------------------------------------------------

Using your interpretation one could argue that nothing is fine since whatever method we employ is prone to error or might be susceptible to erroneous conclusions.

I already explained that it is possible to control for the bias up to a certain level. The problem arises when you attempt to attach significance to events which are so small that they could be the result of uncontrolled-for bias.

Not less confident, no confidence. There is no falsifiable hypothesis for Psi phenomena.
--------------------------------------------------------------------------------

It would be "falsified" if the preponderance of scientific research were to show results which were not of statistical significance.

That is not falsification. That is just lack of supporting evidence.

Of course as I have mentioned many times before, this would not make one iota of a difference to my personnal belief in certain appropriate "paranormal" phenomena. Scientific "falsification" does not invalidate anecdotal evidence and personnel experiences. It's as simple as that. Deal with it.

Scientific evidence does not have to "invalidate" anecdotal evidence or personal experience. They invalidate themselves, by virtue of the fact that they are unreliable.

Without a falsifiable hypothesis, there can be no reliable supporting evidence,
--------------------------------------------------------------------------------

Either there is supporting evidence or there is not. It is meaningless to prefix the word "reliable" in front of "supporting evidence".

I prefixed the word "reliable" in front, because you seem to think that clearly unreliable things, like anecdotes and personal experience, can also constitute "supporting evidence".

Of course what constitutes evidence may cease to be evidence once we have further appropriate data. Like other "skeptics" and so it seems scientists, you do not understand what the word evidence means. Evidence simply means that which renders a certain hypothesis as being more likely being true.

Exactly. And only reliable information can have any relevance to the likelihood of a hypothesis being true. The existence of anecdotal accounts, and your own subjective interpretations of your personal experiences, have absolutely no relevance to the likelihood of a hypothesis being true. That is why they are unreliable.

Evidence for a hypothesis certainly need not entail that it is likely to be true.

Huh? Doesn't this contradict what you just said?

and without reliable supporting evidence, there is no logical reason to believe that such phenomena exist.
--------------------------------------------------------------------------------

The word "logical" in front of reason conveys nothings.

Of course it does. It makes the distinction between a logical reason for believing something, and an emotional or intuitive reason.

The pertinent point is that there are excellent reasons to believe that certain appropriate "paranormal" phenomena exists. Even if it were "falsified".

Like what? Certainly the anecdotal evidence, and personal experiences that you cited, do not qualify.

And how can there be a good reason to believe anything that has been falsified? If there is a significant likelihood that it is true, then it necessarily has not been falsified.

Dr. Stupid

[Interesting Ian]

Originally posted by Stimpson J. Cat

--------------------------------------------------------------------------------
Once again, this establishes your ignorance of the subject. What kind of random distribution? Poisson? Gaussian? Uniform? Uncorrelated? Exponentially correlated? Power-law correlated?
--------------------------------------------------------------------------------

There cannot possibly be different kinds. If there were different kinds it wouldn't be random now would it?? If a RNG was outputting 0's and 1's there could not possibly be any pattern to their distribution as each digit outputted is wholly independent of any of the other digits outputted.
--------------------------------------------------------------------------------



What you have just described is an uncorrelated Poisson process. There are other types of random processes. There are, in fact, many different types of random processes.



No there isn't.



Furthermore, no physical RNG device is a perfect uncorrelated Poisson process, hence my earlier statement that the null-hypothesis is trivially false. We already know for a fact that the RNG is not a perfect uncorrelated Poisson process. The only question is whether it deviates from such a process by more than can be accounted for by the biases and artifacts that we already know to be present.



Fine, is anyone disputing this?



--------------------------------------------------------------------------------
False negatives?
--------------------------------------------------------------------------------

Yup, I was just trying to inject a bit of levity If someone were trying to influence the output of an RNG, and they apparently failed to do so, the "skeptics" would never consider it's because of a possible bias in the machine cancelling out what would have been a positive effect.
--------------------------------------------------------------------------------



Sure they would.



Allow me to exercise a healthy dose of scepticism towards your assertion here.


--------------------------------------------------------------------------------
There is nothing mysterious about XOR, Ian.
--------------------------------------------------------------------------------

Oh dear me! Are you real?
--------------------------------------------------------------------------------



I think so.


quote:
--------------------------------------------------------------------------------
It is defined by the following logic table.
--------------------------------------------------------------------------------

What's a "logic table"??
--------------------------------------------------------------------------------



Oh brother.

Tell me something, Ian. Why would somebody with absolutely no knowledge of how statistical testing works, or even a basic knowledge of formal logic, take it upon themselves to try to argue about the validity of the statistical testing being done by the JREF?



Where did I do this? Indeed how could I do this since I have very little knowledge and indeed interest in thier particular testing protocols?



--------------------------------------------------------------------------------
The point being that if you have a Poisson process
--------------------------------------------------------------------------------

Ah yes! the mysterious poison process! Er . . is it? LMAO!
--------------------------------------------------------------------------------



The fact that you don't know what a Poisson process is doesn't bother me, but the fact that you seem to proud of your ignorance does.



{shrugs} Can't help that.


And the fact that you presume to say something like


quote:
--------------------------------------------------------------------------------
There cannot possibly be different kinds. If there were different kinds it wouldn't be random now would it??
--------------------------------------------------------------------------------



And what I say certainly to seems to be correct. To deny this you would need to employ a differing definition of randomness to myself.



when you are apparently aware that you know nothing about random processes and statistics, says quite a bit about your rationality.



I haven't been educated in such. This does not at all imply that I know nothing about these things. A random process is a random process. There's nowt I need to be educated about.



--------------------------------------------------------------------------------
whose relative probabilities are not 50/50,
--------------------------------------------------------------------------------

The relative probabilitites of what? {Takes wild stab in the dark} You mean the scenario where the output of 0's and 1's are nopt exactly equal to each other??
--------------------------------------------------------------------------------



Yes, the relative probability of getting a zero or a one.


Okie dokie.


--------------------------------------------------------------------------------
the operation of taking the XOR of each pair will produce a Poisson process with closer to 50/50 relative probabilities.
--------------------------------------------------------------------------------

So you claim. But as I have absolutely zero idea of what you're talking about I can't comment. You're going to have to go into much greater detail I'm afraid.
--------------------------------------------------------------------------------



I am not going to teach you basic formal logic and statistics, if that is what you mean. I have neither the time, nor the necessary facilities, to do so. If you really want to know, I suggest you find an introductory text on the subject.



No thanks, I'm not interested.



If you don't really want to know, then I can only ask why you would presume to criticize people for using methods you don't understand,



What methods have I criticized and what don't I understand about them?




and why you would presume to tell experts in a field that you know nothing about, that they are wrong.

For example, you said this above


quote:
--------------------------------------------------------------------------------
I do have to take the "skeptics" word for it because I am not aware of any such studies. I am averse to doing this as I find that they consistantly misrepresent the results.
--------------------------------------------------------------------------------



How would you know?



I don't know. Nevertheless, given that on almost all occasions so called "skeptics" either tell outright lies, or seriously misrepresent the facts, then by inductive reasoning it would be rational for myself to hold any future utterances by them regarding the subject matter of the "paranormal" with a great deal of suspicion.

You know, induction. You have heard of it haven't you?



You clearly do not have the necessary understanding of statistical analysis to be able to understand the arguments put forth by either party,


Listen dickhead! if there is anything wrong with my f*cking arguments I have put forth in this thread then be so good as to point them out. Go on, why don't you do so??

Otherwise shut the f*ck up!

Stimpy,


----
Unfortunately, such tests never show anything. People then demand that meta-analysis be done. They group several studies together, and low and behold, they get a statistically significant result,
----


Which in itself, doesn't mean anything was done wrong. It could be that all the individual effects, while non-significant, were all positive, for example, so when you synthesize all the results, a lot of small non-significant results add up to a very significant overall result.

[Skeptical Greg]

Originally posted by Whodini
Stimpy,


----
Unfortunately, such tests never show anything. People then demand that meta-analysis be done. They group several studies together, and low and behold, they get a statistically significant result,
----


Which in itself, doesn't mean anything was done wrong. It could be that all the individual effects, while non-significant, were all positive, for example, so when you synthesize all the results, a lot of small non-significant results add up to a very significant overall result.

I think the beach is an example of this..:)

[Interesting Ian]

Originally posted by Stimpson J. Cat

I can't comment on your assertions about meta-analysis apart from to say 2 things. Firstly if meta-analysis really does imply nothing whatsoever it seems to me to be extremely surprising that it is so widely used. Secondly, seeing as you're such a dumbf*ck when it comes to philosophical argumentation, why should I suppose your reasoning ability is any better in science??




--------------------------------------------------------------------------------
If you're claiming that therefore we cannot have scientific proof of micro-psychokinesis then {shrugs} fine.
--------------------------------------------------------------------------------



If it exists, then we should be able to find evidence for it. It is just matter of designing the experiment properly. Of course, the very first step should be to construct a testable hypothesis...



Such a hypothesis would be likely to be a metaphysical one. It is difficult to test for metaphysical hypothesis. The truth of the matter is, is that phenomenal consciousness lies outside the purview of science. Therefore if phenomenal consciousness is not amenable to a scientific hypothesis, why do you suppose particular abilities of consciousness are any more susceptible?



--------------------------------------------------------------------------------
I feel the problem is more that of not having an underlying theoretical framework in which to accommodate the putative phenomena though.
--------------------------------------------------------------------------------



That is the entire problem.



Exactly. But it was I who had to point it out.


Without a testable theory, all of this so-called research is a complete waste of time.


Not at all. Patterns can be discerned and that might enable us to dream up better conjecture as to the scope, characteristics and range of these abilities. Also light may be shed on the ultimate nature of consciousness. i think that vastly more money should be ploughed into parapsychological research. They are starved of funding.




--------------------------------------------------------------------------------
Not less confident, no confidence. There is no falsifiable hypothesis for Psi phenomena.
--------------------------------------------------------------------------------

It would be "falsified" if the preponderance of scientific research were to show results which were not of statistical significance.
--------------------------------------------------------------------------------



That is not falsification. That is just lack of supporting evidence.



Ok, only a hypothesis or theory can be "falsified" then.



--------------------------------------------------------------------------------
Of course as I have mentioned many times before, this would not make one iota of a difference to my personnal belief in certain appropriate "paranormal" phenomena. Scientific "falsification" does not invalidate anecdotal evidence and personnel experiences. It's as simple as that. Deal with it.
--------------------------------------------------------------------------------



Scientific evidence does not have to "invalidate" anecdotal evidence or personal experience. They invalidate themselves, by virtue of the fact that they are unreliable.



The fact that evidence is unreliable does not mean that it is not evidence.



--------------------------------------------------------------------------------
Without a falsifiable hypothesis, there can be no reliable supporting evidence,
--------------------------------------------------------------------------------

Either there is supporting evidence or there is not. It is meaningless to prefix the word "reliable" in front of "supporting evidence".
--------------------------------------------------------------------------------



I prefixed the word "reliable" in front, because you seem to think that clearly unreliable things, like anecdotes and personal experience, can also constitute "supporting evidence".


No evidence is absolutely incorrigible, not even scientific evidence.



--------------------------------------------------------------------------------
Of course what constitutes evidence may cease to be evidence once we have further appropriate data. Like other "skeptics" and so it seems scientists, you do not understand what the word evidence means. Evidence simply means that which renders a certain hypothesis as being more likely being true.
--------------------------------------------------------------------------------



Exactly. And only reliable information can have any relevance to the likelihood of a hypothesis being true.



It's not as simple as that. There is not the rigid dichotomy you presume of either reliable or unreliable. The reliablity of evidence I would say rather falls on a continuum from not reliable at all, to a reliability which we can fairly confidently trust.

Thus if a single individual witnessed a phenomenon which clearly contravened our understanding of the world, and moreover that phenomenon had never ever been reported by anyone else, then that would be extremely unreliable. However, if a pheonmenon were reported by numerous witnesses, and moreover this phenomenon had been reported throughout human history and accross all cultures, then we can have a fairly high degree of confidence it really occurred, even though it may have a underlying "normal" explanation.


--------------------------------------------------------------------------------
Evidence for a hypothesis certainly need not entail that it is likely to be true.
--------------------------------------------------------------------------------



Huh? Doesn't this contradict what you just said?



No not at all. For example maybe we could assign a probability of 1% of some hypothesis being true. now some evidence may come to light which raises that probability to 2%. Still very unlikely to be true though!



--------------------------------------------------------------------------------
and without reliable supporting evidence, there is no logical reason to believe that such phenomena exist.
--------------------------------------------------------------------------------

The word "logical" in front of reason conveys nothings.
--------------------------------------------------------------------------------



Of course it does. It makes the distinction between a logical reason for believing something, and an emotional or intuitive reason.



no. An emotional reason or intuitive reason are oxymorons. Note what the word reason means.



--------------------------------------------------------------------------------
The pertinent point is that there are excellent reasons to believe that certain appropriate "paranormal" phenomena exists. Even if it were "falsified".
--------------------------------------------------------------------------------



Like what?



Well apparitions to give an arbitrary example.


Certainly the anecdotal evidence, and personal experiences that you cited, do not qualify.



I have made it clear I don't agree.



And how can there be a good reason to believe anything that has been falsified? If there is a significant likelihood that it is true, then it necessarily has not been falsified.



I agree it might seem so. But if some theory accounting for paranormal phenomena were "falsified" (not that theories can strictly speaking be falsified as one can introduce auxiliary hypotheses, call observations into question etc) another theory accounting for the phenomena might be able to be dreamed up.

[Stimpson J. Cat]

Ian,

What you have just described is an uncorrelated Poisson process. There are other types of random processes. There are, in fact, many different types of random processes.
--------------------------------------------------------------------------------

No there isn't.

:confused:

Furthermore, no physical RNG device is a perfect uncorrelated Poisson process, hence my earlier statement that the null-hypothesis is trivially false. We already know for a fact that the RNG is not a perfect uncorrelated Poisson process. The only question is whether it deviates from such a process by more than can be accounted for by the biases and artifacts that we already know to be present.
--------------------------------------------------------------------------------

Fine, is anyone disputing this?

I was answering your question, Ian. :rolleyes:

Yup, I was just trying to inject a bit of levity If someone were trying to influence the output of an RNG, and they apparently failed to do so, the "skeptics" would never consider it's because of a possible bias in the machine cancelling out what would have been a positive effect.
--------------------------------------------------------------------------------

Sure they would.
--------------------------------------------------------------------------------

Allow me to exercise a healthy dose of scepticism towards your assertion here.

Actually, you are the one who asserted that they would "never consider it". I merely disagreed with your assertion.

Tell me something, Ian. Why would somebody with absolutely no knowledge of how statistical testing works, or even a basic knowledge of formal logic, take it upon themselves to try to argue about the validity of the statistical testing being done by the JREF?
--------------------------------------------------------------------------------

Where did I do this?

In this very thread you have made claims about the validity of the JREF testing procedure. When people tried to explain to you the statistical reasons why it is done the way it is, you insisted that they were wrong.

Indeed how could I do this since I have very little knowledge and indeed interest in thier particular testing protocols?

You tell me. It was your completely baseless criticisms of the JREF testing methods that lead the thread down this path.

And the fact that you presume to say something like

quote:
--------------------------------------------------------------------------------
There cannot possibly be different kinds. If there were different kinds it wouldn't be random now would it??
--------------------------------------------------------------------------------

And what I say certainly to seems to be correct. To deny this you would need to employ a differing definition of randomness to myself.

I have no idea what definition of randomness you are using. Perhaps you should find out what the word really means, before you use it? In particular, since we are discussing statistical hypothesis testing, and science in general, you should find out what the word means in that context, before you make statements about it, of refute other people's statements about it.

when you are apparently aware that you know nothing about random processes and statistics, says quite a bit about your rationality.
--------------------------------------------------------------------------------

I haven't been educated in such. This does not at all imply that I know nothing about these things.

You certainly don't seem to.

A random process is a random process.

That statement conveys no meaning, and you know it.

There's nowt I need to be educated about.

What?

If you don't really want to know, then I can only ask why you would presume to criticize people for using methods you don't understand,
--------------------------------------------------------------------------------

What methods have I criticized and what don't I understand about them?

You have criticized the methods that the JREF uses for deciding whether or not to accept a challenge, and in particular, the statistical demands that they make for a challenge to be accepted. When informed of the formal statistical reasons why those methods are used, you insisted that these requirements are unjustified, even though you clearly have no idea how to even begin to determine what would or would not be justified.

I do have to take the "skeptics" word for it because I am not aware of any such studies. I am averse to doing this as I find that they consistantly misrepresent the results.
--------------------------------------------------------------------------------

How would you know?
--------------------------------------------------------------------------------

I don't know. Nevertheless, given that on almost all occasions so called "skeptics" either tell outright lies, or seriously misrepresent the facts, then by inductive reasoning it would be rational for myself to hold any future utterances by them regarding the subject matter of the "paranormal" with a great deal of suspicion.

Do you have any evidence to back up this slanderous remark?

You clearly do not have the necessary understanding of statistical analysis to be able to understand the arguments put forth by either party,
--------------------------------------------------------------------------------

Listen dickhead! if there is anything wrong with my f*cking arguments I have put forth in this thread then be so good as to point them out. Go on, why don't you do so??

I did. Unfortunately you didn't understand my criticisms, and seem to take great pride in that fact. :rolleyes:

Otherwise shut the f*ck up!

Good idea. There is clearly no point in arguing with an abusive, willfully ignorant, hypocrite.

Dr. Stupid

[Stimpson J. Cat]

Whodini,

Unfortunately, such tests never show anything. People then demand that meta-analysis be done. They group several studies together, and low and behold, they get a statistically significant result,
----

Which in itself, doesn't mean anything was done wrong. It could be that all the individual effects, while non-significant, were all positive, for example, so when you synthesize all the results, a lot of small non-significant results add up to a very significant overall result.

yes, it could mean that. The problem is that there is no way of knowing whether it does mean that. The only way would be to repeat the experiment with a larger sample-size, while controlling for the external biases at an appropriate level. Meta-analysis gives the larger sample-size, but not only does it not assure that the biases were controlled for at a sufficient level, it can actually introduce new biases.


Ian,

I can't comment on your assertions about meta-analysis apart from to say 2 things. Firstly if meta-analysis really does imply nothing whatsoever it seems to me to be extremely surprising that it is so widely used.

It isn't widely used. It is occasionally used in Medical research, but as a method for determining which types of formal studies should be done, not for hypothesis testing. It is not used in other sciences at all. The only field of research where it is widely used is parapsychology.

Secondly, seeing as you're such a dumbf*ck when it comes to philosophical argumentation, why should I suppose your reasoning ability is any better in science??

Nice. I'll tell you what. Why don't you evaluate my scientific reasoning ability from my scientific work? It is all freely available, in peer-reviewed scientific journals.

Or don't. Your opinion of my scientific reasoning ability is about as important to me as your opinion of the flavor of chocolate ice-cream. :rolleyes:

If it exists, then we should be able to find evidence for it. It is just matter of designing the experiment properly. Of course, the very first step should be to construct a testable hypothesis...
--------------------------------------------------------------------------------

Such a hypothesis would be likely to be a metaphysical one. It is difficult to test for metaphysical hypothesis.

Not difficult, impossible. You can't test an untestable hypothesis. And if the only possible hypotheses for the paranormal are untestable ones, then this just means that there will never be any logical reason to believe that the paranormal exists.

The truth of the matter is, is that phenomenal consciousness lies outside the purview of science.

You have no reliable evidence to support this hypothesis. It is simply an irrational belief you hold.

Therefore if phenomenal consciousness is not amenable to a scientific hypothesis, why do you suppose particular abilities of consciousness are any more susceptible?

If it isn't, I wouldn't. But there is absolutely no rational reason to believe it isn't.

Without a testable theory, all of this so-called research is a complete waste of time.
--------------------------------------------------------------------------------

Not at all. Patterns can be discerned and that might enable us to dream up better conjecture as to the scope, characteristics and range of these abilities.

Not using the sloppy and statistically questionable methods they are using.

Also light may be shed on the ultimate nature of consciousness.

I thought you considered that to be impossible? You said "The truth of the matter is, is that phenomenal consciousness lies outside the purview of science." If that is the case, then how could parapsychological research, which claims to be a science, possibly give any insight into the nature of consciousness?

i think that vastly more money should be ploughed into parapsychological research. They are starved of funding.

Let me get this straight. You don't know anything about parapsychological research, and are not interested in it, but you know that it is starved of funding (it is not), and believe that much more money should be spent on it?

That is not falsification. That is just lack of supporting evidence.
--------------------------------------------------------------------------------

Ok, only a hypothesis or theory can be "falsified" then.

Only a falsifiable hypothesis can be falsified. I would have thought that was obvious. :rolleyes:

Scientific evidence does not have to "invalidate" anecdotal evidence or personal experience. They invalidate themselves, by virtue of the fact that they are unreliable.
--------------------------------------------------------------------------------

The fact that evidence is unreliable does not mean that it is not evidence.

It does according to the definition of evidence that you posted! And either way, certainly unreliable evidence is not a logical reason for believing something.

I prefixed the word "reliable" in front, because you seem to think that clearly unreliable things, like anecdotes and personal experience, can also constitute "supporting evidence".
--------------------------------------------------------------------------------

No evidence is absolutely incorrigible, not even scientific evidence.

I said reliable, not infallible.

Exactly. And only reliable information can have any relevance to the likelihood of a hypothesis being true.
--------------------------------------------------------------------------------

It's not as simple as that. There is not the rigid dichotomy you presume of either reliable or unreliable. The reliablity of evidence I would say rather falls on a continuum from not reliable at all, to a reliability which we can fairly confidently trust.

Reliable means, as you said, that it has some bearing on the probability of the hypothesis being true. Of course there is a range of how reliable something is, based on how much bearing it has on that probability. But anecdotes and subjective interpretations of personal experiences, are not reliable at all. They have no bearing on the probabilities.

Thus if a single individual witnessed a phenomenon which clearly contravened our understanding of the world, and moreover that phenomenon had never ever been reported by anyone else, then that would be extremely unreliable.

No, it would be completely unreliable.

However, if a pheonmenon were reported by numerous witnesses, and moreover this phenomenon had been reported throughout human history and accross all cultures, then we can have a fairly high degree of confidence it really occurred, even though it may have a underlying "normal" explanation.

Wrong again. And history has repeatedly proven this.

Huh? Doesn't this contradict what you just said?
--------------------------------------------------------------------------------

No not at all. For example maybe we could assign a probability of 1% of some hypothesis being true. now some evidence may come to light which raises that probability to 2%. Still very unlikely to be true though!

OK, I misinterpreted what you said.

Of course it does. It makes the distinction between a logical reason for believing something, and an emotional or intuitive reason.
--------------------------------------------------------------------------------

no. An emotional reason or intuitive reason are oxymorons. Note what the word reason means.

That depends on whether you are using "reason" as a noun or a verb. "Reason for doing something" uses "reason" as a noun, which is synonymous with "cause". "Using Reason to solve a problem" uses "reason" as a verb, and is synonymous with "logic".

In the context of what I said, reason means cause. Clearly a person can believe something because of logic, or because of emotion or intuition.

Like what?
--------------------------------------------------------------------------------

Well apparitions to give an arbitrary example.

Are you seriously going to claim that apparitions are a logical reason to believe in the paranormal? :eek:

Certainly the anecdotal evidence, and personal experiences that you cited, do not qualify.
--------------------------------------------------------------------------------

I have made it clear I don't agree.

What you haven't made clear, is why.

And how can there be a good reason to believe anything that has been falsified? If there is a significant likelihood that it is true, then it necessarily has not been falsified.
--------------------------------------------------------------------------------

I agree it might seem so. But if some theory accounting for paranormal phenomena were "falsified" (not that theories can strictly speaking be falsified as one can introduce auxiliary hypotheses, call observations into question etc) another theory accounting for the phenomena might be able to be dreamed up.

The specific theory can be falsified (I am not even going to bother addressing your "auxiliary hypothesis nonsense again"). A vague concept like "paranormal phenomena" cannot be. Of course you can always dream up another hypothesis for it. So what?

Dr. Stupid

Originally posted by Diogenes

I think the beach is an example of this..:)


Meta-analysis is sound as can be. I enjoy it when people say

'bahh! meta analysis! crappy woo woo! blah blah abvlh'

It goes to show they don't really understand that issues they think they have mastered.

If they wanted to say there are issues with meta-analysis, (file drawer effect, etc) fine, that is legit criticism. But a meta-analysis in itself doesn't tell you anything about if a study is good or bad.

[BillHoyt]

Originally posted by Interesting Ian

Listen d*khead! if there is anything wrong with my f*cking arguments I have put forth in this thread then be so good as to point them out. Go on, why don't you do so??

Otherwise shut the f*ck up!

Oh, I'm sorry, I thought this was argument lessons...


Cheers,

[BillHoyt]

Originally posted by Whodini



Meta-analysis is sound as can be. I enjoy it when people say

'bahh! meta analysis! crappy woo woo! blah blah abvlh'

It goes to show they don't really understand that issues they think they have mastered.

If they wanted to say there are issues with meta-analysis, (file drawer effect, etc) fine, that is legit criticism. But a meta-analysis in itself doesn't tell you anything about if a study is good or bad.

You've missed the import of both the PRNG and biased RNG comments, didn't you? Stimpy and I have already laid the groundwork. Nobody is simply saying "bah."

Cheers,

[Interesting Ian]

Originally posted by Stimpson J. Cat


In this very thread you have made claims about the validity of the JREF testing procedure. When people tried to explain to you the statistical reasons why it is done the way it is, you insisted that they were wrong.


You tell me. It was your completely baseless criticisms of the JREF testing methods that lead the thread down this path.


You have criticized the methods that the JREF uses for deciding whether or not to accept a challenge, and in particular, the statistical demands that they make for a challenge to be accepted. When informed of the formal statistical reasons why those methods are used, you insisted that these requirements are unjustified, even though you clearly have no idea how to even begin to determine what would or would not be justified.




Listen dipstick, I want to know precisely whereabouts in this thread I have made the comments you accuse me of making. How the f*ck am I able to respond unless you specify what these comments might be?? Please quote them and tell me what issues you have with them. If you refuse then I'll know you're even more full of sh*t than I hitherto had supposed. I'm waiting.

[BillHoyt]

Originally posted by Interesting Ian


Listen dipstick, I want to know precisely whereabouts in this thread I have made the comments you accuse me of making. How the f*ck am I able to respond unless you specify what these comments might be?? Please quote them and tell me what issues you have with them. If you refuse then I'll know you're even more full of sh*t than I hitherto had supposed. I'm waiting.

:rolleyes: Your comments have been legion. Why don't we start classifying them by chapter and verse? Will you then discuss them? :rolleyes:

[Skeptical Greg]

Originally posted by Interesting Ian


Listen dipstick, I want to know precisely whereabouts in this thread I have made the comments you accuse me of making. How the f*ck am I able to respond unless you specify what these comments might be?? Please quote them and tell me what issues you have with them. If you refuse then I'll know you're even more full of sh*t than I hitherto had supposed. I'm waiting.

Calling Dr. Stupid..

I hope you won't bother to do this. ( dredge up the examples )

Wading through it once, is enough..:(

[DrBenway]

This is why democracy sucks:

Interesting Ian:
"I have very little knowledge of parapsychological research, nor indeed am I particularly interested in it."
**************
"i think that vastly more money should be ploughed into parapsychological research. They are starved of funding."
:p

Originally posted by BillHoyt


You've missed the import of both the PRNG and biased RNG comments, didn't you? Stimpy and I have already laid the groundwork. Nobody is simply saying "bah."

Cheers,


And I never called out you and Stimpy...

You protest too much.

[Interesting Ian]

Originally posted by DrBenway
This is why democracy sucks:


quote:
--------------------------------------------------------------------------------

Interesting Ian:
"I have very little knowledge of parapsychological research, nor indeed am I particularly interested in it."
**************
"i think that vastly more money should be ploughed into parapsychological research. They are starved of funding."

--------------------------------------------------------------------------------



There is absolutely no contradiction in those 2 statements whatsoever. Would you care to specify how these 2 statements contradict each other? Or are you a dumbf*ck too? The implications should human beings possess anomalous cognitive abilities could scarcely be exaggerated. If there is any remote possibility that there is anything of any substance behind anecdotal reports then it should be investigated.

[DrBenway]

Originally posted by Interesting Ian

Or are you a dumbf*ck too?

My definition of a dumbf*ck: a person with passionate convictions concerning things he knows very little about.

[Interesting Ian]

Originally posted by DrBenway

My definition of a dumbf*ck: a person with passionate convictions concerning things he knows very little about.

I know more about anomalous phenomena which people would tend to label "paranormal" than the vast majority of people.

A*sewipe.

[RichardR]

Originally posted by Interesting Ian
To Stimpson
if there is anything wrong with my f*cking arguments I have put forth in this thread then be so good as to point them out. Go on, why don't you do so??
Ian, you puzzle me. Don’t you think that’s what he’s been doing?

[Skeptical Greg]

Originally posted by Interesting Ian


I know more about anomalous phenomena which people would tend to label "paranormal" than the vast majority of people.

A*sewipe.

You da' man...;)

[Oso]

Originally posted by Diogenes


Calling Dr. Stupid..

I hope you won't bother to do this. ( dredge up the examples )

Wading through it once, is enough..:( And Stimpy everybody appreciates/enjoys your thorough rebuttals.
Thanks

[Interesting Ian]

Originally posted by RichardR
if there is anything wrong with my f*cking arguments I have put forth in this thread then be so good as to point them out. Go on, why don't you do so??
--------------------------------------------------------------------------------


Ian, you puzzle me. Don’t you think that’s what he’s been doing?



No. Allow me to rephrase myself. What have I uttered which is either incorrect or fallacious??????? :mad:

[Interesting Ian]

Originally posted by Oso
And Stimpy everybody appreciates/enjoys your thorough rebuttals.
Thanks

Well I wish he f*cking would because I've had it up to here with the daft tw*t!!!!!!!!

[DrBenway]

from Interesting Ian:
"I have very little knowledge of parapsychological research, nor indeed am I particularly interested in it."

"I know more about anomalous phenomena which people would tend to label 'paranormal' than the vast majority of people."
*********

Your comments above are contradictory.

[Interesting Ian]

Originally posted by DrBenway
from Interesting Ian:
"I have very little knowledge of parapsychological research, nor indeed am I particularly interested in it."

"I know more about anomalous phenomena which people would tend to label 'paranormal' than the vast majority of people."
*********

Your comments above are contradictory.

they are in fact not at all contradictory.

[DanishDynamite]

Oso:And Stimpy everybody appreciates/enjoys your thorough rebuttals.
ThanksI second that.

What I don't understand is why he keeps replying to abusive, self-contradictory, willfully ignorant ninnies who not only don't have the first clue what they are talking about, but are incapable of defining their own viewpoint coherently, let alone backing it up.

The guy is amazing.