Randi nonsense

[geni]

Geni,

Are you involved in the JREF challenge application process and is this the official line taken by the JREF regarding effect sizes and p values?

P values have been publicaly stated in the past and appear to be fairly consistant. However my point was extrapolated from basicv stats. Once you have a P value any real effect can reach it. Of course if the varation from normal dissribution is very very very small there may be technical problems such as the sun burning out.

 

[davidsmith73]

P values have been publicaly stated in the past and appear to be fairly consistant.

Do you have a link to any written statements?

What were the p values and what experiments were they for?

However my point was extrapolated from basicv stats. Once you have a P value any real effect can reach it. Of course if the varation from normal dissribution is very very very small there may be technical problems such as the sun burning out.

I am aware of how the number of trials effect the p value.

My question is what effect size and p value the JREF would accept for a typical psi experiment to be a success in the challenge.

 

[Rasmus]

Geni,

Are you involved in the JREF challenge application process and is this the official line taken by the JREF regarding effect sizes and p values?

Genrally, none of the posters in this forum do represent JREF.

Usually, though, they have a pretty good idea of how things are or might be handled.

I think it is pretty pointless to argue about percentages and the like, though. As soon as there is a claim and an application it'll be worth anyone's while to work out the details.

 

[davidsmith73]

Genrally, none of the posters in this forum do represent JREF.

Usually, though, they have a pretty good idea of how things are or might be handled.

I think it is pretty pointless to argue about percentages and the like, though. As soon as there is a claim and an application it'll be worth anyone's while to work out the details.

The details are already there for psi experiments in the form of the method section of their papers. In the case of Randi's statement that he would accept Sheldrakes "claim" for the challenge, either the JREF accept an exact replication of his experiment, stats and all, or the criteria for success is changed. I wonder what the JREF would accept? I have emailed Randi. Hope I get a response

 

[davidsmith73]

Well I have to make a retraction. I emailed Randi asking whether the JREF would accept an exact replication of Sheldrakes email telepathy experiment for the challenge.

He replied "yes, with a standard rate of success, nothing more"

I'm not sure what a standard rate of success means so I'm waiting for clarification, but it appears that the JREF will accept at least one psi experiment, as it stands, for the challenge. I still think this should be clearly stated in SWIFT however.

I'm genuinely surprised.

 

[Rasmus]

The details are already there for psi experiments in the form of the method section of their papers. In the case of Randi's statement that he would accept Sheldrakes "claim" for the challenge, either the JREF accept an exact replication of his experiment, stats and all, or the criteria for success is changed. I wonder what the JREF would accept? I have emailed Randi. Hope I get a response

Well, that would depend on what precisely has been "claimed", right? It is at least conceivable that a person would make a claim that goes beyond the findings of their own studies, right? Or maybe make a claim that was simply less precise than what a study would suggest.

So it is still necessary that a specific claim be made as part of an application for the JREF Challenge.

It is,as far as I can see, unjustified to assume that such a claim would correspond to a particular interpretation of a particular study and it is even more unjustified to assume that the JREF should dig through the study and do the interpreting.

As always: Someone should first make a claim and take it from there. Everything else is just pointless.

 

[Gord_in_Toronto]

Is the link not working for you or something? Anyway, this bit from it:

"Viewing the entire tape, we see that the dog responded to every car that drove by, and to every person who walked by." This is simply not true, and Randi now admits that he has never seen the tape.

Did he say "I see"? As I understand it a representative for JREF saw the tape and reported what he saw. Randi, speaking on behalf of JREF said "we" refering to JREF. This is a common way of expressing things in English. Even if the representative is lieing about the dog responding to every pedestrian and car, Randi did not lie.

 

[geni]

Do you have a link to any written statements?

They were mentioned in the commentry.

What were the p values and what experiments were they for?

1000 to one (ok 1024 in the example) for prelim. 1 million to 1 for full.

 

[steve s]

Cleary, parapsychologists are not touching the challenge. I think this is because they know that the JREF will set its expectations too high for an experiment to succeed.

Don't you find it the slightest bit strange that these parapsychologists have absolutely no problem at all getting a positive result when they're in front of TV cameras or a like-minded audience? But let the JREF be present and all of a sudden they can't do better than chance. Aren't you curious about why that is?

When testing dowsers for example, they do an initial run where the dowser knows where the target is. Their magic powers are always working perfectly when they know where it is and have no concerns about setting the bar too high. But once it's hidden they fail miserably. Curious isn't it?

Steve S.

 

[davefoc]

Don't you find it the slightest bit strange that these parapsychologists have absolutely no problem at all getting a positive result when they're in front of TV cameras or a like-minded audience? But let the JREF be present and all of a sudden they can't do better than chance. Aren't you curious about why that is?

When testing dowsers for example, they do an initial run where the dowser knows where the target is. Their magic powers are always working perfectly when they know where it is and have no concerns about setting the bar too high. But once it's hidden they fail miserably. Curious isn't it?

Steve S.

For what it is worth, Steve S. I think davidsmith73 probably has about the same view of this that most of us do. I think davidsmith73 agrees there are a lot of baseless claims and that the dowsing claims might be baseless and what success they think they have is entirely explained by self delusion.

I think his point here is that JREF may not be prepared to test certain kinds of claims that involve effects that produce only slight variations from chance. I am not sure that he is right or wrong about that claim.

Geni says that no matter how small the deviation from chance an effect causes the effect can be detected by running enough tests to arrive at any desired level of confidence, but he goes on to make the point that in fact as a practical matter the time involved to test effects that produce only tiny variations from chance may be too large for practical testing.

I think Geni's point goes to the crux of the question. But I also think it is moot before somebody actually appliies to take the challenge with a claim that would require a very long time to test if JREF stood by its requirement for a very high confidence factor.

My own cut at this is that very long trials looking for very small effects is something that JREF is a little uncomfortable with but none the less I think that if JREF received a clearly worded claim of a small supernatural effect and a clearly worded protocol that described how to test for the effect that JREF would accept the claim for testing.

So in a nutshell I think davidsmith73 is wrong to some degree but not quite about the same thing that you think he is wrong about.

 

[CFLarsen]

Nope. Not taking the bait.

If I open a thread, will you answer it there?

If Randi and many people accociated with the JREF have said that they do not regard any psi experiment to date as offering any evidence of psi, then naturally they are going to demand better results for any potential challenge involving a psi experimental design.

I didn't ask what you thought they will do. I asked how JREF demands - as in real-life cases - "better" results.

Do you have real-life cases, or are you merely speculating?

Is the link not working for you or something? Anyway, this bit from it:

"Viewing the entire tape, we see that the dog responded to every car that drove by, and to every person who walked by." This is simply not true, and Randi now admits that he has never seen the tape.

I'm inclined to believe Sheldrake on this one. If someone printed remarks about me saying I had lied when I had not, I would not let the issue go. Randi has not made any statements saying Sheldrake is wrong about these events as far as I can see.

Did you check out Gord__in_Toronto's post?

 

[davidsmith73]

If I open a thread, will you answer it there?

Ok

I didn't ask what you thought they will do. I asked how JREF demands - as in real-life cases - "better" results.

Do you have real-life cases, or are you merely speculating?

Speculating. And this is the touble. Randi has stated that he can't get any parapsychologist to stand up and be counted. But we are not given any information regarding the details of the JREF terms of any challenge, if there are any. Would the JREF accept an exact replication of a psi experiment for the challenge? Would they insist upon a high hit rate or probability? I can't see this information anywhere, but we need to know it in order to judge whether Randi's comments are fair.

I'm still wating for Randi to clarify what he meant by "standard success" in our email exchange, so I may have a real-life case soon.

Did you check out Gord__in_Toronto's post?

No. What's this about?

 

[CFLarsen]

Ok

Knock yourself out.

Speculating. And this is the touble.

Speculating. You are criticizing JREF from pure speculation. That is the trouble, alright.

Randi has stated that he can't get any parapsychologist to stand up and be counted. But we are not given any information regarding the details of the JREF terms of any challenge, if there are any.

That depends entirely on the claim. If you want to see the details, they are available at JREF. But it isn't as if we are denied the information.

It's hardly Randi's fault if parapsychologists refuse to take the Challenge, is it? Remember, the standards are lower than in the scientific world: Of course, the test itself is scientific, but there is no requirement for peer-review or replication by independent scientists. All they got to do, is do it once.

But they don't.

Would the JREF accept an exact replication of a psi experiment for the challenge? Would they insist upon a high hit rate or probability? I can't see this information anywhere, but we need to know it in order to judge whether Randi's comments are fair.

I'm still wating for Randi to clarify what he meant by "standard success" in our email exchange, so I may have a real-life case soon.

Tell us what he said.

No. What's this about?

Randi's version. Think you might do a bit of research, before you speculate?

 

[Zep]

Just a comment on the subject broadly.

The JREF Challenge is based on the very simple principle of "please show conclusively that you can do what you claim". Now let us look at that more closely.

"Please show...that you can do" means that the claimant must do the showing, and not that JREF must try to repeat or to disprove their claim.

"show conclusively" means that the result has to be clear and unequivocal to any reasonable person, using a process that requires no judgement. This removes all personalities from the decision process - Randi, skeptics, or anyone.

"do what you claim" means the claimant is the person who sets the bar of attainment, not JREF. If the claimant says they can dowse gold with 90% accuracy then that's the bar to be got over when they take the challenge under controlled conditions. Incidentally, from what I have seen over the years, just about every dowser who has ever applied has claimed 80% to 100% accuracy...before the testing...

Of course, it's clear that some claims start to get VERY watered down the moment they come up against reality. I mean, I can pick heads-or-tails 50% of the time in any series of coin-flips just by chance alone. Sometimes I may even get runs of guesses correct, but as any statistician knows, this, too, is predictable by chance as well. So if that's my claim then it's a piss-weak one, and hardly likely to get past square one with JREF as "extraordinary".

The problem for most of the "psi" claims nows seems to be that the claimed "effect" is only very occasionally just above random chance, and often not even that. These incredible remote-viewing metal-bending time-warping effects somehow seem to be just hidden away in the mush of random noise of experimentation, always elusive and mercurial, just out of the grasp of investigators. Certainly just far enough out of their grasp to require ongoing pourings of buckets of grant money and ongoing university tenures in the certain hope that it will indeed be isolated "REAL SOON NOW".

The point being, davidsmith73, that it is well nigh impossible to set a "standard test method" for claimants in advance an examination of their claims and their expected levels of success.


Oh, and incidentally, for those who are not aware of this yet: Victor Zammit is off-his-trolley, barking, froot-loop bonkers. In the nicest eccentric way.

 

[steenkh]

Well I have to make a retraction. I emailed Randi asking whether the JREF would accept an exact replication of Sheldrakes email telepathy experiment for the challenge.

He replied "yes, with a standard rate of success, nothing more"

Are you sure he was talking about an exact replication of Sheldrakes experiment? As I have understood, the controls were extremely lax, to the point of letting the test persons decide on who called them even after having lifted the receiver! I cannot imagine that Randi would accept the exact same experiment without stricter controls.

 

[bjb]

Reading through James Randi's book, "Flim-Flam!", there are many examples of the tests he has conducted over the decade. In general, he lets the claimant set the standard for success, and most of them say their success rate is in the 90-100% range. However, the many tests show them achieving no better than random chance would predict, far below this 90-100% value. Since the claimants routinely fail so spectacularly, no detailed statistical analysis is necessary to show the claimiant has failed.

One problem I have with Sheldrake and the other parapsychologists is that they tell their stories as if no one has ever been tested by James Randi. The fact is the many hundreds of people who have taken Randi's challenge have agreed to his conditions, so why are the parapsychologists like Sheldrake so unwilling to be tested? Instead, they hide behind statistics and imaginary test protocols that they say are unacceptable. How do they know Randi's test is unfair when they won't even bother to apply?

Randi says pretty much the same thing when he points out that he can't get any parapsychologists to apply for the challenge. But davesmith73 declares Randi's statements to be 'snide rhetoric'. That is untrue because Randi is making a statement of fact. Parapsychologists won't take his challenge, and that brings me to my next point.

In my opinion, parapsychologists like Sheldrake are basically dishonest in their beliefs. Others like the dowsers and many psychics honestly believe in their abilities and are perfectly willing to be tested by Randi. They have no problem agreeing to Randi's conditions and controls so I believe it is because they honestly believe they have paranormal abilities. They alays fail the tests, which shows they're deluded, but at least they honestly believe in themselves. However, the parapsychologists don't have guts like the dowsers do. Instead of responding to the challenge, they come up with all sorts of b.s. as to why the test isn't fair, or why Randi himself is dishonest. Their statements about him are based on their own speculation, not direct evidence. If they really want to prove the challenge is bogus, they should take the challenge and document all of Randi's misconduct.

 

[T'ai Chi]

, so why are the parapsychologists like Sheldrake so unwilling to be tested? Instead, they hide behind statistics and imaginary test protocols that they say are unacceptable.

A person deeply emotionally immersed in the organized skeptical movement, such as yourself?, might read something sinister or telling in their refusal.

People with their heads out of... the water, simply see it as the parapsychologists, one the whole, might be more interested in persuing the standard channels of science, not a challenge from an organization in the skeptical movement, with someone known to be hostile to such investigation.

 

[drkitten]

People with their heads out of... the water, simply see it as the parapsychologists, one the whole, might be more interested in persuing the standard channels of science,

... except they don't do that, either.

When was the last time a parapsychologist made Brain? Or Nature? Or JAMA?

 

[steenkh]

When I was at the World Skeptic Congress a couple of years ago, there were actually a couple of brave parapsychologists who stood up and defended their cause. Apparently, the meagre results have forced the few remaining parapsychologists to develop methods of distinguishing the real parapsychologically gifted people from the psychologically challenged, and until now, they had only found people of the latter category.

 

[CFLarsen]

When I was at the World Skeptic Congress a couple of years ago, there were actually a couple of brave parapsychologists who stood up and defended their cause. Apparently, the meagre results have forced the few remaining parapsychologists to develop methods of distinguishing the real parapsychologically gifted people from the psychologically challenged, and until now, they had only found people of the latter category.

Less than two years, mate....

Fifth World Skeptics Congress
Abano Terme, Italy, October 8-10, 2004