Information on Protocol for Pavel Ziborov Applicant - Page 4

Peristarkawan · 13th August 2008, 10:30 AM

Originally Posted by pavel_do

I cant claim 29 out of 40.. cause that not beating 1.000 odds..

Just to make sure you understand this, it is possible to claim a success rate lower than 75% and still have 1:1000 odds. A lower success rate would just require more trials to prove. For example, as I noted above, the odds of getting 60 or more correct out of 90 are also about 1:1000. The real problem with that is that it would be a much longer test.

Startz · 13th August 2008, 10:45 AM

Originally Posted by Peristarkawan

Just to make sure you understand this, it is possible to claim a success rate lower than 75% and still have 1:1000 odds. A lower success rate would just require more trials to prove. For example, as I noted above, the odds of getting 60 or more correct out of 90 are also about 1:1000. The real problem with that is that it would be a much longer test.

Pavel understands this. He originally proposed more trials, but JREF felt this was problematic. The current design is intended to accommodate concerns raised by JREF.

Rodney · 13th August 2008, 04:16 PM

Originally Posted by Peristarkawan

And what if Pavel gets only 10 correct out of 40? The odds of that are the same as the odds of getting 30 correct. Do you claim Pavel should also be retested in that case?

Now that you mention it, yes.

Rodney · 13th August 2008, 04:25 PM

Originally Posted by Startz

Pavel understands this. He originally proposed more trials, but JREF felt this was problematic. The current design is intended to accommodate concerns raised by JREF.

Which is why he definitely should be retested if he gets 29 -- or even 27 or 28 -- out of 40. If the JREF is serious about the paranormal, it can't ignore unusual results that fall just below an arbitrary standard that it establishes.

RemieV · 13th August 2008, 05:27 PM

Originally Posted by Rodney

Which is why he definitely should be retested if he gets 29 -- or even 27 or 28 -- out of 40. If the JREF is serious about the paranormal, it can't ignore unusual results that fall just below an arbitrary standard that it establishes.

Please stay on topic, particularly in this thread. The topic of the thread is the protocol for Pavel Ziborov's claim. Potential future testing has no place here.

Gr8wight · 13th August 2008, 06:11 PM

Originally Posted by Rodney

Which is why he definitely should be retested if he gets 29 -- or even 27 or 28 -- out of 40. If the JREF is serious about the paranormal, it can't ignore unusual results that fall just below an arbitrary standard that it establishes.

Oh, Rodney, we've all had this discussion before. Fifteen or twenty times, if I recall correctly. The JREF is not serious about the paranormal. They are serious about challenging those people who make claims about the paranormal to demonstrate their abilities. That's all we're asking Pavel, or any other applicant, to do. Demonstrate the exact ability you claim to have. If Pavel claims to be able to correctly identify 30 of 40 photographs, then 29 does not meet his own conditions.

Of course, until Pavel undertakes his test, your position on this is moot. What happens if he only gets 26 correct? Or 25? Or 19? Where do you draw the line? Or do you not recognise why it is necessary to draw a line?

Do you?

Rodney · 13th August 2008, 06:28 PM

Originally Posted by Gr8wight

Oh, Rodney, we've all had this discussion before. Fifteen or twenty times, if I recall correctly. The JREF is not serious about the paranormal. They are serious about challenging those people who make claims about the paranormal to demonstrate their abilities. That's all we're asking Pavel, or any other applicant, to do. Demonstrate the exact ability you claim to have. If Pavel claims to be able to correctly identify 30 of 40 photographs, then 29 does not meet his own conditions.

Of course, until Pavel undertakes his test, your position on this is moot. What happens if he only gets 26 correct? Or 25? Or 19? Where do you draw the line? Or do you not recognise why it is necessary to draw a line?

Do you?

Let's continue this discussion in the existing thread "Odds Standard for Preliminary Test."

GzuzKryzt · 14th August 2008, 03:15 PM

Originally Posted by Rodney

Let's continue this discussion in the existing thread "Odds Standard for Preliminary Test."

In said thread, Pavel made the following post:

Originally Posted by pavel_do

He is right, though I think I will be able to perform 30 out of 40, will do my best for that. If I was sure I cant do that I would say I am not going for it… I was the one who asked for 40 pairs.. At least..
But the conditions is not as good for me as the other tests that would give me enough time, less stress in order to perform positive results.
I know some one will say now.. you see.. he prepared already an excuse in case he fails..
THERE WILL BE NO EXCUSES… no one forced me to apply and take test.. It is my own decision..
Just that I want test to happen.. And if I will not step out of my way and test conditions that would make me really comfortable etc.. that the test will never happen.. cause it take too many hours.. cause I have advantage, cause it is too many photos etc.. CAUSE IT IS POSSIBLE FOR ME TO PASS IT WITH THOSE CONDITONS..
I stepped way to far just to move some how the negotiations and make test possible..
I would be more comfortable with 2-3 trials ( let say by 30 pairs in each) where achieving positive results in any would mean I passed it by covering any way 1.000 odds that needed to pass preliminary test.. I am sure during all 2-3 trial I would perform evident results that I am real deal and able to see and perform what I am claiming to! Having 2-3 trials also reduce a stress level that is the main issue for me in order to be as calm as I can during a test… Every one would feel calmer and more comfortable knowing that he can perform 2-3 trial.. But not 1.. I have explained that many times.
Same with more pairs.. that is still covers 1.000 that is way far from getting by a chance and I see no problem in having a few hour test with allot of pairs that I am the one who pays for any way..
Etc.. many reasons that I have explained before many times that is just ignored, and I am in front of the fact that it will be not allowed full stop.
Like passing it make them to give me the money! that is just a preliminary test as we all know.
Just that passing it will mean that is possible to pass the second one and get the money.

I will be honest with you. If the JREF will not let me to have results revealed immediately after each pair is named, I am not taking this test. Cause that would mean I have agreed to everything that will make test almost impossible to pass and I have to rely in chance to pass it..
Ill just drop it, and will be looking for other ways of testing ( maybe JREF offer something reasonable too). There is a time.. my claim valid for 1 year if I am not mistaken.
I really want test to happen and doing my best to make it possible. Cause I know I will be able to show that such things is real and can exist and has to be studied etc.. There is a fake psychics.. but that dose not mean there is no real ones!

I highlighted one relevant part.

pavel_do · 14th August 2008, 04:49 PM

Originally Posted by GzuzKryzt

In said thread, Pavel made the following post:

I highlighted one relevant part.

I am sure most of people read it there, same as JREF..

I see no reasons to copy and post it here too.. Or... Oh yes.. cause I said if JREF will not allow the envelopes to be opened immediately.. (that is not affect odds in any ways), so I am not taking this test.
And I am not taking this test with pairs, as I agreed to all that they wanted so far and that the last thing that would help me to show good results… and I am not getting off that.. If it cant not be agreed, that we have to find the other test to test me and I will do my best for that as I am doing any ways. So what the problem? I am being honest , as always but..

Every body, have a nice weekend.
Regards,
Pavel

IXP · 14th August 2008, 04:52 PM

Originally Posted by GzuzKryzt

I highlighted one relevant part.

What is your point? Pavel has always maintained that he foresees what will happen when the envelope is opened and this is why he needs each envelope to be opened immediately after the guess, so it is not confuse it with others.

IXP

pavel_do · 14th August 2008, 05:14 PM

Originally Posted by IXP

What is your point? Pavel has always maintained that he foresees what will happen when the envelope is opened and this is why he needs each envelope to be opened immediately after the guess, so it is not confuse it with others.

IXP

His point is obvious.. to twist it and make look like I have no intentions to have any test and sort of wasting my time and the others etc..

I am not even going to argue it.. it is pointless. I am just sick of it to be honnest. When some one don’t want to see (admit) the truth, he will not see or admit it.. and the truth is that I am not wasting anyone’s time, and I really want to have proper good test and show that what I am claiming I am able to perform. And I will do my best to make it happen. I just cant as a lame sheep, agree to all the conditions that would put me in the position to lose the test, but not being able to perform it is best…
Any way I am sure who want to see it he sees it..

Or that was another joke? Or the way to encourage me? I am sorry but maybe due to a cultural difference and some others.. I don’t take it as a joke, and obviously not taking it as an encouraging…

IXP · 14th August 2008, 05:45 PM

I am with you Pavel.

I think you have been quite reasonable in trying to determine exactly what you can do and under what circumstances. You are one of the few who has not changed your story or come up with new demands each time JREF proposes something. I hope you get a test that you believe you can pass.

IXP

p.s. This does not mean that I think you will pass!

Startz · 14th August 2008, 05:50 PM

Originally Posted by IXP

I am with you Pavel.

I think you have been quite reasonable in trying to determine exactly what you can do and under what circumstances. You are one of the few who has not changed your story or come up with new demands each time JREF proposes something. I hope you get a test that you believe you can pass.

IXP

p.s. This does not mean that I think you will pass!

I fully agree with IXP's position.

Akhenaten · 14th August 2008, 10:46 PM

Hi Pavel.

Welcome to город skeptiks

It was always going to be tough for you to deal with the rigourous examination of your ability here. It's just the nature of the place that it will sometimes become quite adversarial, and you seem to have done a brilliant job so far of not taking it personally, but rather in the spirit of transparency in which most criticisms and suggestions have been made.

Please believe that nobody here will attempt to twist your words with intent to discredit you. That's just not what the folks here do. Well, mostly

But people will insist, quite rightly, that all the cards are on the table, so to speak.

Your honesty has set you apart from the vast majority of previous claimants, and I'm sure this will continue until a mutually agreeable test is designed.

I understand your need to see the envelopes opened as you make your choices, and I also understand the objections to this, but sadly I don't know nearly enough to help with a solution. I just wanted to add a voice of support for the way you've gone about this, regardless of the outcome.

Good Luck Mate,

Dave

Ravenwood · 14th August 2008, 11:10 PM

I can't believe I missed this! Pavel, you said earier that you could not use the blank/white photos because you confused it sometimes with the cross picture. What if you used only blank/white photos & black/exposed photos as your "targets" in your test? if you had one black & one white in the pair, perhaps you would "see" one or the other more distinctively. If that is the case, then you could do say sets of 5 envelopes, but only one contains the black or white photo(whichever you can feel strongest about), the other 4 contain the opposite color photo. That would be a 1 in five chance, or 20% as opposed to the 50% chance per pair, meaning less runs. I would be really interested in seeing what your success with that protocol is, I'm sure one of our number crunchers will come up with how many runs you would have to do to meet the odds, assuming that you are sticking to the 75% accuracy claim...

Peristarkawan · 15th August 2008, 10:39 AM

Originally Posted by Ravenwood

I can't believe I missed this! Pavel, you said earier that you could not use the blank/white photos because you confused it sometimes with the cross picture. What if you used only blank/white photos & black/exposed photos as your "targets" in your test? if you had one black & one white in the pair, perhaps you would "see" one or the other more distinctively. If that is the case, then you could do say sets of 5 envelopes, but only one contains the black or white photo(whichever you can feel strongest about), the other 4 contain the opposite color photo. That would be a 1 in five chance, or 20% as opposed to the 50% chance per pair, meaning less runs. I would be really interested in seeing what your success with that protocol is, I'm sure one of our number crunchers will come up with how many runs you would have to do to meet the odds, assuming that you are sticking to the 75% accuracy claim...

The odds of getting 7 correct out of 10 would be 0.00086, about 1:1157.
At a 60% accuracy rate, the odds of getting 9 correct out of 15 would be 0.00078, about 1:1274.

Startz · 15th August 2008, 12:23 PM

There are a large number of statistical designs that can be used for these tests, as well as a large number of experimental procedures. Forumites have made a number of helpful suggestions along both lines that Pavel has used in coming up with a protocol.

As of right now, there is a specific proposed protocol that was emailed to JREF just over a week ago. It was also posted here in the forum. While waiting for JREF to respond, it would be most helpful for folks here to comment on the current draft protocol.

Remember that:

(1) Pavel's claimed ability is to "see" the content of envelopes that will be opened after his identification. He "sees" the envelopes being opened. He does not claim to see into the sealed envelopes. This is a claim of clairvoyance, as opposed to "x-ray vision."

(2) Pavel has always said his that his ability is imperfect.

(3) JREF wants to hold down the number of envelopes involved, and has expressed some preference for the test to be done in pairs of envelopes.

pavel_do · 15th August 2008, 12:42 PM

Thanks Startz, I wouldn't answer better!

petre · 15th August 2008, 03:10 PM

Originally Posted by Startz

There are a large number of statistical designs that can be used for these tests, as well as a large number of experimental procedures. Forumites have made a number of helpful suggestions along both lines that Pavel has used in coming up with a protocol.

As of right now, there is a specific proposed protocol that was emailed to JREF just over a week ago. It was also posted here in the forum. While waiting for JREF to respond, it would be most helpful for folks here to comment on the current draft protocol.

Remember that:

(1) Pavel's claimed ability is to "see" the content of envelopes that will be opened after his identification. He "sees" the envelopes being opened. He does not claim to see into the sealed envelopes. This is a claim of clairvoyance, as opposed to "x-ray vision."

(2) Pavel has always said his that his ability is imperfect.

(3) JREF wants to hold down the number of envelopes involved, and has expressed some preference for the test to be done in pairs of envelopes.

Thanks for the summary! I might have missed it in my skimming, or it might have gotten a bit obfuscated in translation, but it wasn't until this thread that I realized the claimed ability was clairvoyance rather than some flavor of remote viewing. That makes it greatly more clear why the post-selection reveal might impact performance.

The following have probably already been suggested but...
It would seem likely that with an ability like this, testing several differentation classes would be much easier (one would be less likely to confuse the results of multiple trials). For example, do one trial with six envelopes containing pictures of zenner cards. Pavel points the one out from a list that will be chosen, then a die is rolled and that number envelope opened and the item shown to Pavel. A second trial is then done with pictures of six different colors of full-page rectangles. Another trial might use pictures of numerical digits, etc.

Perhaps he has more difficulty picking out one item from among a similar class of items though, so his own testing will have to guide the protocol. I was just hoping to offer a helpful suggestion based on my recent new understanding of the described ability.

pavel_do · 17th August 2008, 09:28 AM

Originally Posted by petre

Thanks for the summary! I might have missed it in my skimming, or it might have gotten a bit obfuscated in translation, but it wasn't until this thread that I realized the claimed ability was clairvoyance rather than some flavor of remote viewing. That makes it greatly more clear why the post-selection reveal might impact performance.

The following have probably already been suggested but...
It would seem likely that with an ability like this, testing several differentation classes would be much easier (one would be less likely to confuse the results of multiple trials). For example, do one trial with six envelopes containing pictures of zenner cards. Pavel points the one out from a list that will be chosen, then a die is rolled and that number envelope opened and the item shown to Pavel. A second trial is then done with pictures of six different colors of full-page rectangles. Another trial might use pictures of numerical digits, etc.

Perhaps he has more difficulty picking out one item from among a similar class of items though, so his own testing will have to guide the protocol. I was just hoping to offer a helpful suggestion based on my recent new understanding of the described ability.

Thank you, good advise.. But I think for JREF such test will be too complicated as it will again involve allot of different pictures etc... and they are not research team that interested in researching and testing.. it is a challenge..

pavel_do · 17th August 2008, 01:55 PM

I know it is out of topic.. but still would like to post it..
There is a web site offering a phone readings to there clients.. and there is a top pay “psychic” that charges 20$ per MINUTE… I am shocked.. Nevertheless.. here is one of the comments.. from the client.. that made me laugh so I would like to share it with you.

“He started by saying things were really good for me & he was going to tell me why they were. He repeated that statement 3 times and then it sounded like he turned on the water and stepped into the shower--while still trying to talk to me. When I told him I could no longer hear him he said he would speak louder. Louder wasn't the problem--he sounded all garbled like he was drowning. Total waste of $$$$. If you don't have the time to do a proper reading don't be available for calls.”

Moochie · 17th August 2008, 03:17 PM

Originally Posted by pavel_do

I know it is out of topic.. but still would like to post it..
There is a web site offering a phone readings to there clients.. and there is a top pay “psychic” that charges 20$ per MINUTE… I am shocked.. Nevertheless.. here is one of the comments.. from the client.. that made me laugh so I would like to share it with you.

“He started by saying things were really good for me & he was going to tell me why they were. He repeated that statement 3 times and then it sounded like he turned on the water and stepped into the shower--while still trying to talk to me. When I told him I could no longer hear him he said he would speak louder. Louder wasn't the problem--he sounded all garbled like he was drowning. Total waste of $$$$. If you don't have the time to do a proper reading don't be available for calls.”

Good one

Reminds me of those sex lines which are staffed by all kinds of people -- grannies, guys with feminine sounding voices, and so on. I guess it's a living.

M.

RemieV · 19th August 2008, 07:40 PM

At this point, I'm going to need to have some discussions with other JREF staff members.

They are trickling back from the Galapagos trip. I'll update soon.

Startz · 19th August 2008, 09:55 PM

Originally Posted by RemieV

At this point, I'm going to need to have some discussions with other JREF staff members.

They are trickling back from the Galapagos trip. I'll update soon.

Remie,

Thanks for keeping us all in the loop.

Myriad · 21st August 2008, 03:13 PM

The following would appear to address most if not all of the various issues and objections:

There are 27 trials.

There are 3 opaquely lined envelopes per trial.

There is 1 photograph with an image and 2 blank photographs in each trial's set of envelopes.

Pavel chooses one envelope per trial after examining all three as he desires (if necessary, within a pre-set time limit).

A successful trial (hit) is choosing the photograph with an image.

The result of each trial is revealed immediately after each trial.

The image(s) can be the same for all the trials, or different for each trial, whatever Pavel thinks would be most helpful. He may be shown a copy of the trial's image or told in words what the image depicts at the start of each trial if that would be helpful.

18 hits in the 27 trials (2/3 or better correct) is a successful test. The probability of success by chance is well under 1 in 1000 (.000407). The probability of 17 hits or better by chance is .00154.

The success rate needed for success corresponds to getting 27 or more hits out of 40, something that appears to be more comfortably within Pavel's claimed success rate than 30 out of 40.

Note that the total number of envelopes involved is only one more than in the 40-trials protocol currently under consideration.

(If I'm interpreting Pavel's comments correctly, he suggests failures occur when he is unable to perceive the future opening of an envelope showing a target picture when the envelope does in fact contain the target picture -- in other words, false negatives. If that's the case, then the presence of the two blanks instead of one in each trial set should not decrease the probability of success per trial, given that the examination time he would have available per envelope would be the same.)

Respectfully,
Myriad

Startz · 21st August 2008, 03:42 PM

Originally Posted by Myriad

The following would appear to address most if not all of the various issues and objections:

There are 27 trials.

There are 3 opaquely lined envelopes per trial.

There is 1 photograph with an image and 2 blank photographs in each trial's set of envelopes.

Pavel chooses one envelope per trial after examining all three as he desires (if necessary, within a pre-set time limit).

A successful trial (hit) is choosing the photograph with an image.

>snip
Respectfully,
Myriad

I rather like this. It greatly reduces the probability of a false negative, down to about 40 percent. I think this is better than the revised protocol that Pavel (with a minor assist from me) submitted.

Pavel:

If you like this, perhaps you should send it in to Remie.
-Dick

Rodney · 21st August 2008, 05:51 PM

Originally Posted by Myriad

The probability of 17 hits or better by chance is .00154.

So, under your proposal, would 17 out of 27 be a failure or would the preliminary test be extended?

Gr8wight · 21st August 2008, 07:04 PM

Originally Posted by Rodney

So, under your proposal, would 17 out of 27 be a failure or would the preliminary test be extended?

The applicant may submit the protocol design. The JREF will decide if 17 out of 27, or 18 out of 27, or some other result will satisfy their statistical requirements. In any case, there are two possible outcomes: a) the applicant is successful, or: b) the applicant is unsuccessful. Giving the applicant a do-over for being kinda close is not provided for in the rules.

But why am I wasting my time writing this all out, Rodney? You already know this. You're just being deliberately obtuse for trolling purposes.

Myriad · 21st August 2008, 07:23 PM

Originally Posted by Rodney

So, under your proposal, would 17 out of 27 be a failure or would the preliminary test be extended?

Please note that my proposal is a suggestion from a third party, not an offer from JREF.

Pavel (with Startz's continued help) must negotiate these things. I would suggest -- still as a bystander -- that he choose ONE of the following possiblities, whichever he feels is most important to him, to negotiate for:

1. Explicit provision for a guaranteed extension of the test, within a set range of future dates, should a certain number of hits short of the requirement be made.

2. Allowance for a chance probability value for success somewhat over .001, such as allowing 17 of 27 to be counted as a success.

3. Slightly increased number of trials. With the trials as I suggested them, if there were 30 trials instead of 27 (that's ten more envelopes total than the 40-trial protocol, but fewer steps for the testers to go through during the test than the 40-trial protocol), 19 hits would be sufficient for p < .001 (.000737). That's essentially three more chances to make one more hit. (That seems surprising, but it's how the math works out.) Overall 19 out of 30 is approximately the same hit expectation per trial as 25 out of 40.

Respectfully,
Myriad

pavel_do · 21st August 2008, 08:12 PM

Originally Posted by Myriad

The following would appear to address most if not all of the various issues and objections:

There are 27 trials.

There are 3 opaquely lined envelopes per trial.

There is 1 photograph with an image and 2 blank photographs in each trial's set of envelopes.

Pavel chooses one envelope per trial after examining all three as he desires (if necessary, within a pre-set time limit).

A successful trial (hit) is choosing the photograph with an image.

The result of each trial is revealed immediately after each trial.

The image(s) can be the same for all the trials, or different for each trial, whatever Pavel thinks would be most helpful. He may be shown a copy of the trial's image or told in words what the image depicts at the start of each trial if that would be helpful.

18 hits in the 27 trials (2/3 or better correct) is a successful test. The probability of success by chance is well under 1 in 1000 (.000407). The probability of 17 hits or better by chance is .00154.

The success rate needed for success corresponds to getting 27 or more hits out of 40, something that appears to be more comfortably within Pavel's claimed success rate than 30 out of 40.

Note that the total number of envelopes involved is only one more than in the 40-trials protocol currently under consideration.

(If I'm interpreting Pavel's comments correctly, he suggests failures occur when he is unable to perceive the future opening of an envelope showing a target picture when the envelope does in fact contain the target picture -- in other words, false negatives. If that's the case, then the presence of the two blanks instead of one in each trial set should not decrease the probability of success per trial, given that the examination time he would have available per envelope would be the same.)

Respectfully,
Myriad

Thank you very much!
Sounds good.. as even the results would be more evident than 50/50 chance in pairs... For that reason before I was offering also 1 out of 3.. and even 2 out of 5.. I can try this one tomorrow or Saturday.. and It still might be better for me if there will be 3 photos.. not 2 blank.. But sound “easy” and I done things like this before.. just that there was 3 photos.. not 2 blank and 1 photo.. But I think it does not matter at all for the odds..

So we could prepare with Startz a new proposal for JREF and see what they would say..

Jackalgirl · 21st August 2008, 08:31 PM

Hi, Pavel --

Play around with different combinations of regular photos and blank photos and see what works best. In the end, it doesn't really matter, as long as there's some kind of target image (or non-image) that you can identify out of two other images or non-images (which need not necessarily be the same as each other).

I love the idea of three photos. This thread has inspired me to take a stats & probabilities class this next term. What a cool collaborative effort!

Startz · 21st August 2008, 10:17 PM

Originally Posted by pavel_do

Thank you very much!
Sounds good.. as even the results would be more evident than 50/50 chance in pairs... For that reason before I was offering also 1 out of 3.. and even 2 out of 5.. I can try this one tomorrow or Saturday.. and It still might be better for me if there will be 3 photos.. not 2 blank.. But sound “easy” and I done things like this before.. just that there was 3 photos.. not 2 blank and 1 photo.. But I think it does not matter at all for the odds..

So we could prepare with Startz a new proposal for JREF and see what they would say..

Pavel:

This sounds good. I think that to keep the negotiations on track you should put some weight on keeping things simple. You might do well to stick exactly to Myriad's suggestion, perhaps using 3 photos if that is really important.

Startz · 21st August 2008, 10:19 PM

Originally Posted by Jackalgirl

This thread has inspired me to take a stats & probabilities class this next term. What a cool collaborative effort!

Proving the success of the E in JREF!

gibsonbaud · 24th August 2008, 06:38 PM

Im not sure if I am the only person who noticied this or if it is something that has been covered before, or if im wong on this, but the statistics most of you have been posting seem faulty...

10 sets of two envelopes: probability of getting them in two distinct piles... 50%. Every time you start a new pair the probability of getting it right goes back to one out of two... (i.e. 1 out of 2, 2, out of 4, 3 out of 6), it's a constant 50% probability. It would be different if a stack of 20 was handed to them and they picked the correct one of 20 each time, but 1 of 2 each time stays at 50%.

Same goes for the three envelopes at a time method only its a 33.3% chance of getting them correct.

Each time a new one is done the number correct or incorrect goes up, but so does the total number of envelopes taken in to consideration.

EDIT:

PS: I would suggest making it so the envelopes cannot be picked up, ie, they are taped to the table or are fastened down, allowing them to be tuched but not held.

Gr8wight · 24th August 2008, 08:29 PM

Originally Posted by gibsonbaud

Im not sure if I am the only person who noticied this or if it is something that has been covered before, or if im wong on this, but the statistics most of you have been posting seem faulty...

10 sets of two envelopes: probability of getting them in two distinct piles... 50%. Every time you start a new pair the probability of getting it right goes back to one out of two... (i.e. 1 out of 2, 2, out of 4, 3 out of 6), it's a constant 50% probability. It would be different if a stack of 20 was handed to them and they picked the correct one of 20 each time, but 1 of 2 each time stays at 50%.

Same goes for the three envelopes at a time method only its a 33.3% chance of getting them correct.

Each time a new one is done the number correct or incorrect goes up, but so does the total number of envelopes taken in to consideration.

Yes, but...

My chances of success in the first pair is 50%. My chances of success in the second pair is 50%. However, my chances of getting both right are only 25%.

gibsonbaud · 24th August 2008, 08:39 PM

Not true, you have a 50% chance on envelopes 1 and 2, and a 50% chance on envelopes 3 and 4. No matter how many times you do it, if you are choosing between a correct and an incorrect it will always be 50%.

You could do it 500 times. The chance of getting them all correct is always 50%. It would be different if you had 4 cards, chose a correct one from the 4, then chose a correct one from the remaining 3... Then the overall probability would change, but as long as the total number of envelopes increases at a rate of 2 for each one attempt, the probability does not change.

Like I said... I could be wrong... Just how it appears to me...

Startz · 24th August 2008, 09:18 PM

Originally Posted by gibsonbaud

Not true, you have a 50% chance on envelopes 1 and 2, and a 50% chance on envelopes 3 and 4. No matter how many times you do it, if you are choosing between a correct and an incorrect it will always be 50%.

You could do it 500 times. The chance of getting them all correct is always 50%. It would be different if you had 4 cards, chose a correct one from the 4, then chose a correct one from the remaining 3... Then the overall probability would change, but as long as the total number of envelopes increases at a rate of 2 for each one attempt, the probability does not change.

Like I said... I could be wrong... Just how it appears to me...

What Gr8wight said is right. This very easy for you to check. Take a coin, guess heads or tails. Do it again. If you're right both times, score it as a success. Now run this two-toss experiment a few dozen times. You'll see that you get both tosses right about one time in four.

More formally, the joint probability of a sequence of independent events is the product of the individual probabilities.

gibsonbaud · 24th August 2008, 09:56 PM

ahh, im an idiot... Guss I've been out of school too long...lol.

EHocking · 25th August 2008, 08:45 AM

Originally Posted by gibsonbaud

ahh, im an idiot... Guss I've been out of school too long...lol.

I've posted this a number of times, but do find it a handy reference for checking the odds of random chance for simple tests such as being proposed for Pavel's challenge.
http://www.automeasure.com/chance.html

Jeff Wagg · 27th August 2008, 03:54 PM

Folks, I'm going to step in here.

This protocol has been going on forever. It seems like Pavel keeps trying things and then failing, and people are trying to modify the test so that he has a chance of winning. This is NOT what the challenge is about. The claimant should have a clear claim and confidence that he can accomplish what he claims before he applies.

RemieV recently went over a protocol with me that looked like it might be workable, and then I see here that ANOTHER one has been proposed and Pavel likes it. Now we're essentially back to square one.

It is now incumbent upon Pavel, who I admit seems sincere, to state what he can do and allow us to test it. It seems that he applied way before he was ready, and now the JREF is expending too many resources to try to make something work.

There are other applicants waiting.

The JREF will accept one more full protocol from Pavel (with Startz's kind help, if he's willing) and then we will move on to another candidate.

13th August 2008, 10:30 AM	#121
Peristarkawan New Blood Join Date: Aug 2008 Posts: 7	Originally Posted by pavel_do I cant claim 29 out of 40.. cause that not beating 1.000 odds.. Just to make sure you understand this, it is possible to claim a success rate lower than 75% and still have 1:1000 odds. A lower success rate would just require more trials to prove. For example, as I noted above, the odds of getting 60 or more correct out of 90 are also about 1:1000. The real problem with that is that it would be a much longer test.
Peristarkawan New Blood Join Date: Aug 2008 Posts: 7

13th August 2008, 04:16 PM	#123
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942	Originally Posted by Peristarkawan And what if Pavel gets only 10 correct out of 40? The odds of that are the same as the odds of getting 30 correct. Do you claim Pavel should also be retested in that case? Now that you mention it, yes.
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942

13th August 2008, 04:25 PM	#124
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942	Originally Posted by Startz Pavel understands this. He originally proposed more trials, but JREF felt this was problematic. The current design is intended to accommodate concerns raised by JREF. Which is why he definitely should be retested if he gets 29 -- or even 27 or 28 -- out of 40. If the JREF is serious about the paranormal, it can't ignore unusual results that fall just below an arbitrary standard that it establishes.
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942

13th August 2008, 05:27 PM	#125
RemieV Philosopher Join Date: Jul 2006 Posts: 5,292	Originally Posted by Rodney Which is why he definitely should be retested if he gets 29 -- or even 27 or 28 -- out of 40. If the JREF is serious about the paranormal, it can't ignore unusual results that fall just below an arbitrary standard that it establishes. Please stay on topic, particularly in this thread. The topic of the thread is the protocol for Pavel Ziborov's claim. Potential future testing has no place here.
RemieV Philosopher Join Date: Jul 2006 Posts: 5,292

13th August 2008, 06:11 PM	#126
Gr8wight red-shirted crewman Join Date: Jul 2004 Posts: 1,661	Originally Posted by Rodney Which is why he definitely should be retested if he gets 29 -- or even 27 or 28 -- out of 40. If the JREF is serious about the paranormal, it can't ignore unusual results that fall just below an arbitrary standard that it establishes. Oh, Rodney, we've all had this discussion before. Fifteen or twenty times, if I recall correctly. The JREF is not serious about the paranormal. They are serious about challenging those people who make claims about the paranormal to demonstrate their abilities. That's all we're asking Pavel, or any other applicant, to do. Demonstrate the exact ability you claim to have. If Pavel claims to be able to correctly identify 30 of 40 photographs, then 29 does not meet his own conditions. Of course, until Pavel undertakes his test, your position on this is moot. What happens if he only gets 26 correct? Or 25? Or 19? Where do you draw the line? Or do you not recognise why it is necessary to draw a line? Do you?
	__________________ Aurora Walking Vacation "A point of view can be a dangerous luxury when substituted for insight and understanding."--Marshall McLuhan

13th August 2008, 10:45 AM	#122
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Originally Posted by Peristarkawan Just to make sure you understand this, it is possible to claim a success rate lower than 75% and still have 1:1000 odds. A lower success rate would just require more trials to prove. For example, as I noted above, the odds of getting 60 or more correct out of 90 are also about 1:1000. The real problem with that is that it would be a much longer test. Pavel understands this. He originally proposed more trials, but JREF felt this was problematic. The current design is intended to accommodate concerns raised by JREF.

13th August 2008, 06:28 PM	#127
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942	Originally Posted by Gr8wight Oh, Rodney, we've all had this discussion before. Fifteen or twenty times, if I recall correctly. The JREF is not serious about the paranormal. They are serious about challenging those people who make claims about the paranormal to demonstrate their abilities. That's all we're asking Pavel, or any other applicant, to do. Demonstrate the exact ability you claim to have. If Pavel claims to be able to correctly identify 30 of 40 photographs, then 29 does not meet his own conditions. Of course, until Pavel undertakes his test, your position on this is moot. What happens if he only gets 26 correct? Or 25? Or 19? Where do you draw the line? Or do you not recognise why it is necessary to draw a line? Do you? Let's continue this discussion in the existing thread "Odds Standard for Preliminary Test."
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942	Last edited by Rodney; 13th August 2008 at 06:44 PM.

14th August 2008, 03:15 PM	#128
GzuzKryzt Philosopher Join Date: Aug 2005 Posts: 6,363	Originally Posted by Rodney Let's continue this discussion in the existing thread "Odds Standard for Preliminary Test." In said thread, Pavel made the following post: Originally Posted by pavel_do He is right, though I think I will be able to perform 30 out of 40, will do my best for that. If I was sure I cant do that I would say I am not going for it… I was the one who asked for 40 pairs.. At least.. But the conditions is not as good for me as the other tests that would give me enough time, less stress in order to perform positive results. I know some one will say now.. you see.. he prepared already an excuse in case he fails.. THERE WILL BE NO EXCUSES… no one forced me to apply and take test.. It is my own decision.. Just that I want test to happen.. And if I will not step out of my way and test conditions that would make me really comfortable etc.. that the test will never happen.. cause it take too many hours.. cause I have advantage, cause it is too many photos etc.. CAUSE IT IS POSSIBLE FOR ME TO PASS IT WITH THOSE CONDITONS.. I stepped way to far just to move some how the negotiations and make test possible.. I would be more comfortable with 2-3 trials ( let say by 30 pairs in each) where achieving positive results in any would mean I passed it by covering any way 1.000 odds that needed to pass preliminary test.. I am sure during all 2-3 trial I would perform evident results that I am real deal and able to see and perform what I am claiming to! Having 2-3 trials also reduce a stress level that is the main issue for me in order to be as calm as I can during a test… Every one would feel calmer and more comfortable knowing that he can perform 2-3 trial.. But not 1.. I have explained that many times. Same with more pairs.. that is still covers 1.000 that is way far from getting by a chance and I see no problem in having a few hour test with allot of pairs that I am the one who pays for any way.. Etc.. many reasons that I have explained before many times that is just ignored, and I am in front of the fact that it will be not allowed full stop. Like passing it make them to give me the money! that is just a preliminary test as we all know. Just that passing it will mean that is possible to pass the second one and get the money. I will be honest with you. If the JREF will not let me to have results revealed immediately after each pair is named, I am not taking this test. Cause that would mean I have agreed to everything that will make test almost impossible to pass and I have to rely in chance to pass it.. Ill just drop it, and will be looking for other ways of testing ( maybe JREF offer something reasonable too). There is a time.. my claim valid for 1 year if I am not mistaken. I really want test to happen and doing my best to make it possible. Cause I know I will be able to show that such things is real and can exist and has to be studied etc.. There is a fake psychics.. but that dose not mean there is no real ones! I highlighted one relevant part.
GzuzKryzt Philosopher Join Date: Aug 2005 Posts: 6,363

14th August 2008, 04:49 PM	#129
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	Originally Posted by GzuzKryzt In said thread, Pavel made the following post: I highlighted one relevant part. I am sure most of people read it there, same as JREF.. I see no reasons to copy and post it here too.. Or... Oh yes.. cause I said if JREF will not allow the envelopes to be opened immediately.. (that is not affect odds in any ways), so I am not taking this test. And I am not taking this test with pairs, as I agreed to all that they wanted so far and that the last thing that would help me to show good results… and I am not getting off that.. If it cant not be agreed, that we have to find the other test to test me and I will do my best for that as I am doing any ways. So what the problem? I am being honest , as always but.. Every body, have a nice weekend. Regards, Pavel
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363

14th August 2008, 04:52 PM	#130
IXP Graduate Poster Join Date: Oct 2004 Posts: 1,395	Originally Posted by GzuzKryzt I highlighted one relevant part. What is your point? Pavel has always maintained that he foresees what will happen when the envelope is opened and this is why he needs each envelope to be opened immediately after the guess, so it is not confuse it with others. IXP
IXP Graduate Poster Join Date: Oct 2004 Posts: 1,395	__________________ "When reason sleeps, monsters are produced" -- Goya, title of etching that is my avatar

14th August 2008, 05:14 PM	#131
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	Originally Posted by IXP What is your point? Pavel has always maintained that he foresees what will happen when the envelope is opened and this is why he needs each envelope to be opened immediately after the guess, so it is not confuse it with others. IXP His point is obvious.. to twist it and make look like I have no intentions to have any test and sort of wasting my time and the others etc.. I am not even going to argue it.. it is pointless. I am just sick of it to be honnest. When some one don’t want to see (admit) the truth, he will not see or admit it.. and the truth is that I am not wasting anyone’s time, and I really want to have proper good test and show that what I am claiming I am able to perform. And I will do my best to make it happen. I just cant as a lame sheep, agree to all the conditions that would put me in the position to lose the test, but not being able to perform it is best… Any way I am sure who want to see it he sees it.. Or that was another joke? Or the way to encourage me? I am sorry but maybe due to a cultural difference and some others.. I don’t take it as a joke, and obviously not taking it as an encouraging…
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	Last edited by pavel_do; 14th August 2008 at 05:18 PM.

14th August 2008, 05:45 PM	#132
IXP Graduate Poster Join Date: Oct 2004 Posts: 1,395	I am with you Pavel. I think you have been quite reasonable in trying to determine exactly what you can do and under what circumstances. You are one of the few who has not changed your story or come up with new demands each time JREF proposes something. I hope you get a test that you believe you can pass. IXP p.s. This does not mean that I think you will pass!
IXP Graduate Poster Join Date: Oct 2004 Posts: 1,395	__________________ "When reason sleeps, monsters are produced" -- Goya, title of etching that is my avatar

14th August 2008, 05:50 PM	#133
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Originally Posted by IXP I am with you Pavel. I think you have been quite reasonable in trying to determine exactly what you can do and under what circumstances. You are one of the few who has not changed your story or come up with new demands each time JREF proposes something. I hope you get a test that you believe you can pass. IXP p.s. This does not mean that I think you will pass! I fully agree with IXP's position.

14th August 2008, 10:46 PM	#134
Akhenaten Heretic Pharaoh Join Date: Nov 2007 Location: Pi-Broadford, Australia Posts: 29,692	Hi Pavel. Welcome to город skeptiks It was always going to be tough for you to deal with the rigourous examination of your ability here. It's just the nature of the place that it will sometimes become quite adversarial, and you seem to have done a brilliant job so far of not taking it personally, but rather in the spirit of transparency in which most criticisms and suggestions have been made. Please believe that nobody here will attempt to twist your words with intent to discredit you. That's just not what the folks here do. Well, mostly But people will insist, quite rightly, that all the cards are on the table, so to speak. Your honesty has set you apart from the vast majority of previous claimants, and I'm sure this will continue until a mutually agreeable test is designed. I understand your need to see the envelopes opened as you make your choices, and I also understand the objections to this, but sadly I don't know nearly enough to help with a solution. I just wanted to add a voice of support for the way you've gone about this, regardless of the outcome. Good Luck Mate, Dave
	__________________ Life is mostly Froth and Bubble - Adam Lindsay Gordon

14th August 2008, 11:10 PM	#135
Ravenwood Muse Join Date: Jun 2002 Posts: 614	I can't believe I missed this! Pavel, you said earier that you could not use the blank/white photos because you confused it sometimes with the cross picture. What if you used only blank/white photos & black/exposed photos as your "targets" in your test? if you had one black & one white in the pair, perhaps you would "see" one or the other more distinctively. If that is the case, then you could do say sets of 5 envelopes, but only one contains the black or white photo(whichever you can feel strongest about), the other 4 contain the opposite color photo. That would be a 1 in five chance, or 20% as opposed to the 50% chance per pair, meaning less runs. I would be really interested in seeing what your success with that protocol is, I'm sure one of our number crunchers will come up with how many runs you would have to do to meet the odds, assuming that you are sticking to the 75% accuracy claim...
Ravenwood Muse Join Date: Jun 2002 Posts: 614	__________________ You can choose a ready guide in some celestial voice. If you choose not to decide, you still have made a choice. You can choose from phantom fears and kindness that can kill; I will choose a path that's clear- I will choose Free Will. -Rush, "Free Will"

15th August 2008, 10:39 AM	#136
Peristarkawan New Blood Join Date: Aug 2008 Posts: 7	Originally Posted by Ravenwood I can't believe I missed this! Pavel, you said earier that you could not use the blank/white photos because you confused it sometimes with the cross picture. What if you used only blank/white photos & black/exposed photos as your "targets" in your test? if you had one black & one white in the pair, perhaps you would "see" one or the other more distinctively. If that is the case, then you could do say sets of 5 envelopes, but only one contains the black or white photo(whichever you can feel strongest about), the other 4 contain the opposite color photo. That would be a 1 in five chance, or 20% as opposed to the 50% chance per pair, meaning less runs. I would be really interested in seeing what your success with that protocol is, I'm sure one of our number crunchers will come up with how many runs you would have to do to meet the odds, assuming that you are sticking to the 75% accuracy claim... The odds of getting 7 correct out of 10 would be 0.00086, about 1:1157. At a 60% accuracy rate, the odds of getting 9 correct out of 15 would be 0.00078, about 1:1274.
Peristarkawan New Blood Join Date: Aug 2008 Posts: 7

15th August 2008, 12:23 PM	#137
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Protocol under consideration There are a large number of statistical designs that can be used for these tests, as well as a large number of experimental procedures. Forumites have made a number of helpful suggestions along both lines that Pavel has used in coming up with a protocol. As of right now, there is a specific proposed protocol that was emailed to JREF just over a week ago. It was also posted here in the forum. While waiting for JREF to respond, it would be most helpful for folks here to comment on the current draft protocol. Remember that: (1) Pavel's claimed ability is to "see" the content of envelopes that will be opened after his identification. He "sees" the envelopes being opened. He does not claim to see into the sealed envelopes. This is a claim of clairvoyance, as opposed to "x-ray vision." (2) Pavel has always said his that his ability is imperfect. (3) JREF wants to hold down the number of envelopes involved, and has expressed some preference for the test to be done in pairs of envelopes.

15th August 2008, 12:42 PM	#138
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	Thanks Startz, I wouldn't answer better!
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363

15th August 2008, 03:10 PM	#139
petre Muse Join Date: Sep 2005 Posts: 887	Originally Posted by Startz There are a large number of statistical designs that can be used for these tests, as well as a large number of experimental procedures. Forumites have made a number of helpful suggestions along both lines that Pavel has used in coming up with a protocol. As of right now, there is a specific proposed protocol that was emailed to JREF just over a week ago. It was also posted here in the forum. While waiting for JREF to respond, it would be most helpful for folks here to comment on the current draft protocol. Remember that: (1) Pavel's claimed ability is to "see" the content of envelopes that will be opened after his identification. He "sees" the envelopes being opened. He does not claim to see into the sealed envelopes. This is a claim of clairvoyance, as opposed to "x-ray vision." (2) Pavel has always said his that his ability is imperfect. (3) JREF wants to hold down the number of envelopes involved, and has expressed some preference for the test to be done in pairs of envelopes. Thanks for the summary! I might have missed it in my skimming, or it might have gotten a bit obfuscated in translation, but it wasn't until this thread that I realized the claimed ability was clairvoyance rather than some flavor of remote viewing. That makes it greatly more clear why the post-selection reveal might impact performance. The following have probably already been suggested but... It would seem likely that with an ability like this, testing several differentation classes would be much easier (one would be less likely to confuse the results of multiple trials). For example, do one trial with six envelopes containing pictures of zenner cards. Pavel points the one out from a list that will be chosen, then a die is rolled and that number envelope opened and the item shown to Pavel. A second trial is then done with pictures of six different colors of full-page rectangles. Another trial might use pictures of numerical digits, etc. Perhaps he has more difficulty picking out one item from among a similar class of items though, so his own testing will have to guide the protocol. I was just hoping to offer a helpful suggestion based on my recent new understanding of the described ability.
petre Muse Join Date: Sep 2005 Posts: 887

17th August 2008, 09:28 AM	#140
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	Originally Posted by petre Thanks for the summary! I might have missed it in my skimming, or it might have gotten a bit obfuscated in translation, but it wasn't until this thread that I realized the claimed ability was clairvoyance rather than some flavor of remote viewing. That makes it greatly more clear why the post-selection reveal might impact performance. The following have probably already been suggested but... It would seem likely that with an ability like this, testing several differentation classes would be much easier (one would be less likely to confuse the results of multiple trials). For example, do one trial with six envelopes containing pictures of zenner cards. Pavel points the one out from a list that will be chosen, then a die is rolled and that number envelope opened and the item shown to Pavel. A second trial is then done with pictures of six different colors of full-page rectangles. Another trial might use pictures of numerical digits, etc. Perhaps he has more difficulty picking out one item from among a similar class of items though, so his own testing will have to guide the protocol. I was just hoping to offer a helpful suggestion based on my recent new understanding of the described ability. Thank you, good advise.. But I think for JREF such test will be too complicated as it will again involve allot of different pictures etc... and they are not research team that interested in researching and testing.. it is a challenge..
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363

17th August 2008, 01:55 PM	#141
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	I know it is out of topic.. but still would like to post it.. There is a web site offering a phone readings to there clients.. and there is a top pay “psychic” that charges 20$ per MINUTE… I am shocked.. Nevertheless.. here is one of the comments.. from the client.. that made me laugh so I would like to share it with you. “He started by saying things were really good for me & he was going to tell me why they were. He repeated that statement 3 times and then it sounded like he turned on the water and stepped into the shower--while still trying to talk to me. When I told him I could no longer hear him he said he would speak louder. Louder wasn't the problem--he sounded all garbled like he was drowning. Total waste of $$$$. If you don't have the time to do a proper reading don't be available for calls.”
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363

17th August 2008, 03:17 PM	#142
Moochie Philosopher Join Date: Aug 2005 Location: Australia Posts: 5,491	Originally Posted by pavel_do I know it is out of topic.. but still would like to post it.. There is a web site offering a phone readings to there clients.. and there is a top pay “psychic” that charges 20$ per MINUTE… I am shocked.. Nevertheless.. here is one of the comments.. from the client.. that made me laugh so I would like to share it with you. “He started by saying things were really good for me & he was going to tell me why they were. He repeated that statement 3 times and then it sounded like he turned on the water and stepped into the shower--while still trying to talk to me. When I told him I could no longer hear him he said he would speak louder. Louder wasn't the problem--he sounded all garbled like he was drowning. Total waste of $$$$. If you don't have the time to do a proper reading don't be available for calls.” Good one Reminds me of those sex lines which are staffed by all kinds of people -- grannies, guys with feminine sounding voices, and so on. I guess it's a living. M.

19th August 2008, 07:40 PM	#143
RemieV Philosopher Join Date: Jul 2006 Posts: 5,292	At this point, I'm going to need to have some discussions with other JREF staff members. They are trickling back from the Galapagos trip. I'll update soon.
RemieV Philosopher Join Date: Jul 2006 Posts: 5,292

19th August 2008, 09:55 PM	#144
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Originally Posted by RemieV At this point, I'm going to need to have some discussions with other JREF staff members. They are trickling back from the Galapagos trip. I'll update soon. Remie, Thanks for keeping us all in the loop.

21st August 2008, 03:13 PM	#145
Myriad The Clarity Is Devastating Join Date: Nov 2006 Location: Betwixt Posts: 19,977	The following would appear to address most if not all of the various issues and objections: There are 27 trials. There are 3 opaquely lined envelopes per trial. There is 1 photograph with an image and 2 blank photographs in each trial's set of envelopes. Pavel chooses one envelope per trial after examining all three as he desires (if necessary, within a pre-set time limit). A successful trial (hit) is choosing the photograph with an image. The result of each trial is revealed immediately after each trial. The image(s) can be the same for all the trials, or different for each trial, whatever Pavel thinks would be most helpful. He may be shown a copy of the trial's image or told in words what the image depicts at the start of each trial if that would be helpful. 18 hits in the 27 trials (2/3 or better correct) is a successful test. The probability of success by chance is well under 1 in 1000 (.000407). The probability of 17 hits or better by chance is .00154. The success rate needed for success corresponds to getting 27 or more hits out of 40, something that appears to be more comfortably within Pavel's claimed success rate than 30 out of 40. Note that the total number of envelopes involved is only one more than in the 40-trials protocol currently under consideration. (If I'm interpreting Pavel's comments correctly, he suggests failures occur when he is unable to perceive the future opening of an envelope showing a target picture when the envelope does in fact contain the target picture -- in other words, false negatives. If that's the case, then the presence of the two blanks instead of one in each trial set should not decrease the probability of success per trial, given that the examination time he would have available per envelope would be the same.) Respectfully, Myriad
	__________________ A zømbie once bit my sister...

21st August 2008, 03:42 PM	#146
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Originally Posted by Myriad The following would appear to address most if not all of the various issues and objections: There are 27 trials. There are 3 opaquely lined envelopes per trial. There is 1 photograph with an image and 2 blank photographs in each trial's set of envelopes. Pavel chooses one envelope per trial after examining all three as he desires (if necessary, within a pre-set time limit). A successful trial (hit) is choosing the photograph with an image. >snip Respectfully, Myriad I rather like this. It greatly reduces the probability of a false negative, down to about 40 percent. I think this is better than the revised protocol that Pavel (with a minor assist from me) submitted. Pavel: If you like this, perhaps you should send it in to Remie. -Dick

21st August 2008, 05:51 PM	#147
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942	Originally Posted by Myriad The probability of 17 hits or better by chance is .00154. So, under your proposal, would 17 out of 27 be a failure or would the preliminary test be extended?
Rodney Illuminator Join Date: Aug 2005 Posts: 3,942

21st August 2008, 07:04 PM	#148
Gr8wight red-shirted crewman Join Date: Jul 2004 Posts: 1,661	Originally Posted by Rodney So, under your proposal, would 17 out of 27 be a failure or would the preliminary test be extended? The applicant may submit the protocol design. The JREF will decide if 17 out of 27, or 18 out of 27, or some other result will satisfy their statistical requirements. In any case, there are two possible outcomes: a) the applicant is successful, or: b) the applicant is unsuccessful. Giving the applicant a do-over for being kinda close is not provided for in the rules. But why am I wasting my time writing this all out, Rodney? You already know this. You're just being deliberately obtuse for trolling purposes.
	__________________ Aurora Walking Vacation "A point of view can be a dangerous luxury when substituted for insight and understanding."--Marshall McLuhan

21st August 2008, 07:23 PM	#149
Myriad The Clarity Is Devastating Join Date: Nov 2006 Location: Betwixt Posts: 19,977	Originally Posted by Rodney So, under your proposal, would 17 out of 27 be a failure or would the preliminary test be extended? Please note that my proposal is a suggestion from a third party, not an offer from JREF. Pavel (with Startz's continued help) must negotiate these things. I would suggest -- still as a bystander -- that he choose ONE of the following possiblities, whichever he feels is most important to him, to negotiate for: 1. Explicit provision for a guaranteed extension of the test, within a set range of future dates, should a certain number of hits short of the requirement be made. 2. Allowance for a chance probability value for success somewhat over .001, such as allowing 17 of 27 to be counted as a success. 3. Slightly increased number of trials. With the trials as I suggested them, if there were 30 trials instead of 27 (that's ten more envelopes total than the 40-trial protocol, but fewer steps for the testers to go through during the test than the 40-trial protocol), 19 hits would be sufficient for p < .001 (.000737). That's essentially three more chances to make one more hit. (That seems surprising, but it's how the math works out.) Overall 19 out of 30 is approximately the same hit expectation per trial as 25 out of 40. Respectfully, Myriad
	__________________ A zømbie once bit my sister... Last edited by Myriad; 21st August 2008 at 07:24 PM.

21st August 2008, 08:12 PM	#150
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363	Originally Posted by Myriad The following would appear to address most if not all of the various issues and objections: There are 27 trials. There are 3 opaquely lined envelopes per trial. There is 1 photograph with an image and 2 blank photographs in each trial's set of envelopes. Pavel chooses one envelope per trial after examining all three as he desires (if necessary, within a pre-set time limit). A successful trial (hit) is choosing the photograph with an image. The result of each trial is revealed immediately after each trial. The image(s) can be the same for all the trials, or different for each trial, whatever Pavel thinks would be most helpful. He may be shown a copy of the trial's image or told in words what the image depicts at the start of each trial if that would be helpful. 18 hits in the 27 trials (2/3 or better correct) is a successful test. The probability of success by chance is well under 1 in 1000 (.000407). The probability of 17 hits or better by chance is .00154. The success rate needed for success corresponds to getting 27 or more hits out of 40, something that appears to be more comfortably within Pavel's claimed success rate than 30 out of 40. Note that the total number of envelopes involved is only one more than in the 40-trials protocol currently under consideration. (If I'm interpreting Pavel's comments correctly, he suggests failures occur when he is unable to perceive the future opening of an envelope showing a target picture when the envelope does in fact contain the target picture -- in other words, false negatives. If that's the case, then the presence of the two blanks instead of one in each trial set should not decrease the probability of success per trial, given that the examination time he would have available per envelope would be the same.) Respectfully, Myriad Thank you very much! Sounds good.. as even the results would be more evident than 50/50 chance in pairs... For that reason before I was offering also 1 out of 3.. and even 2 out of 5.. I can try this one tomorrow or Saturday.. and It still might be better for me if there will be 3 photos.. not 2 blank.. But sound “easy” and I done things like this before.. just that there was 3 photos.. not 2 blank and 1 photo.. But I think it does not matter at all for the odds.. So we could prepare with Startz a new proposal for JREF and see what they would say..
pavel_do Critical Thinker Join Date: Aug 2007 Posts: 363

21st August 2008, 08:31 PM	#151
Jackalgirl Graduate Poster Join Date: Aug 2004 Posts: 1,801	Hi, Pavel -- Play around with different combinations of regular photos and blank photos and see what works best. In the end, it doesn't really matter, as long as there's some kind of target image (or non-image) that you can identify out of two other images or non-images (which need not necessarily be the same as each other). I love the idea of three photos. This thread has inspired me to take a stats & probabilities class this next term. What a cool collaborative effort!

21st August 2008, 10:17 PM	#152
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Originally Posted by pavel_do Thank you very much! Sounds good.. as even the results would be more evident than 50/50 chance in pairs... For that reason before I was offering also 1 out of 3.. and even 2 out of 5.. I can try this one tomorrow or Saturday.. and It still might be better for me if there will be 3 photos.. not 2 blank.. But sound “easy” and I done things like this before.. just that there was 3 photos.. not 2 blank and 1 photo.. But I think it does not matter at all for the odds.. So we could prepare with Startz a new proposal for JREF and see what they would say.. Pavel: This sounds good. I think that to keep the negotiations on track you should put some weight on keeping things simple. You might do well to stick exactly to Myriad's suggestion, perhaps using 3 photos if that is really important.

21st August 2008, 10:19 PM	#153
Startz Muse Join Date: Nov 2004 Location: Santa Barbara, CA Posts: 669	Originally Posted by Jackalgirl This thread has inspired me to take a stats & probabilities class this next term. What a cool collaborative effort! Proving the success of the E in JREF!

24th August 2008, 06:38 PM	#154
gibsonbaud Guest Join Date: Aug 2008 Posts: 7	Im not sure if I am the only person who noticied this or if it is something that has been covered before, or if im wong on this, but the statistics most of you have been posting seem faulty... 10 sets of two envelopes: probability of getting them in two distinct piles... 50%. Every time you start a new pair the probability of getting it right goes back to one out of two... (i.e. 1 out of 2, 2, out of 4, 3 out of 6), it's a constant 50% probability. It would be different if a stack of 20 was handed to them and they picked the correct one of 20 each time, but 1 of 2 each time stays at 50%. Same goes for the three envelopes at a time method only its a 33.3% chance of getting them correct. Each time a new one is done the number correct or incorrect goes up, but so does the total number of envelopes taken in to consideration. EDIT: PS: I would suggest making it so the envelopes cannot be picked up, ie, they are taped to the table or are fastened down, allowing them to be tuched but not held.
gibsonbaud Guest Join Date: Aug 2008 Posts: 7	Last edited by gibsonbaud; 24th August 2008 at 06:45 PM.

24th August 2008, 08:29 PM	#155
Gr8wight red-shirted crewman Join Date: Jul 2004 Posts: 1,661	Originally Posted by gibsonbaud Im not sure if I am the only person who noticied this or if it is something that has been covered before, or if im wong on this, but the statistics most of you have been posting seem faulty... 10 sets of two envelopes: probability of getting them in two distinct piles... 50%. Every time you start a new pair the probability of getting it right goes back to one out of two... (i.e. 1 out of 2, 2, out of 4, 3 out of 6), it's a constant 50% probability. It would be different if a stack of 20 was handed to them and they picked the correct one of 20 each time, but 1 of 2 each time stays at 50%. Same goes for the three envelopes at a time method only its a 33.3% chance of getting them correct. Each time a new one is done the number correct or incorrect goes up, but so does the total number of envelopes taken in to consideration. Yes, but... My chances of success in the first pair is 50%. My chances of success in the second pair is 50%. However, my chances of getting both right are only 25%.
	__________________ Aurora Walking Vacation "A point of view can be a dangerous luxury when substituted for insight and understanding."--Marshall McLuhan