Beth Clarkson, Complex Protocol

[Ashles]

Beth Clarkson's experimental protocol (http://forums.randi.org/showthread.php?s=&postid=1870688146#post1870688146)

That seems incredibly over-elaborate and has a couple of real no-nos.

The sequence of trials (controls and tests) need to be unknown to the observer and anyone who makes measurements. It should also be determined in advance. I recommend that the subject determine the sequence and give the sequence to the observer in a sealed envelope or a password protected computer file prior to each experiment. The envelope or file would not be opened until after the experiment is complete and all measurement are complete and have been accepted.
So Beth decides the order of controls and tests? Why should she be deciding the order of anything here?

Absolute value of measurements vs. -90° to 270°. There are 360° in a circle. The target is set at 0°. The measurements can occur anywhere on the circumference of the circle. It isn’t appropriate to use 0 to 360 as the measurements, because measurements that occur close to the target, but on opposite sides would then be recorded as being very far apart rather than close together.
Bizarre over complexity - how is anyone supposed to measure exact degrees when we are talking about a small wax ring being melted?
And why -90 to 270 degrees?
Maybe the next point adresses this:

The target is oriented to the subject’s preference.
Erm, how about, no way whatsoever.
Why don't we just allow the subject to move the candle to the target area with her hand?

This whole claim is really looking like an absolute non-starter.

I predict the actual testing will not happen as the required protocol will be too strict for Beth (or her partners).
And by too strict I mean, of course, it won't allow cheating.

This is the point at which I imagine many paranormal claims get good results - because they allow the protocol to be dictated by the claimant and the experiment turns out to be far more elaborate and confusing than it needs to be.

[Ashles]

I'm not claiming a "power" sufficient to convince anyone, even me, in a single trial. I thought you guys wanted scientific evidence! Scientific evidence requires repeated trials, careful measurements and statistical analysis to establish whether a seemingly subtle difference between groups is real or just random chance. So what's the objection to using those in an experiment? - Beth
This test is never going to happen.



And KRAMER, could you ask her how she discovered this skill in the first place?

Can anyone think of any possible way anyone could 'discover' they had a skill as minute and untestable as this?

If it will require many trials and statistical analysis to discover whether there is any effect there at all WHY IS SHE IN ANY WAY CONVINCED THAT HER ABILITY EXISTS?

If it is too small to be noticed in a single trial then obviously she must have randomly decided she had this ability and then spent ages staring at flames trying to convince herself that she really had it.

I am convinced this is diplacement activity to distract her from the pressing problems in her life.

[Prolix]

What's preventing Prof. Clarkson from performing her own test and making it as thorough and elaborate as she feels necessary to be convincing, and then publish it to the public, including posting it in this forum, thereby challenging JREF to replicate it?

I hereby pledge $10,000 to the Clarkson challenge, wherein Clarkson awards JREF if JREF replicates her own completed test.

[apoger]

I pointed out in earlier discussion that the complex nature of this procedure was an issue.

I honestly have no idea how the JREF can reasonably deal with this.

Instead I will approach from another angle.
The claimant seems willing to look at the possiblity that she is in error. This is a good start. I propose a simple protocol that may help the claimant come to determine the validity of her "power" without all the complexity of the candle & wax fanfare.

The claim is that she can move things with her mind. It would seem that pushing a flame was chosen due to the limited force needed to make it move. I propose using a less complex target.

For this test the claimant will need a piece of lined paper, and some scissors.

Take the scissors and cut some tiny bits off the corner of the paper. Make certain that they are very tiny. So small that you might call them specks. So small that you have trouble handling them because static makes them stick to things.

Take three tiny specks and place them in a row, on a line, on the paper.

Puff a small breath near the specks.
See how they blow away as if slammed by a mighty force!
It takes almost no force at all to move things so small.

Put the three tiny specks back in a row, on a line, on the paper.

Now here is the test: Try to push the middle speck forward until it is off the line. The two outer specks are controls. If they move with the inner speck, it's probably a minute air current at work.

Success would be the ability to push ONLY the center speck off-line. It takes only the tiniest shard of force. Unlike a candle flame, the speck won't flicker or move back. Thus there will be less confusion about the result.

While not as stringent nor as definitive as a proper test, this is a test that can be done simply and inexpensively. Perhaps by using a simple test like this, the claimant can better determine if more complex procedures are called for.

[rppa]

I sympathize with Beth Clarkson, and on reading that thread so does Kramer, because she really does seem like she's trying to be rigorous.

She thinks she has a weak power. She thinks she can achieve statistically significant results. She's familiar with the language and mathematics of hypothesis testing.

The JREF standard ordinarily seems to be that the observed results should have less than 0.001 chance of occurring if random. That is, not only are they significant, but they're significant at the 0.999 level. She seems to be saying she can achieve that, given enough trials, but due to the amount of noise the number of trials needs to be large indeed.

As to the -90 to 270 thing. It sounds like she's trying to figure out how best to deal with the old problem of angle averaging. What's the average of -89 and +269 degrees? Do a naive average (add the numbers and divide by 2) and you get (-89 + 269)/2 = 180/2 = 90. But draw these two points on a graph and you'll realize they're only 2 degrees apart and their average is -90.

This is a solvable problem. Meteorologists do it all the time when working out average wind direction.

I think the other question you raise, about the subject choosing the order, is more important. I"m not clear on what is being ordered, but the JREF gold standard is usually double-blindness. How can you do double-blind testing if the power is "I can make the candle flame move in direction X?" You have to know what X is.

[Ashles]

As to the -90 to 270 thing. It sounds like she's trying to figure out how best to deal with the old problem of angle averaging. What's the average of -89 and +269 degrees? Do a naive average (add the numbers and divide by 2) and you get (-89 + 269)/2 = 180/2 = 90. But draw these two points on a graph and you'll realize they're only 2 degrees apart and their average is -90.
Sorry, I didn't make myself clear.
I understand that you don't want to have the measurements cross the 360 point as it buggers up the calculations.

But wouldn't -180 to +180 make more sense?

Combined with "The target is oriented to the subject’s preference" it just felt to me like she has a preferred part of the ring that she generally aims for, and the -90 to 270 seemed to reflect that.

If she has a preferred part of the wax ring then it looks even more likely that she is using air currents somehow.

[Moose]

Originally posted by apoger
For this test the claimant will need a piece of lined paper, and some scissors.

I can't resist: where does the rock come in? :D

[rppa]

Originally posted by Ashles
Sorry, I didn't make myself clear.
I understand that you don't want to have the measurements cross the 360 point as it buggers up the calculations.

But wouldn't -180 to +180 make more sense?

Combined with "The target is oriented to the subject’s preference" it just felt to me like she has a preferred part of the ring that she generally aims for, and the -90 to 270 seemed to reflect that.

If she has a preferred part of the wax ring then it looks even more likely that she is using air currents somehow.

That's not how I read it. It was just a longish way of trying to describe the best statistical averaging procedure, and she was worried about what to do if the measurements cover all 4 quarters whatever those quarters are. But she also said they usually only occur in 3 quarters and you can just use whatever choice makes all 3 quarters positive.

Almost certainly she's unconsciously using air currents. This is self-delusion by what seems to be a pretty smart and well-trained person. It's an aspect of woo-woo that always saddens me, because you want to let them down gently but there might not be any to do that.

On the other hand, according to Randi dowsers are usually pretty nice, sincere folks who are completely self-deluded, and they seem to weather the failure of rigorous tests with their self-esteem intact.

Maybe I shouldn't worry about Beth. If she's training as a scientist, she has to be prepared to see a dearly-beloved hypothesis break against the experimental evidence. It's how science is done.

[Yaotl]

What if she tries to keep a candle flame steady in the presence of air currents?

[alfaniner]

Originally posted by Yaotl
What if she tries to keep a candle flame steady in the presence of air currents?

Kind of like a drunk man walking during an earthquake...

[Kimpatsu]

It's all a nonsense to me. She can only perform on one out of every three tries? That's worse than chance. Hey, maybe she emanates negative psychic energy! Hey, I have a way to test her: Put her in a room with Uri Geller, and see if he can no longer bend spoons.
Errmmm...

[davefoc]

Kimpatsu and KRAMER seem to have the same view with regard to the one in three thing.

When MS. Clarkson's initial email and KRAMER's response about this I thought KRAMER had misunderstood what Ms. Clarkson was saying.

If somebody can telekinetically alter a flame even one in a thousand times that's still a paranormal event and worthy of the prize.

Clearly, if there is any genuine paranormal effect it is likely to be subtle since if it was obvious it would likely have been validated by now. So if KRAMER is going to reject the testing of every claimant just because the claimed effect is subtle he is making a decision to fail to test what are the most likely kinds of claims to be true.

The difficulty is in detecting a subtle effect. In this case the woman proposes that she can telekinetically alter a flame about one time in three that she tries. That seems like a highly testable proposition to me. Unfortunately based on the subsequent emails that I read (I didn't read them all) it seems like the difficulty is that the effect is so subtle that it is not clear when the effect is observed whether it was just a random flame variation or the result of telekinetic influence.

It seemed though that the woman had a protocol in mind to deal with this difficulty but that the protocol was so complicated that it exceeded JREF's patience and or resources to deal with. Too bad, but I suppose JREF's view is that there are so many claims for effects that are easily tested that fooling with ones like this isn't a useful use of their resources.

But if there was any such thing as a genuine paranormal effect it is more likely to be something like this than one where the claim is so strong that it is almost certain that if it was real somebody would have been able to demonstrate it by now.

[roger]

I agree, dave. Once again Kramer has made the mistake of confusing the chance of doing something with success rate. If I can predict the throw of a dice 1 in 3 times, that's paranormal (or I'm cheating). Moving a candle flame 1 in 3 times, if true, is an earth shattering change in how we understand the world works. She'd get the Nobel for it, easy peasy. It's a stunning achievement.

I also have sympathy for her complex protocol. Since when do scientists perform an experiment that alters the direction of physics in an afternoon? Sometimes - sure. If I say I can dowse with 85% success rate, then I better be able to deliver over a few trials, period. If my claim is that gravity exists, well, I drop my beer and tada, proved. On the other hand, if my claim is neutrinos exists, I set up a multi-million dollar detector and run it for years, hoping to detect that 1 in a trillion interaction with ordinary matter (or whatever the real odds are).

Hey, I don't blame JREF a second for not wanting to spend weeks on a single test, but the 'show me or shut up' (paraphrased) attitude really isn't a valid retort to someone claiming a subtle effect.

Of course, her experiment is flawed. She states she has to "close the room off or unexpected breezes can ruin the experiment". BAM. It's all over. No cookie for you. If you aren't controlling for an effect, you have proved nothing. Not to mention all the other possible confounding variables I and others have mentioned in the past. But Randi's test is flawed too, as it tests something she does not claim she is able to do.

[apoger]

Once again Kramer has made the mistake of confusing the chance of doing something with success rate.

Sadly, I must agree with this.
If I claimed that I could jump a mile straight up, but I needed about three tries to do it, it would still be a fantastic feat. Moving a flame with your mind seems like a smaller feat, but is no less fantastic.

Let's be honest. Most of us think the claimant is either misinterpreting chance events, or is biasing the experiment by influencing the flame with minute air currents. She has chosen a test method that is incredibly vulnerable to such.

I still feel the best way to go is not to dismiss the claim, but to negotiate a protocol that is acceptable.

Again, the claim is telekinesis, not flame pushing. The flame pushing was just the overly complicated way that the claimant thought up to test the telekinesis. What we need is a better test for telekinesis that both parties will find agreeable.

How about this:
A plane of glass is coated with a layer of very fine dust.
Very fine dust being chosen becuase it's incredibly light. A force that could waver a flame, could tap a particle of dust. Unlike a flame, dust particles don't flicker nor reset themselves. I think we can all agree that dust particles are a better target than a flame.
The plane is placed on a table.
A target location is chosen on the plane. Perhaps it can be indicated with a laser pointer.
The claimant can then attempt to push particles in any direction, away from the target location. The claimant will be given as many attempts as she can muster during a time limit. Thus there is no issue with the power only working one in three times. For the sake of argument, let's say that we afford the claimant three hours to poke at the target location.
Success would be if the dust remains undisturbed on the plane, with the exception of the target location which would show clearing, lumping, smudging, or any obvious indication that particles have been moved.



Well, those are my best suggestions on the matter.
I hope all parties involved can come to an argeement that leaves all satisfied.

[Kimpatsu]

Apoger, your protocol won't work. She doesn't direct her telekinesis in a straight line like a beam; she skips through supspace to directly affect the flame as if there were no spatial separation. (That's one excuse.)
Is she saying that her TK can't influence the direction of the flame? Because if it can, then the answer is simple. Have her pull the flame towards her on each try. Then, even accepting her 3-in-1 oil hypothesis, a statistically significant result is possible; on each group of three tries, she must at least once pull the flame towards her. (Having negated the possibility of external air currents by, for example, placing the candle in a bell jar.
See? Not so difficult.

[webfusion]

"If I can predict the throw of a dice 1 in 3 times, that's paranormal (or I'm cheating)" --

Um, I hate to burst anyone's bubble, but putting my money where my mouth is, I have done this at the Mohegan Sun Casino numerous times. It's called CRAPS. a little game of anticipating the outcome on throws of two dice. On so-called proposition bets, you are "predicting" what the Next Roll will be!
I happen to be fairly good at it. C'mon snake eyes!! (pays 30-to-1)

But I cannot even imagine what JREF would say if I came to them with the proposition that they come stand around and watch how my chip stack grows and after 16 straight hours of play, pay me the million, based on the 1-in-3 "paranormal effect" being proven!!! I wish...
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ \\\\\\\\\\\\

Also ---- Her claim is telekinesis, not flame pushing Nope, her claim is flame pushing.

She has not even made one reference of ability to alter at-rest position of a solid such as dust, paper, grains of sand or other minor particles. Where is there any indication of Beth Clarkson even hinting at "moving" anything other than a flickering candle flame?
(by the way, wax melting and pooling under the wick itself can alter the direction that the flame flickers, causing it to "bend sideways" -- try it and you'll see).

[B.S]

I can't imagine a more useless ability.

At least if you could bend spoons you could make your supper host cranky.

[roger]

Originally posted by webfusion
Um, I hate to burst anyone's bubble, but putting my money where my mouth is, I have done this at the Mohegan Sun Casino numerous times. It's called CRAPS. a little game of anticipating the outcome on throws of two dice. On so-called proposition bets, you are "predicting" what the Next Roll will be!
I happen to be fairly good at it. C'mon snake eyes!! (pays 30-to-1) Don't worry about my bubble, it's doing fine.

If you can beat the predicted odds, then yes, JREF will test you. Go ahead and apply.

[69dodge]

When trying to measure small forces, it's important to eliminate the effects of air currents. That's why analytical balances (http://www.sfu.ca/chemistry/courses/chem110-111/techniques/analytical_balance.htm) are generally enclosed.

A flickering candle flame is just about the worst thing to use; all it is, after all, is hot air currents.

Extremely sensitive torsion balances (http://www.elgi.hu/museum/tinga_.htm) have been around for hundreds of years. I made a really simple one by tying a string around the middle of a toothpick and hanging it inside an empty 2 liter Coke bottle. Screwing the cap on holds the string in place and completely seals out external air currents. Even so, if I put it on top of my stereo amplifier, the toothpick twists slowly back and forth if the amp is on. I assume this is because air currents are set up inside the bottle due to uneven heating by the amp. If the amp has been off for a while and has cooled down, the toothpick is motionless.

According to my back-of-the-envelope calculations, a force of only 10 micrograms or so is enough to twist it noticeably. And the string I used is not especially thin. Using fine thread would make it even more sensitive.

(UnrepentantSinner's link in the other thread also involved torsion balances, although it didn't call them that.)

Ms. Clarkson is a statistician, so it's understandable that she's focusing on the statistical analysis. But it's often much cheaper and easier to reduce the noise in the data directly by better experimental technique rather than by increasing the number of trials and averaging it out. The standard deviation goes like the square root of the number of trials; if I want to reduce the noise by a factor of ten, I need a hundred times as many trials. And, of course, no amount of repetition will eliminate systematic error caused by poor experimental technique.

[roger]

Excellent points, 69dodge. I think JREF should take your post nearly verbetim, and send it Beth, as an explanation for why they are declining this challenge unless she can exhibit gross effects with the flame, or subtle effects with a better apparatus.

[webfusion]

"If you can beat the predicted odds, then yes, JREF will test you. Go ahead and apply."

I am not making any open claim of paranormal abilities, rog. It just seemed appropriate to point out rolls of the dice that are bet as "propositions" and yet win consistently can be attributed to other reasons besides 'cheating' (um, like random luck). When I win ten times in a row (which I've done), I can assure you the house is not thrilled, but that's the game. It's their dice, and I'm not cheating. Over time, they'll get their money back if I continue to try and overcome those astronomical house odds against me
(trust me on that).

But when a streak happens... it is a really sweet thing to see!
The people around the table are actually cheering me on, and I'm like a celebrity for a few moments!
roger, if I was able to walk into a casino or horsetrack with the ability to really see in advance (or be able to paranormally alter) the outcomes, then you can be sure james randi is gonna be the LAST person on Earth to be told about it!!! *grin*

http://www.temptatts4u.com/giant-flaming-dice.jpg
Here's what I would love ---- Beth Clarkson showing us how she can set dice on fire! That would do it for me!
Otherwise, she's just gonna remain another delusional in a long line of rejected applicants...

[apoger]

Originally posted by Kimpatsu
Apoger, your protocol won't work. She doesn't direct her telekinesis in a straight line like a beam; she skips through supspace to directly affect the flame as if there were no spatial separation. (That's one excuse.)
Is she saying that her TK can't influence the direction of the flame? Because if it can, then the answer is simple. Have her pull the flame towards her on each try. Then, even accepting her 3-in-1 oil hypothesis, a statistically significant result is possible; on each group of three tries, she must at least once pull the flame towards her. (Having negated the possibility of external air currents by, for example, placing the candle in a bell jar.
See? Not so difficult.


The subspace excuse is exactly the sort of nonsense that claimants will come up with. Agreed.

On the other hand we can just as easily use the same sort of excuse making to completely dismiss your suggestion of impossing a bell jar. After all the jar would block my hand on the way to the flame. She could just as easily say that the jar will defeat the telekinetic push.


We will not get far at all if we are to dismiss protocols on the assumption that each one will get shot down with nonsense.

[apoger]

Nope, her claim is flame pushing.

If you check her opening claim it starts with: Claim to be demonstrated: TELEKINESIS

I think she made herself quite clear.

[apoger]

Extremely sensitive torsion balances have been around for hundreds of years. I made a really simple one by tying a string around the middle of a toothpick and hanging it inside an empty 2 liter Coke bottle.

Outstanding! I am all for moving the test to a more sensitive and definitive target. I really think that this is the only way you will get any satisfaction from this claim.

[KRAMER]

[QUOTE]Originally posted by davefoc
So if KRAMER is going to reject the testing of every claimant just because the claimed effect is subtle he is making a decision to fail to test what are the most likely kinds of claims to be true.

=============================================

Good point. But, what constitutes "true"? This notion of the existence of numerous truths sounds somewhat Zen to me.
While I personally respect such notions, they simply do not apply to the JREF Challenge, wherein only a single "truth" applies when asking the prime question: DOES IT WORK, or DOESN'T IT?

Are you actually suggesting that if someone can succeed in demonstrating something once in three attempts, that this is a successful demonstration?

This applicant is NOT claiming a paranormal effect because it works less than chance would dictate. This is the very foundation of JREF's requirements in proving a paranormal claim.

In point of fact, she is really claiming that she can NOT do it, except on occasion.

The more I think about this, the more I feel convinced that this is simply not a paranormal claim. Randi agrees 100%.

This is not about semantics. I understand some of the criticisms of JREF's position regarding this claim, but the Challenge begins with one basic question that we, as the organization offering the prize, must decipher the answer to, and that is...

IS THIS A PARANORMAL CLAIM?

Randi will not entertain any suggestion that a one-out-of-three demonstration constitutes a success.

People, this is a One Million Dollar prize. Do you really think we're going to write a check in that amount for anything less than conclusive proof? That kind of money is not awarded for a demonstration of "subtle effects" that may or may not suggest paranormal phenomenon.

Quite the contrary. Read the Challenge rules. If you disagree with the stringency of the Challenge requirements, you certainly have the right to do so. It will, however, have no impact upon what we require, as per the Challenge rules.

That having been said, should you ever decide to offer your
own Paranormal Challenge and cash prize based upon your own set of requirements, I have a file full of applicants who failed the JREF preliminary test whom I am certain would just love to apply for it.

[apoger]

Are you actually suggesting that if someone can succeed in demonstrating something once in three attempts, that this is a successful demonstration?

It absolutely depends on the effect being generated.

Let's say that I claim to have the power to walk through walls.
Are you telling me that if I made three attempts, and if only successful once, that you would discount this success as being less than chance?



Randi will not entertain any suggestion that a one-out-of-three demonstration constitutes a success.


In the past Randi has made it very clear that each claim must be evaluated so as to determine what constitutes success.

Has this changed?

[KRAMER]

Originally posted by apoger
In the past Randi has made it very clear that each claim must be evaluated so as to determine what constitutes success.
Has this changed?

Absolutely not.

If you say that you could walk through walls, and achieve it even ONCE out of TEN times via means that are clearly not involving sledgehammers, heavy equipment or other physical means,
that would be paranormal.

Let's be real, here. OK?

The hypothetical claim you have offered as a comparison is obviously a paranormal one. FYI: We've never had such a claim.
If we did, we'd want to see a videotape before accepting it for testing. I think such an approach to such an extraordinary claim is entirely reasonable, from any viewpoint.

This specific claim submitted by Beth Clarkson, on the other hand, has been carefully evaluated, independent of all other claims, and it has been determined that nothing paranormal is involved.

Do you disagree with that assessment, or are you simply trying to draw me into a pointless debate over JREF's refusal to cater to
absurd demands from self-deluded applicants making nutty claims?

That's right - I'm in a rather foul mood here this Saturday evening at the JREF.

[Kimpatsu]

Originally posted by webfusion
[B Also ---- Nope, her claim is flame pushing. [/B]
Which is telekinesis. Her TK only affects fire, however. Like Pyro out of the X-Men.

[webfusion]

apoger and kimpatsu are both right, and I stand corrected.
What I meant was that her ability was only to be proved by flame movement, and nothing about moving stuff around (as Nina Kulagina).

Hey KRAMER, can you ask Ms Clarkson to move aside these flames ? (see photo)

(just trying to make your evening more enjoyable -- forgive the sophomoric attempt at humor)

[Kimpatsu]

Originally posted by webfusion
Hey KRAMER, can you ask Ms Clarkson to move aside these flames ? (see photo)

(just trying to make your evening more enjoyable -- forgive the sophomoric attempt at humor)
Yeah, we're really burned up about it... :p
She's hot stuff, BTW.

[Brian]

Ask her if she can push the flame of a blow-torch. Drafts won't make it move and if she does move it, it will be obvious.

[davefoc]

KRAMER Wrote:
Good point. But, what constitutes "true"? This notion of the existence of numerous truths sounds somewhat Zen to me.
While I personally respect such notions, they simply do not apply to the JREF Challenge, wherein only a single "truth" applies when asking the prime question: DOES IT WORK, or DOESN'T IT?
KRAMER,
Thank you for your response. I don't want to belabor this. Apoger summed up my thoughts on this with his walk through walls analogy.

I think the issues associated with this testee are similar to the kind of issues the ganzfield experiments engender. There is a very subtle effect claimed that can only be detected with large numbers of trials. I have wondered how JREF would respond to proposals that are searching for this kind of effect. This case is perhaps a little different in that there is the confounding issue of determining unambiguously whether a positive effect has been achieved.

I haven't read through all the communication associated with this so perhaps I am wasting people's time here but I would be looking to reduce some of the ambiguity in determing that a positive effect has occurred. Perhaps by placing a thermocouple near the flame and looking for a temperature rise when the testee attempts to move the flame.

I might get fancy and setup a computer to record the thermocouple output versus time and to record via a button push when she was attempting to move the flame.

At this point one could do lots of testing and get lots of results. This is vaguely similar to the kind of testing I did occasionally as an engineer. Then when the test started producing results I'd adjust the test to investigate various hypothesis. Of course, I pretty much started all those kind of investigations with an assumption that there was a nonparanormal explanation for whatever problem I was investigating but it seems like the approach that I applied there might apply in this case also.

Just for the record since you don't know me, I am and always have been as far as I can remember a complete skeptic about this kind of thing. I agree with apoger's ealier statement that the likelyhood that there is a genuine paranormal effect here is about as likely that apoger can jump three miles straight up.

[CurtC]

Originally posted by Ashles
If it will require many trials and statistical analysis to discover whether there is any effect there at all WHY IS SHE IN ANY WAY CONVINCED THAT HER ABILITY EXISTS? This reminds me of usenet conversations I used to have on rec.gambling.craps a few years ago, and since the subject of craps has already been brought up, I'll share it. There are nuts out there who claim to be able to "set" dice, meaning throw the numbers they want to throw. Note that in craps, the dice must bounce off the rear wall of the table, which is covered with pyramid shapes to make the rolls even more unpredictable. I would challenge this assertion, and the conversation would typically go like this:

Dice-setter: I can set dice.
Me: You can? Then why are you here, and not on your mega-yacht?
DS: I don't like to use it that way, plus it doesn't work that way.
Me: Can you roll the numbers you want to roll?
DS: Well, not every time. But I can do it often enough to win overall.
Me: Oh yeah? How often is that?
DS: Well, Bob Johnson calculated that decreasing the odds of rolling a seven from 16.667% down to 15% would negate the house advantage, and actually tip it in your favor.
Me: Well, let's see... (figures a while)... to actually measure that small of a change, you would need to keep track of 320,000 dice rolls, writing down the result of each and then tallying them all up. Of course, if you have not done this, then there is no way for you yourself to know whether you really have this skill. Have you kept these logs?

And that would be the last I heard of that person, or he would claim to just "know" what he can do without measurements. I wonder if I ever got through to any of them. Most of my time there was spent trying to convince people that there is no successful craps "system." No ordering of individually negative-expectation bets can result in an overall positive-expectation game. I had more success with that.

[WaterD]

I agree that the claim cannot be accepted under the contest rules, but if the porpouse of randy is instead look for some paranormal event, then i would give some attention to the matter. The point is that i did hear and read several times that people minds, focusing on affect a random event bring a difference of 1% or 2% in favor of the result the people where thinking about, and this after numeros test. I'm very sceptikal, still i'm open that something could exist in that matter. I mean if one wants to know if that claim is true, it requieres a hard work, but what if it's true?

[Kimpatsu]

Where is Water D fropm? If his mother tongue isn't English, I can forgive the above post, but otherwise... :rolleyes:

[WaterD]

No, my english is very bad i'm from argentina, that's the main reason i usually don't post even i read the forums since 2003

[Prolix]

I'm very sceptikal, still i'm open that something could exist in that matter.

Kinda like Randi, the difference being that Randi responsibly backs his "openess that something could exist" with a monetary award.

[rppa]

Originally posted by WaterD
No, my english is very bad i'm from argentina, that's the main reason i usually don't post even i read the forums since 2003

Could you please edit your post to include what you meant to say in Spanish? My Spanish isn't fluent, but I'll bet that there are enough people here who understand Spanish to help you with the English syntax and let you contribute.

Por favor puede cambiar su mensaje y incluir lo mismo en espan~ol?

[Czarcasm]

Originally posted by WaterD
No, my english is very bad i'm from argentina, that's the main reason i usually don't post even i read the forums since 2003
And the reason you cannot look at the top of this page and see that the man's name is "James Randi" and not "randy" is......?

[IXP]

Originally posted by webfusion
"If I can predict the throw of a dice 1 in 3 times, that's paranormal (or I'm cheating)" --

Um, I hate to burst anyone's bubble, but putting my money where my mouth is, I have done this at the Mohegan Sun Casino numerous times. It's called CRAPS. a little game of anticipating the outcome on throws of two dice. On so-called proposition bets, you are "predicting" what the Next Roll will be!
I happen to be fairly good at it. C'mon snake eyes!! (pays 30-to-1)

But I cannot even imagine what JREF would say if I came to them with the proposition that they come stand around and watch how my chip stack grows and after 16 straight hours of play, pay me the million, based on the 1-in-3 "paranormal effect" being proven!!! I wish...


Webfusion,

Stanford Wong, who has written extensively on how to make money at blackjack, used to offer a $10,000 prize for anyone who can play craps for a specified period of time (I cannot remember what it is) and come out ahead. And you do not have to claim paranormal ability, just win. You might want to contact him to see if this is still offered. Here is his web site and contact info:

http://www.bj21.com/

If you decide to accept, I would be willing to make a side bet against you. Craps is a negative expectation game, being "good" at it only means you lose as little as possible, and the strategy is trivial, you bet only "pass line" and the maximum odds bet.

[IXP]

This effect is not impossible or even difficult to test.

If Beth can affect the flame one out of three times above chance, you can test her in an equal probability binary test with 30 trials.

The probability of getting 24 right out of 30 is .0007, less than 1 in a 1000. To hit 24 she needs only the 15 expected chance hits and 9 hits via PK, less than the 1 out of 3 she claims.

[Ashles]

I think this quote by CurtC is very interesting and relevant:
Dice-setter: I can set dice.
Me: You can? Then why are you here, and not on your mega-yacht?
DS: I don't like to use it that way, plus it doesn't work that way.
Me: Can you roll the numbers you want to roll?
DS: Well, not every time. But I can do it often enough to win overall.
Me: Oh yeah? How often is that?
DS: Well, Bob Johnson calculated that decreasing the odds of rolling a seven from 16.667% down to 15% would negate the house advantage, and actually tip it in your favor.
Me: Well, let's see... (figures a while)... to actually measure that small of a change, you would need to keep track of 320,000 dice rolls, writing down the result of each and then tallying them all up. Of course, if you have not done this, then there is no way for you yourself to know whether you really have this skill. Have you kept these logs?

And that would be the last I heard of that person, or he would claim to just "know" what he can do without measurements. I wonder if I ever got through to any of them. Most of my time there was spent trying to convince people that there is no successful craps "system." No ordering of individually negative-expectation bets can result in an overall positive-expectation game. I had more success with that.

As I wondered earlier on: how has Beth ever got the impression that she can do this, if the effect is as faint at is seems to be, and it is only detectable by a long-term series of trials and subsequent statistical analysis?
Logically she must have decided she had this ability before she ever saw any evidence for it.

Also (as mentioned by other posters) can we clarify what the claim is? If it is (very limited) telekinesis, then creating an experiment shouldn't be too hard (good suggestions above by other posters).

I think the JREF should try and accomodate the claimant as far as possible, but if Beth is determined that her protocol is the only way to test her very specific candle flame moving ability then I can't see how the JREF is the right organisation to test this.
This needs the type of testing that only a University (or similar organisation) is set up to carry out.

If the JREF ever were inclined to cary out a series of tests as long-term and elaborate as this I feel it would only be for a claimant who sounded like they really might have the claimed ability. This claimant just doesn't.

[Yaotl]

Originally posted by Ashles
I think this quote by CurtC is very interesting and relevant:

As I wondered earlier on: how has Beth ever got the impression that she can do this, if the effect is as faint at is seems to be, and it is only detectable by a long-term series of trials and subsequent statistical analysis?
Logically she must have decided she had this ability before she ever saw any evidence for it.

Also (as mentioned by other posters) can we clarify what the claim is? If it is (very limited) telekinesis, then creating an experiment shouldn't be too hard (good suggestions above by other posters).

I think the JREF should try and accomodate the claimant as far as possible, but if Beth is determined that her protocol is the only way to test her very specific candle flame moving ability then I can't see how the JREF is the right organisation to test this.
This needs the type of testing that only a University (or similar organisation) is set up to carry out.

If the JREF ever were inclined to cary out a series of tests as long-term and elaborate as this I feel it would only be for a claimant who sounded like they really might have the claimed ability. This claimant just doesn't.

Indeed, it sounds like she really wants a test to see whether or not she has a testable ablility in the first place. She seems unsure and might be wanting to use the JREF as her method to see if she has the ablility or not, not to prove her ablility.

[69dodge]

Originally posted by IXP
This effect is not impossible or even difficult to test.

If Beth can affect the flame one out of three times above chance, you can test her in an equal probability binary test with 30 trials.

The probability of getting 24 right out of 30 is .0007, less than 1 in a 1000. To hit 24 she needs only the 15 expected chance hits and 9 hits via PK, less than the 1 out of 3 she claims.The hard part is deciding, in any given trial, whether the flame has been affected. We need to know, with much more confidence than the p-value we're aiming for, what flames do when they're not being psychically affected.

Running a few control trials and saying, "well, it looks sort of Gaussian, so I'll use a t-test," as she appears to have done, is not good enough. A p-value of 0.1 % from such a test is entirely meaningless unless we're much surer than 99.9 % that the assumptions underlying the test are justified.

Or, regarding your suggestion of an equal probability binary test, what are the two possible results, and how can we be sure that they really are equally probable? If they're not, and especially if successive trials are not independent, any p-value we may get goes out the window.

It's not easy to do right, which, presumably, is why she's been getting positive results.

[davefoc]

69didge put forth my understanding of the problem (together with some statistical comments that I don't mean to claim I understand).

The problem seems to be that sometimes the flame moves randomnly in such a way as to mimic the effect of being moved with psychic power.

That's why I put forth the idea of a thermocouple connected to a computer to monitor flame movement. This way one could automatically detect flame motion with and without telekinetic efforts and easily collect quantified data to monitor the flame motion.

A possible protocol now might be just as simple as running the experiment over something like ten sessions half with telekinetic efforts and half with out. The challenge would then be for the testee to determine which tests involved telekinetic effort and which didn't based on the computer data. One could then set the success criteria at a certain number correct to achieve the reliability of the test result that was desired.

[KRAMER]

Originally posted by webfusion

Hey KRAMER, can you ask Ms Clarkson to move aside these flames ? (see photo)


Perhaps I should try first, just to see if it's even possible.

[KRAMER]

Originally posted by Ashles

If the JREF ever were inclined to cary out a series of tests as long-term and elaborate as this I feel it would only be for a claimant who sounded like they really might have the claimed ability. This claimant just doesn't.

This assessment seems to nail it on the head.

Beth has basically come right out and said that she "needs JREF's help in determining if this effect is real or not".

Is this really what JREF is supposed to be doing?

I think she can figure this out on her own by conducting the test on her own. But will she accept the results?

[IXP]

Originally posted by 69dodge
The hard part is deciding, in any given trial, whether the flame has been affected. We need to know, with much more confidence than the p-value we're aiming for, what flames do when they're not being psychically affected.

Running a few control trials and saying, "well, it looks sort of Gaussian, so I'll use a t-test," as she appears to have done, is not good enough. A p-value of 0.1 % from such a test is entirely meaningless unless we're much surer than 99.9 % that the assumptions underlying the test are justified.

Or, regarding your suggestion of an equal probability binary test, what are the two possible results, and how can we be sure that they really are equally probable? If they're not, and especially if successive trials are not independent, any p-value we may get goes out the window.

It's not easy to do right, which, presumably, is why she's been getting positive results.

69Dodge,

Her protocol involves determining where a drop of wax first falls from a ring of wax that is placed centered above the candle.

To make it an equal probabily binary test, all you have to do pick a target direction at random, then divide the ring into two equal parts with the target at the center of one. If the drop falls from that half of the ring it is a success, if it falls from the other half it is failure. Ambiguous "on the line" results should be discarded. With the target directions randomized, this will give equal probabilities even if there is a bias toward some partiticular direction in the experimental setup.

As far a making the trials independent, using a new candle and wax ring for each trial should do. Use a new glass also if you think non-uniform heating or soot deposits might affect it.

If she can perform better than p<0.001, this should be considered sucessful for the preliminary test. For the million, I believe the p value has to be reduced to 1 in a million, which would take a lot more trials, but I wouldn't worry about the need to do this, since she is highly likely to fail the prelimary as everyone else has.

[Startz]

As a long time JREF board fan, here are a couple of comments on the Beth Clarkson application that I think might be helpful. Some of these points have already been made by others, but perhaps saying them another way will be helpful.

First, when Clarkson says "level of confidence" she's using a standard term of art from statistics. "Level of confidence" refers to the chance that a statistical test gives the right answer. It doesn't have anything to do with anyone's emotional state. Clarkson is using the term in the correct scientific sense, although she certainly isn't explaining it well.

Second, it makes sense to perform repeated tests to detect a weak - but real - power. Suppose I claimed that I could make a fair coin land heads but that my power only worked 1/3 of the time. If my claim were valid and we tossed the coin 12 times we'd expect 4 heads from the times I succeeded in exerting control, plus 4 heads and 4 tails from random chance from the tosses where my control failed. So if my claim is true, we expect 8 heads as compared to a fair coin with only 6 heads.

If I succeeded in getting 8 or more heads out of 12 would JREF be willing to say I'd passed the preliminary challenge? Of course not, because even a fair coin will give 8 or more heads quite often. One would insist on many more than 12 throws.

Suppose I said my power worked all the time. Would 12 throws be sufficient? Probably yes, because the chance of getting 12 heads from a fair coin is 1 in 4096, which meets the standards for a preliminary challenge.

So a weak power does require more trials, just as Clarkson says. A weak effect would be just as "paranormal" as a strong effect. It's just requires much more data to prove. This, of course, is the applicant's problem, not JREF's.

Dick Startz

[Ashles]

So a weak power does require more trials, just as Clarkson says. A weak effect would be just as "paranormal" as a strong effect. It's just requires much more data to prove.
I think we are all agreed that if anyone is to test Beth in the way she desires many tests would be required.

I feel the biggest concern though is that this is a claimant who does not appear to even be sure that she has the ability she is talking about.

There is no logical way she can suddenly have discovered she has this ability as, by her own claim, it requires repeated trials and statistical analysis of the data to confirm it either way.

I.e. one day she just decided she had this ability, without ever having actually observed it in action.

Her subsequent tests may have been highly subjective, and it is obvious she wants to believe she has this ability. Her 'associates'/'partners' seem to be encouraging her (wish we knew more about them).

To take your coin example -
Even if I actually had this mild ability,would I ever notice it? We toss coins rarely and if I happened to get 8 heads out of 12 tosses I would never think twice about it.
I would have to be tossing coins all the time (for some strange reason) to even notice I had this ability. That's if I even noticed in the first place. (I honestly believe if I had the ability to guarantee heads one time in three I would still never notice I had this ability)
And this example talks about something whuch has very clear known odds - 2 possible options. Heads or tails.
A candle flame, on the other hand, is an extremely random thing.
How could anyone ever possibly detect they had this ability?

The JREF does not serve as some kind of organisation which you can go to if you feel you might have a paranormal ability, but you're not sure, and you want someone to test you to find out and tell you if you do or not.

[IXP]

Originally posted by Startz

Second, it makes sense to perform repeated tests to detect a weak - but real - power. Suppose I claimed that I could make a fair coin land heads but that my power only worked 1/3 of the time. If my claim were valid and we tossed the coin 12 times we'd expect 4 heads from the times I succeeded in exerting control, plus 4 heads and 4 tails from random chance from the tosses where my control failed. So if my claim is true, we expect 8 heads as compared to a fair coin with only 6 heads.

If I succeeded in getting 8 or more heads out of 12 would JREF be willing to say I'd passed the preliminary challenge? Of course not, because even a fair coin will give 8 or more heads quite often. One would insist on many more than 12 throws.



You are correct. I need to revise my previous estimate because I was counting on getting half of all trials by chance in additon to the PK ones. The correct calculation is to subtract the PK affected trials from the trials first (as you did) and then add half the remaining ones.

The new requirements to prove you have the power to affect an equal probability binary outcome 1/3 of the time is:

62 correct of 93 trials to get a confidence level of 1 in a 1000 (suitable for preliminary test)

138 correct of 207 trials to get a confidence level of 1 in 1,000,000 (suitable for prize)

[KRAMER]

I've had RAndi look at many of the most recent comments supporting the possibility of testing this claim, and he comments as follows:

===========================================

Understood, but if someone is claiming to get results less than chance would call for, she has NO ability!

Should we test her for having a huge failure rate?

James Randi.

[Startz]

Originally posted by KRAMER
I've had RAndi look at many of the most recent comments supporting the possibility of testing this claim, and he comments as follows:

===========================================

Understood, but if someone is claiming to get results less than chance would call for, she has NO ability!

Should we test her for having a huge failure rate?

James Randi.

One question of course is when JREF should spend time trying to untangle an applicant's not very clear claim. I'll just defer to Randi and Kramer's experience on this.

But there are two other questions.

I took Clarkson's claim to be that she could move a candle by more than chance would suggest, just that she couldn't do it all the time. If someone could make a coin land heads 2/3rds of the time, it would be an astounding achievement.

The second issue is should "we" test her for having a huge failure rate. Let me respectfully disagree with Randi. Suppose Clarkson claimed to be able to throw more heads than would be dictated by chance and then proceeded to throw 100 tails in a row. That would be just as much evidence that she has a paranormal power, even though it would be the opposite of the power she claimed.

Of course since Clarkson is trained in statistics, she ought to be able to design a proper experiment herself. (And if she does that one might hope that the result would be that she'd realize that she has no powers at all.)

[Beady]

Originally posted by Startz
But there are two other questions.

There's actually a third other question. Perhaps I'm stirring the pot, here, but still...

This is highly theoretical, but suppose that someone manages once in every "n" trials to do something that, heretofore, has always been considered completely impossible. A 1% success rate is certainly worse than chance would dictate, but wouldn't doing it at all be worth something?

[Kimpatsu]

Originally posted by KRAMER
Should we test her for having a huge failure rate?
Hey, negative psychic powers! :D

[Ashles]

Understood, but if someone is claiming to get results less than chance would call for, she has NO ability!

Should we test her for having a huge failure rate?

James Randi.
Whilst I don't in any way think this will turn out to be a useful way for the JREF to spend it's time and energy in investigating, what Beth is actually claiming is certainly higher than chance.

I don't think Randi has read everything to do with the claim and if he has he must understand that claiming less than 1/2 success rate does not mean less than chance as explained by many posters above.

I think it is fairly clear that I, personally, do not think this claim has any grounds whatsoever, but to reject it because Beth is claiming less than chance is incorrect. She isn't.

By all means reject it because of the uneccessary over-complexity of the test, or the ridiculous length of time it would take to test, or on the grounds that she has apparently not even demonstrated this ability to herself, or even the fact that she may not be in the soundest psychological state to take the test (worried as she is about personal issues).

But please don't reject her because she is claiming less than 1/2 chance rate, because that would be unfair in this particular instance. And statistically incorrect as far as I read it.

KRAMER, surely you can see that she isn't claiming less than chance rate - please pass that on to Randi as it is a silly reason to reject the claim.

[roger]

We'd better fire all the major league baseball players, then, cause almost to a man they aren't batting at a 1 in 3 success rate. Who knew they were doing worse than chance?

[Kimpatsu]

Apples and oranges, Roger.
For starters, in baseball, the pitcher is deliberately trying to prevent the batter from hitting the ball; do you think the flame is consciously trying to resist being pushed around by Beth Clarkson?

[Beady]

[Sorry. I posted then decided I had nothing important to say, but I can't just delete the message.]

[Stereolab]

Any chance of getting Beth Clarkson on this forum? I have a couple of questions I'd really like to ask her.

[davefoc]

Nicely said Ashles.

[Dr Adequate]

Originally posted by Beady
There's actually a third other question. Perhaps I'm stirring the pot, here, but still...

This is highly theoretical, but suppose that someone manages once in every "n" trials to do something that, heretofore, has always been considered completely impossible. A 1% success rate is certainly worse than chance would dictate...
If the effect is held to be impossible (finding a yeti on a yeti-hunting expedition) then the probability of it happening would be 0%. Hence, a 1% success rate would be better than chance.

[Beady]

Originally posted by Dr Adequate
If the effect is held to be impossible (finding a yeti on a yeti-hunting expedition) then the probability of it happening would be 0%. Hence, a 1% success rate would be better than chance.

Well, then, if it is held that Beth can't affect the flame at all, but she does it one out of three tries...

Oh, never miind. It's still a pretty useless ability to have.

[Kimpatsu]

Originally posted by Beady
Oh, never miind. It's still a pretty useless ability to have.
To be fair, "uselessness" isn't the issue. Paranormal ability is the issue, and so far, she ain't demonstrated none.

[KRAMER]

Originally posted by Startz
Suppose Clarkson claimed to be able to throw more heads than would be dictated by chance and then proceeded to throw 100 tails in a row. That would be just as much evidence that she has a paranormal power, even though it would be the opposite of the power she claimed.

Agreed, but this is NOT what she is claiming. She's actually claiming to do a demonstration that would prove nothing.

If she said she could make a flame move south, but then succeeded in moving it north 100 times, it would certainly warrant closer investigation. Contrary to the insistence of the woo-woo's,
JREF would NOT look away from evidence that something paranormal was taking place just because the demonstration didn't precisely match the claim.

If we saw something - ANYTHING - we'd look more deeply.

[KRAMER]

Originally posted by Stereolab
Any chance of getting Beth Clarkson on this forum? I have a couple of questions I'd really like to ask her.

I have tried several times to convince Beth to join the forum, but she has said that she is "not in the mood" to get bashed and called "delusional" by "complete strangers".

She has also admitted that she may indeed be deluded regarding her beliefs. She wants us to help her to decipher truth from fantasy here.

Again, is this really what the Million Dollar Challenge is all about?

Yes, we are an Educational Foundation, but the Challenge was neither designed nor intended to support an individual's delusions, and agreeing to "help" applicants to verify whether or not something paranormal is involved in what they THINK they are seeing, would open some torrential floodgates, and permanently alter the very nature of the Challenge.

It was also not intended to be a free reserach think-tank for anyone who thinks they MIGHT have powers. Quite the contrary.

Happy Thanksgiving.

[Startz]

Originally posted by KRAMER
Agreed, but this is NOT what she is claiming. She's actually claiming to do a demonstration that would prove nothing.

If she said she could make a flame move south, but then succeeded in moving it north 100 times, it would certainly warrant closer investigation. Contrary to the insistence of the woo-woo's,
JREF would NOT look away from evidence that something paranormal was taking place just because the demonstration didn't precisely match the claim.

If we saw something - ANYTHING - we'd look more deeply.

Let me quote three things Clarkson has said
The applicant will demonstrate control of a candle flame by directing the heat of the flame towards a specific target. The results of the attempted control, even when successful, are subtle. Success occurs on less than half of attempts. ... my results indicate that the effect I am producing is subtle (perhaps 30 degrees difference on average when successful) and somewhat irregular (I estimate that I am successful in only 30% to 50% of trials)

Seems to me that this statement supports Kramer's point.

Later Clarkson said
What I was trying to communicate is that what I consider a "success" occurs on only about 1/3 to 1/2 of my attempts, thus multiple trials are necessary to establish my claim.

A success should always lead to a positive result. Random chance will lead to a positive result 50% of the time and a negative result 50% of time. If I am "successful" on one third of my trials, I can expect to see (+) results on 2/3 of all trials (1/3 from actual successes and 1/3 from random chance) and (-) results on 1/3 of all trials. This is a testable deviation from the results of random chance alone.

Now that seems like a perfectly reasonable and very testable suggestion.

Clarkson also says
I currently teach mathematics at a university and am pursuing a Ph.D. in statistics. For me, whether we win or lose the challenge, public revelation of my involvement in such an activity is likely to be detrimental to my professional future.
I am qualified and competent to design and analyze such tests.

But she hasn't actually offered up such a test, now has she?

And winning the JREF prize would not only guarantee a successful career in statistics, given what academic salaries are like it the prize would pretty much cover her salary for a couple of decades at least!

So to a researcher, a reasonable response might be "Ms. Clarkson, you're pursuing a PhD. YOU need to propose a protocol describing a trial, how success will be decided for each trial, and the details of the statistical analysis to be conducted after the data is gathered."

But I'm reminded of something Randi has said many times. Scientists aren't very good at designing this kind of test. We think the world "makes sense" and research is conducted under that assumption. JREF applicants frequently don't make sense of any kind.

I hope Kramer and Randi would encourage Clarkson to submit a seriously testable application rather than just reject what she's said so far. But it's not JREF's responsibility.

Let me make a small, serious, offer to help out. I would be happy to contact Ms. Clarkson's dissertation supervisor, explain the question at hand, and ask if (s)he would help Clarkson design an appropriate protocol. Obviously, I would only do this with Clarkson's very explicit permission.

[KRAMER]

Originally posted by Startz
I would be happy to contact Ms. Clarkson's dissertation supervisor, explain the question at hand, and ask if (s)he would help Clarkson design an appropriate protocol. Obviously, I would only do this with Clarkson's very explicit permission.

I'd wager my left foot that Beth wouldn't want ANYONE from her place of employment to know what she is up to here.

More from Randi:

"If I said I could tell you whether a randomly selected playing card would come up red or black only one out of 3 times,
would that be a paranormal claim?"

I do believe that the answer would be a solid and resounding NO.

That said, Startz, I've forwarded your previous posting to Randi, in its entirety, for comment. I will post it upon receipt.

[Ashles]

More from Randi:

"If I said I could tell you whether a randomly selected playing card would come up red or black only one out of 3 times,
would that be a paranormal claim?"

I do believe that the answer would be a solid and resounding NO.

I don't understand why Randi is being like this. He should know perfectly well that this is not a comparable example.

If it was shown that 1 in every 3 times Beth could make a flame hit a target point on a ring that would be clear evidence of paranormal ability. There isn't really any doubt or debate about that.

This random 1 in 2 probablity representing all chance events has come out of nowhere and is flat out incorrect.

If Randi wants to liken it to cards ask him if a person could predict the exact card from a deck 1 in every 3 times, would that be above chance?
Obviously it would. So once again I am unclear why he does not understand that what Beth is claiming (regardless of the reality or likelihood of the claim to be real) is definitely above chance.

[Startz]

Originally posted by KRAMER
I'd wager my left foot that Beth wouldn't want ANYONE from her place of employment to know what she is up to here.

More from Randi:

"If I said I could tell you whether a randomly selected playing card would come up red or black only one out of 3 times,
would that be a paranormal claim?"

I do believe that the answer would be a solid and resounding NO.

That said, Startz, I've forwarded your previous posting to Randi, in its entirety, for comment. I will post it upon receipt.

Kramer,

No argument at all about this particular applicant.

I think a couple of us have a disagreement with you and Randi about some principles of statistical testing. I hope the sense comes through that this is friendly input from admirers of your work.

Maybe you could ask Randi to do the following: Take a deck of cards (two decks would be better), call red or black for each card before turning it over and keep track of the number of times he's right. I bet Randi can't manage to get it right only one-third of the time! If he does enough cards, even trying to fail, he'll get about half of them right.

(Okay, okay. Randi probably can draw 53 aces from a deck of 52 cards. So if he wants to give this a try he has to pretend to be unAmazing for a bit.)

[KRAMER]

Originally posted by Startz
I think a couple of us have a disagreement with you and Randi about some principles of statistical testing. I hope the sense comes through that this is friendly input from admirers of your work.

OF COURSE, OF COURSE, OF COURSE. I can tell the difference between you and Peter Morris - trust me.

I'm forwarding many of these comments on Beth Clarkson to Randi for his input, as he is much too busy to visit the forum.

For such an utterly useless claim, it's certainly spawned some of the most intriguing debate I've seen here in the forum.

I wonder how it will all come turn out. For now, Beth's latest email has suggested that she will contact us again when she can perform with a higher rate of success, which could mean never.

[Brian]

I don't mean to sound like a broken record, but has she said that the flame has to come from a candle? Can she do it with a blowtorch?

[Smike]

I think the problem here is whether or not the flame will move by itself anyway, and if so, at what probability.

Beth appears to be assuming about 0%.

If the candle is contained to avoid wind currents, I would expect it to be pretty low. In this case, a succes rate of 1/3 is pretty good.

Howver, if the flame will randomly flop about at a probabilty of .5, then she's not doing anything special.

It shouldn't be too difficult to set up a test which lowers the random element a lot. Assuming the TK works through glass or whatever, which she appears to claim it does.

[Ashles]

Smike, I love the avatar.

KRAMER, I do think the claim should be rejected simply on the grounds of how small the claimed effect is and how complicated and long-term the analysis would be.

I think most of the concern expressed here is merely to do with how certain people on this and other sites view any rejection of an applicant.
Very few people would have issue with Randi rejecting this claimant for several different reasons (as outlined several times before by myself and other posters).
But the reason should not be because of Randi misunderstanding the statistical analysis of her claim.

And don't worry, we don't think you are thinking of us as Peter Morris types.:)

[Beady]

Originally posted by Ashles
I do think the claim should be rejected simply on the grounds of how small the claimed effect is and how complicated and long-term the analysis would be.

Is that in the rules? If not, how do you propose to avoid a justifiable charge of arbitrariness?

[Ashles]

It wouldn't be an arbitrary change. The test is set up to produce a clear and unequivocal result, firstly, in a preliminary challenge, then in the main test.

Beth's protocol, firstly produces no clear result. Complex statistical analysis to determine if an effect exists is not part of the JREF challenge or methodology.

Secondly, Beth's protocol requires long term testing. This is not how the JREF test is set up. It requires a clear demonstration in a single test or series of tests. It would just not be practical to perform the JREF tests in this way.

The rules state that both parties have to be happy with the testing. If the JREF is not happy with the protocol it is under no obligation to agree to the testing. It can deny the claimant if it is unhappy with the testing for any reason.

If the believers wish to see this as unsatisfactory that's really too bad. Anyone should be able see this whole claim looks highly suspect anyway.

[roger]

Originally posted by KRAMER
For such an utterly useless claim, it's certainly spawned some of the most intriguing debate I've seen here in the forum. Well, it's a fascinating case, and complicated. An educated, smart, apparently normal person, who for some reason is investing a lot of effort, and perhaps risking her academic career, for a 'skill' she has yet to adequately demonstrate to herself. Yet she claims 1/3 success. Why do so many people think they know something without actually knowing it? It's a fascinating question.

And then the effect, extremely subtle if by some miracle does exist, requires significant maths to prove. You couldn't keep us away from this topic with a stick :D.

And there is the frustrating feeling of this being a real claim. Not that it's real as in the ability is real, but that this isn't somebody who should be in medical care. It would be great to see the test run, and assuming she failed, to see the reaction. Does she recant like the person mentioned in the recent commentary, or does she go on searching? Anyway, how much more interesting than the applications where the applicant can't even describe what they are going to do, yet alone devise a way to test it.

Of course with the test how it is, there's no way JREF can pursue it. Oh well.

[Ashles]

If you wanted to have paranormal powers but had never seen any evidence that you actually had them, would would be the most faint, hard to detect thing you could think of?

I find it hard to think of anything fainter than the ability to slightly deflect a candle flame one in three times.

You know, just the sort of ability you stumble across by chance.

[Beady]

Originally posted by Ashles
I find it hard to think of anything fainter than the ability to slightly deflect a candle flame one in three times.

You know, just the sort of ability you stumble across by chance.

Don't you mean "outrageous chance"?

Think about it. How did she first "discover" she might have this ability? I don't know about anyone else here, but I keep a couple of emergency candles in the car and the house, but can't remember the last time I lit one. How often does Beth encounter burning candles, in places where there is so little movement that the flame is perfectly still? What led Beth to set up a laboratory experiment? Was she on a date, in a restaurant, dining by candlelight? Was her companion so boring that she found the candle so interesting she couldn't take her eyes off it? Was the restaurant's business so dead that night that no one, including Beth and her companion, moved around enough to cause air currents, thus enabling the flame to remain perfectly steady?

Or did she just wake up one morning and wonder if she had a strong enough TK ability to move a candle flame, and was curious enough to set up a lab experiment?

[IXP]

Originally posted by Beady
Don't you mean "outrageous chance"?

Think about it. How did she first "discover" she might have this ability? I don't know about anyone else here, but I keep a couple of emergency candles in the car and the house, but can't remember the last time I lit one. How often does Beth encounter burning candles, in places where there is so little movement that the flame is perfectly still? What led Beth to set up a laboratory experiment? Was she on a date, in a restaurant, dining by candlelight? Was her companion so boring that she found the candle so interesting she couldn't take her eyes off it? Was the restaurant's business so dead that night that no one, including Beth and her companion, moved around enough to cause air currents, thus enabling the flame to remain perfectly steady?

Or did she just wake up one morning and wonder if she had a strong enough TK ability to move a candle flame, and was curious enough to set up a lab experiment?

Actually, if you read all the correspondance on this, it seems likely that her friend and partner suggested that she had this ability to affect physical objects with her mind and set up the candle movement demonstration. I think it is quite possible that there is a trick here that the friend is quite aware of and is using to convince a skeptical Beth of this paranormal ability. Beth might then be responding by seeking another opinion in the form of JREF. Pure speculation, but I think it fits the known facts well.

[KRAMER]

Originally posted by Beady
Think about it. How did she first "discover" she might have this ability? What led Beth to set up a laboratory experiment? Was she on a date, in a restaurant, dining by candlelight?...Or did she just wake up one morning and wonder if she had a strong enough TK ability to move a candle flame, and was curious enough to set up a lab experiment?

Beth has been most forthcoming with me regarding certain events in her life that preceded her present pursuits in this area, and, although she has requested that I not divulge the specific circumstances, I really don't think it would be betraying her confidence or breaching her privacy if I said that her "belief" was precipitated by a powerful emotional trauma.

Sound familiar?

I've tried to get her to see that the two events are most likely very connected. She feels it's entirely coincidental.

This claim keeps me up at night. It's very sad.

[rppa]

Originally posted by Kimpatsu
Hey, negative psychic powers! :D

Sadly, I've heard this term. I read once of a psychic demo in which members of a group were tested for ESP using Zener cards. Pure chance suggests that you should get the right symbol 20% of the time. The psychic called any score above 20% evidence of psychic powers, and any score below 20% evidence of "negative psychic powers", but still psychic.

If you guess one out of five choices randomly, the chance of getting exactly 20 right in 100 trials is about 9.9%. Does that mean 90.1% of the population is either positively or negatively psychic?

[IXP]

Originally posted by rppa
Sadly, I've heard this term. I read once of a psychic demo in which members of a group were tested for ESP using Zener cards. Pure chance suggests that you should get the right symbol 20% of the time. The psychic called any score above 20% evidence of psychic powers, and any score below 20% evidence of "negative psychic powers", but still psychic.

If you guess one out of five choices randomly, the chance of getting exactly 20 right in 100 trials is about 9.9%. Does that mean 90.1% of the population is either positively or negatively psychic?

No, they alternate between the two randomly and cannot predict before any trial which it will be.

[Beady]

Fellas? Something just occurred to me.

This thread has been going on for a while, and a lot of people who know a lot more than I do about Chance and general statistics have been arguing back and forth and getting nowhere. IIRC, isn't one of the rules for the Challenge something on the order that the results of any test have to be self-evident and non-ambiguous? I think about the only thing that has come out of this discussion is that the proposed test can *only* yield ambiguous results. Or, at best, even the experts can't agree on what a self-evident result would be.

Isn't it time to chuck the flame idea and start over?

Just off the top of my head, it seems to me that anyone who could move a flame via TK could also roll a penny standing on edge along a level surface or, alternatively, knock the penny over. There, you would have a result that is both self-evident and easily quantifiable.

[Kimpatsu]

Originally posted by Beady
Just off the top of my head, it seems to me that anyone who could move a flame via TK could also roll a penny standing on edge along a level surface or, alternatively, knock the penny over. There, you would have a result that is both self-evident and easily quantifiable.
But what if her power only affects flames, like Pyro out of the X-Men?

[Skeptical Greg]

Alas, is the world forever to be denied the benefits of candle flame moving.:(

[KRAMER]

I've suggested numerous things (other than a candle flame) for Beth to consider demonstrating her powers with, to no avail.

She insists that the candle flame is the best way to demonstrate her "powers".

I think we're getting nowhere, fast.

[Beady]

Originally posted by KRAMER
I've suggested numerous things (other than a candle flame) for Beth to consider demonstrating her powers with, to no avail.

She insists that the candle flame is the best way to demonstrate her "powers".

I think we're getting nowhere, fast.

Well, then, it seems to me this is *her* problem, not yours. Or ours. Why are we even discussing this? As I read the rules, isn't the claimant (or whatever) responsible for designing the demonstration, with the caveats that it be performed under controlled observation and the results be self-evident? She is not providing for self-evident results; hell, ISTR even she is having trouble describing how to read the results. When I was student teaching, my mentor told me the reason you can't explain something is that you don't understand it; if she doesn't understand it, how is anyone else supposed to? For that matter, how would she know whether she had succeeded and you and Randi weren't just keeping the money? How could she prove it?

[BanthaTracks]

Isn't the bottom line her, is that she needs to demonstrate to herself that she has this ability and then once convinced, approach a thirdy party to verify her results and THEN approach JREF?

Why bother discussing elaborate protocols when she hasn't even convincingly demonstrated the ability to herself.

There are any number of good, less complex setups/experiments suggested in the forum she can try to first demonstrate the ability to herself. Time to move on and have her come back later with some results.

[Beth]

Hello,

Kramer has encouraged me to post here looking for suggestions. As many of you have surmised, no, I am not certain this ability exists. Nor am I certain that it does not. That’s why I’m running experiments.

Currently, my experiments are on hiatus due to being busy with other things this time of year. However, I do hope to start up again after the holidays and would like to work out any changes to my experimental protocol in the interim. I have managed to find a few people willing to act as observers (lack of observers is the main problem with my current experiments and the reason I submitted a claim to JREF in the first place – if my claim had been accepted, they would have provided me with observers). I’m hoping to have some additional subjects participating as well.

I’m looking at making changes to the set-up, based on the suggestions that I’ve heard from JREF. I hoping that you will provide me with some additional advice regarding my experimental procedure. Here are the issues I’m currently trying to resolve and would appreciate your suggestions for improvement.

I’m currently working with a candle inside a glass and a series of 8 uniformly molded flat wax rings (actually washers might be a better description) placed on top of the glass. I use 4 for controls and 4 for tests in a randomized order. I measure the angle from a target for the first drop of melted wax that falls from each ring. Unconscious manipulation via breath is eliminated because the entire set-up is raised until the flame is at eye-level and the only opening is above the subject’s head.

In addition to protecting from outside air currents, the glass also serves as a jig for uniform positioning of both the candle and the wax ring. This positioning is not perfect, but is done to the best of my ability. The candles have enough variation (wicks are not perfectly centered within the candle) that I’m not going to be able to do any better than this until I switch to a different type of flame.

Type of flame: I should be able to switch from a candle to an oil lamp or butane lighter. I’m currently looking into what’s available, the cost, and what will work best with my set up. Any reasons to prefer one over the other?

Air flow/currents: Would a larger container with a separate jig to hold the flame and wax ring in position placed entirely inside of it be better? If so, what can be used as a jig? I need something small (it must go inside the glass container) and inexpensive. Further, would this type of set up really be less susceptible to air currents than the current one? Currently, the top of the wax ring is exposed, but the underside is not. There is only a small (1� diameter) opening, so the flame itself has very little exposure to outside air currents. I have found some reasonably priced large clear glass tubes (chimneys for an oil lamp) that would be suitable. But I am concerned that with a larger container, the larger diameter will mean more exposure of the flame and wax ring to outside air currents than with my current set up.

Vibration: How can the effects of vibration best be minimized? A rubber mat? A foam pad? A dish filled with sand? A large stack of books? I’m not at all familiar with how to damp vibrations, so suggestions in this area are greatly appreciated.

Measuring angles: What is the best method of averaging angles? I’m currently leaning towards an absolute value measurement as it simplifies things. Computations with my current data have a higher p-value when I use absolute values rather than positive and negative ones, so it seems the more conservative measure. However, it has the disadvantage of underestimating the variance and thus won’t be appropriate for certain types of analysis.

Analysis: For each trial, I compute the mean for the control rings (4 independent measurements) and the mean for the test rings (another 4 independent measurements). I then compute the difference between those means: the control mean minus the test mean.

Currently, I think the binomial analysis is the best. It’s conservative and robust. For the binomial distribution, I only consider the sign of the difference between the means. A positive value is a “success�, a negative value a “failure�. (For those unfamiliar with statistics, the binomial distribution requires that results be classified into one of two groups, traditionally termed “success� and “failure�.) I compute the probability of my results under the null hypothesis of no effect, i.e. that the probability of a “success� is 0.50. The only assumption needed for this analysis – and I think it’s a reasonable one – is that under the hypothesis of no psychic effect it is equally likely for the control mean or the test mean to be closer to the target.

Trial to trial data seems to follow a uniform distribution (as would be expected). Deviations from the mean within a trial (looking just at control data to establish what would be typical) have a nice bell-shaped distribution. I have nearly 60 sets of control data from my experiments. While I have changed some features of the testing over those trials, the control data would be unaffected by those changes. A t-test seems appropriate for analyzing the data from a single trial, but I won’t accept the results of a single trial as being definitive, despite having a very low p-value for some trials.

The within trial variance is not constant from one trial to the next. I presume this is due to the problem of imperfect centering of the flame. Thus, though a paired t-test would seem appropriate to compare the sample means, I am reluctant to place much confidence in those results due the non-constant variance. If I can solve that problem, I should be able to use that type of analysis.

I do have another statistician who’s agreed to take a look at my data over the semester break. However, I would like to hear from others regarding their thoughts on appropriate analysis techniques. I would also be willing to provide my current set of data to anyone who would like to review it. It is stored in an EXCEL spreadsheet.

[Skeptical Greg]

May I ask what you suppose the usefullness of this ability is, if it exists?

[roger]

Hi Beth, welcome to the forum. I'm glad you've joined us.

There are many issues with your experimental design. The main problem is you are not directly measuring what you claim you are doing. Your claim is that you can move a flame, but what you are measuring is heat conduction and the change of state of a material (wax). This set up is tricky enough that it really would require a lot of experimentation to determine exactly all the independent variables, and more experimentation to ensure that you have adequately controlled for them. By experimentation I don't mean fiddling around, but full on, lab conditions, multiple runs, etc., such as you would find in any competent lab. Given that, I honestly can't see why JREF would ever test this claim as it stands - we're talking months of work, just to prove that all conditions are controlled for.

But let's proceed with some of the issues with the current set up. To do so, let's make sure we understand the physics. The lit candle generates light, heat, and various residues (hot gases, soot, etc). I will discount light from futher analysis because I can't envision a way for it to effect the wax melting. The heat of the candle heats the surrounding air, as well as the candle itself. This heat is transferred through the air to the 1) wax ring; 2) the glass chamber; 3) the air outside the glass chamber; 4) the table top; 5) the candle (separate effect from the candle heating itself).

There at least 3 other heat sources in this experiment: 1) the room air which can enter through the hole in the top; 2) the room air which cools the glass chamber; 3) the desk top.

These varying heat sources all contribute to the heating of the wax ring. The various sources all start at different temperatures, so the heat flow model inside that glass chamber is quite complex, perhaps even chaotic. I'm using the mathematical definition of chaotic, in which the system is deterministic, yet unpredictable.

Let me enumerate some of the issues that could effect the wax melting on a trial by trial basis. This does not even begin to be exhaustive:

* outside air temperature - contrary to your claim, unconscious breathing towards the apparatus will alter the outside air temperature, which will alter the heat convection.

* glass temperature - can be modified by 1) your handling while setting up the apparatus 2) breathing on it during the test

* centering of the candle in the apparatus

* vertical orientation of the candle in the apparatus

* dimensional and density unifomity of the wax ring

* table top temperature - each trial will add heat to the table top, meaning it will be non-uniform across trials

* positioning of the ring on the glass.

* air currents entering through the hole in the top

* candle height at the start of the burn, and how long after the burn that the trial starts.

etc.

That's a lot that you need to experimentally prove that you have controlled for. Clearly, a statement that the hole in the ring is small, and thus probably doesn't allow room air currents to effect the results cannot be accepted at face value - this must be proven. I can imagine just that proof alone taking several weeks or even months of clever experimental design and execution.

---------------

That you are allowed to handle the apparatus is a huge red flag, and honestly, probably what is causing you to see results that make you think you are affecting the flame. Very tiny errors in placement could cause large variation in melt position. Do you have experimental evidence for, say, how much the melt position changes for a 0.1mm error in candle position in the X axis effects the result? What are your QC requirements for manufacturing the wax disks? 6 sigma? How do you measure the positioning of the ring, and what is the error rate of measurement? How much will that error cause variations in the melt position? Etc. I can easily imagine that completely unconsciously you might position the wax ring differently, whether based on horizontal position, or the wax's density, based on the target. Do you know what the position of the target is going to be before you set up the ring? Do you know if it is going to be a control trial or real trial before you position the apparatus? Do you know the temperature of all materials and gases prior to the start of the trial, and do you control for that in your analysis?

Those are all rhetorical questions. I hope it is clear that you have designed a system which is exquisitely sensitive to a multitude of
variables, none which you are controlling, fully understand, or have data on their impact on the wax melt point.

I can't imagine, based on Kramer's statement on here, quite reasonable statements, that JREF will not invest weeks in investigating subtle phenomena like this, that this challenge will ever be accepted in it's present form.

--------------------------------------

So let's examine what you are claiming, and see if there is not a better way to measure it. Your claim is telekinesis of a flame. The first thing I would ask is there any reason to believe your ability is restricted to flames? I ask because there are well established ways to measure very tiny forces. The design of a torsional balance that is isolated from external influence, for example, is well understood. It's still time consuming, and expensive, and difficult to do well, but we know how to do it. No one has ever studied your particular set up, so all we can do is speculate about it.

Another, easier set up would be to place 3 tiny pieces of styrofoam in a sealed chamber, isolated from vibration and heat sources. Your job is to move one and only one of the pieces on command. This is a set up Randi has used before, and so he would be likely to accept it as a challenge demonstration.


If for whatever reason your skill is limited to flame, then I would move to an apparatus with less complex heat flow. A flame from a consistant source (some kind of gas flame), in a large chamber (so it is not coupled to the walls of the chamber, with a long burn prior to the start of the trial (so temperatures are stable, and convection, if any, is laminar), rigged up with vibration and temperature sensors (so you can discard trials with abnomalities), and a camera and computer system set up to measure flame movement. You still have the issue of the proximity of the table top (heat coupling), but hopefully you can remove that from consideration using the temperature sensors to prove steady state conditions.

I really don't see how you can prove your claim with less than months of experimentation, and I don't see how JREF could accept your challenge on just your say-so that you have done these experiments and identified all the contributing factors to the flame movement or heat conduction/convection. And JREF isn't going to do the months of experiments either.

I'm somewhat pessimistic, and so sometimes assume things will be harder than they really are, so I hope somebody else sees flaws in my limited analysis above.

One assumption I have made in the above, and it may not be warranted, is that the movement of the flame due to your mind is very small, and not trivially visible to the naked eye. For example, if I blow at a candle with a strong puff of air, it is immediately visible because the flame jumps. People would shout out "you just blew at the flame" even if they didn't see me do it. If you are getting those kinds of gross movements, than it is a much easier experiment to design, and a simple camera system coupled with your isolation from the air in the test apparatus would do. But I'm guessing the candle wavers during the control run, and wavers during the real trials, and you think maybe the wavering is different when you are trying to move it with your mind.

[Beth]

[QUOTE]Originally posted by roger
[B]Hi Beth, welcome to the forum. I'm glad you've joined us.

Thank you. It is nice to be welcomed and you obviously put a lot of thought into your response.

I think I need to clear up a few misconceptions here. First of all, I don’t expect JREF to be doing this testing. That’s okay. I’m doing these experiments for my own benefit, simply because I wish to know with more certainty what might or might not be happening. I had submitted an application in the hopes of getting some help in the form of observers. That hasn’t happened, but I have gotten some good feedback on the experimental design, so that’s helped.

Second of all, I don't have access to a lab. My experiments are limited to what I can reasonably conduct at my dining room table. That’s a constraint I am working under and that won’t be changing in the foreseeable future.

Third, I don’t really care what’s being affected, whether it’s the heat, the flame, the wax, etc. Theoretically, I shouldn’t be able to affect any of that. What I’m interested in is determining whether I am having any effect or not.

You brought up a lot of good issues, but with the exception of the positioning of the ring on the glass, randomization should equalize the effect of all of those uncontrolled variables. That is, after all, the purpose of the randomization. If I can get observers, they would be placing the rings on the glass, without prior knowledge of which rings were tests and which controls.

Your suggestions, though theoretically good, are not feasible for me given the resources I have to work with.

Do you have any suggestions on how to improve on my current setup? Or a better setup that's feasible for me to implement?

[davefoc]

Beth,
Welcome to the forum. I hope you will participate in some of the other fora . The science forum has math and even statistic threads that occurr occasionaly that your expertise would be appreciated in.

Thank you for your comments.

I was attracted to KRAMER's report about your claim because it struck me as the kind of claim a skeptical objective person might make if they were reporting on a possible paranormal phenomena.

Yours is the first claim that has struck me like that in the several years that I have been participating in this forum.

I suggested above that one possible way to test your idea would be to use thermocouples. I am not quite sure how well this idea would work, but what I had in mind was to place one or two thermocouples above and just beside the flame. This way if the flame moved in the direction of the thermocouple a sudden increase in temperature would be detected. Thermocouples are readily available in an electronic supply store. You would need a meter to plug them into. It would be nice if the meter had a computer link so the temperature could be recorded throughout your experiment. Years ago I bought such a device for about $200. I don't have it right now or I would have fooled around with the idea. It might also work to put the thermocouple just above the flame and look for a sudden cooling.

Anyway, I wish you well with your experiment. I hesitate to say this because I don't wish to discourage you from looking into something that interests you, but I also must admit that I am a lifelong skeptic about things like this and I think it is very unlikely that there is any paranormal effect going on here. But if it helps at all, I have been wrong many many times in my 55 years so maybe there's room for one more time here.

Best Regards,
Dave

[69dodge]

Hi, Beth.

Not sure if you read the whole thread, but I mentioned earlier that I made myself a torsion balance by hanging a toothpick from a string inside an empty Coke bottle. Not terribly high-tech, but still remarkably sensitive. I estimated, from its frequency of oscillation, that a force of just 10 micrograms would deflect it noticeably. If you use thin sewing thread or something like that, instead of string, you could probably do even better.

Anyway, assuming you want to stick with the flame and wax ring: I think a good way to average the angles is not to think of them as angles at all. Instead think of the position of the wax drop as a unit vector. (The origin is at the ring's center). Averaging vectors is easy. Success is if the test average is closer to the target than the control average is, treating the target position as a vector too.

Actually, if you're using a binomial test, I'm not sure why you'd need to average anything. Is there any reason to have four controls and four tests per trial? Why not one and one? It's more powerful than spending eight rings on each trial.

On the other hand, I'd be wary of t-tests and such, which rely on normality. How can you tell the drops are normally distributed, from just four data points? Does it make sense to combine different trials' data if it appears, from the varying variance, that the data don't all come from the same distribution?

[davefoc]

I just realized that I hadn't read Beth's first post or Roger's response before I made my post.

The only thing I might add after reading those posts is that Roger's idea of a camera on the flame during the test seems like a reasonable idea to me. It is even less intrusive than the thermocouple idea and by hooking it up to a computer the flame location could be closely monitored for movement.

I also liked the idea of a gas flame for producing a flame with less of the confounding randomness of a candle. I was thinking of an oil lamp with a wick for the same reason.

I understand (at least somewhat) Beth's idea that the randomness of the candle flame and other effects can be dealt with by comparing different runs statistically to determine if the telekinesis is affecting the outcome, but I also like the idea of just making the experiment as simple as possible so that random variation is more easily eliminated.

[Beth]

Originally posted by davefoc
Beth,
Welcome to the forum. I hope you will participate in some of the other fora . The science forum has math and even statistic threads that occurr occasionaly that your expertise would be appreciated in.

Thank you for your comments.

I was attracted to KRAMER's report about your claim because it struck me as the kind of claim a skeptical objective person might make if they were reporting on a possible paranormal phenomena.

Yours is the first claim that has struck me like that in the several years that I have been participating in this forum.

I suggested above that one possible way to test your idea would be to use thermocouples. I am not quite sure how well this idea would work, but what I had in mind was to place one or two thermocouples above and just beside the flame. This way if the flame moved in the direction of the thermocouple a sudden increase in temperature would be detected. Thermocouples are readily available in an electronic supply store. You would need a meter to plug them into. It would be nice if the meter had a computer link so the temperature could be recorded throughout your experiment. Years ago I bought such a device for about $200. I don't have it right now or I would have fooled around with the idea. It might also work to put the thermocouple just above the flame and look for a sudden cooling.

Anyway, I wish you well with your experiment. I hesitate to say this because I don't wish to discourage you from looking into something that interests you, but I also must admit that I am a lifelong skeptic about things like this and I think it is very unlikely that there is any paranormal effect going on here. But if it helps at all, I have been wrong many many times in my 55 years so maybe there's room for one more time here.

Best Regards,
Dave

Thank for the welcome and the suggestions. I've been lurking in a few other threads and the discussions do seem interesting.

I like the idea of thermocouple. It would certainly speed up the testing process to have a continuous recording of the temperature at various points around the circumference of the flame. The wax rings are simply the best way, thus far, that I have found to measure objectively the direction of a flame.

Unfortunately, even though $200 is a modest expense, it's beyond what I can justify allocating for my experiments at this time. I'm also pretty clueless about how to set up and use such a device, so I think I'll table that idea for now. If I still get positive results after I've used observers and made other improvements, I'll reconsider it.

[Beth]

[QUOTE]Originally posted by davefoc
[B]I just realized that I hadn't read Beth's first post or Roger's response before I made my post.

The only thing I might add after reading those posts is that Roger's idea of a camera on the flame during the test seems like a reasonable idea to me. It is even less intrusive than the thermocouple idea and by hooking it up to a computer the flame location could be closely monitored for movement.

I also liked the idea of a gas flame for producing a flame with less of the confounding randomness of a candle. I was thinking of an oil lamp with a wick for the same reason.

I like the idea of a camera as well, but I'm not sure how the flame location could be analyzed, even with a digital recording. That is beyond my abilities. I've also considered using a camera to record where the first drop of wax falls from, but have the same problem. How do I get from a video recording to an accurate objective measurement?

An oil lamp with a wick is something that I should be able to set up in my dining room. I'm not sure I can manage a gas flame, though I am planning to look into it. Is there any particular reason to prefer gas over oil?

[Beth]

Originally posted by 69dodge
Hi, Beth.

Not sure if you read the whole thread, but I mentioned earlier that I made myself a torsion balance by hanging a toothpick from a string inside an empty Coke bottle. Not terribly high-tech, but still remarkably sensitive. I estimated, from its frequency of oscillation, that a force of just 10 micrograms would deflect it noticeably. If you use thin sewing thread or something like that, instead of string, you could probably do even better.


Anyway, assuming you want to stick with the flame and wax ring: I think a good way to average the angles is not to think of them as angles at all. Instead think of the position of the wax drop as a unit vector. (The origin is at the ring's center). Averaging vectors is easy. Success is if the test average is closer to the target than the control average is, treating the target position as a vector too.

Actually, if you're using a binomial test, I'm not sure why you'd need to average anything. Is there any reason to have four controls and four tests per trial? Why not one and one? It's more powerful than spending eight rings on each trial.

On the other hand, I'd be wary of t-tests and such, which rely on normality. How can you tell the drops are normally distributed, from just four data points? Does it make sense to combine different trials' data if it appears, from the varying variance, that the data don't all come from the same distribution?


I did read about the torsion balance, but at this point, it is the influencing of a flame that I wish to establish.

While it's true, I don't require the use of four controls and four tests for the binomiial analysis, using the means gives me with more confidence in the results. I feel I'm less likely to get a false positive. I had originally thought I could use a paired t-test on the data. Means will always follow a normal distribution if the underlying data is even approximately bell-shaped. It's the non-constant variance that's the problem. That can be dealt with using more sophisticated analytical techniques, but it's not easy and I'm not as confident in my ability to do that analysis without error. I'm hoping that if I can improve the uniformity of my set-up, I can eliminate that problem.

I've also considered using some nonparametric analytical methods, but since they will likely be less powerful than the binomial model I'm using, I haven't given that a great deal of thought.

[roger]

Okay, several things, mostly off the top of my head.

1) randomization does not control for many of the things I wrote about in my post. I reread it, and realized I left much implied. Any successful experiment must demonstrate that Beth is not moving the flame via physical means. So, for example, breathing on the glass during the test changes the boundry conditions that the air column inside the glass experiences. It is quite feasible that this could alter the flame's movement. Randomization will not correct for this. Please note that my assumption is that Beth is doing this unconsciously, but purposefully. Meaning, one way or another she behaves differently during a real trial vs. the control. She may lean closer, tap the table top because she is concentrating, or may have learned unconsciously how to move the flame.


2) It's hard to opine about this setup without studying what it does. So last night at 1230am I had a candle set in a short glass, and measured the temperature with a very sensitive J-style thermocouple. My set up was NOT the same as Beth's as I believe the glass was shorter, and I didn't have the wax ring. Nonetheless, I believe I gathered some suggestive data.

I believe what happens is air is sucked in from underneath and the side of the flame, gets heated by the candle, and raises above the flame. At the same time, the burning wick is generating gases, which also rise. I established this in 2 ways. First, I took a lot of temperature measurements around the flame. I could place the probe almost in the flame on the side, and I'd get reading in the 140 to 155 range (all temps F). Lifting up the probe slightly would quickly register higher values - up to 180 to 190, but no more. Secondly, the heat caused visual distortions in the surrounding air, and I was able to observe the flow to some extent with my eye.

As you move the probe above the flame, the temps increased rapidly. I backed off at 450F, as I didn't want to damage the probe. I got those readings about an inch above the flame.

Moving about 3 inches above the top of the flame, the readings were very erratic. In one second intervals, you might see: 148 173 179 199 223 257 221 199 178 170 158 ... (that data is completely made up).

In other words, very significant swings. The swings by far exceeded the movements of the flame, which was just dancing around a bit like a candle is wont to do.

Moving the probe just a bit off center, just a 1/2 inch, would result in a 100degree or more temperature drop. Basically there was a very narrow tunnel of hot gases rising, surrounded by a much cooler boundary of cool air.

However, this tunnel was very susceptible to small perturbations in the environment. For example, if I would very lightly tap the counter, the temperature would vary wildly. I didn't measure the force of the tap, but we are talking half way between just sitting down your finger without any force, and the kind of tap you might do unconsciously. This tap was not significant enough to produce visually obvious movements in the flame - sometimes you'd think you see a reaction, but sometimes not.

So, much as I suspected, Beth has constructed a device that vastly amplifies tiny movements in the flame. In a sense, this is good, because it's easier to measure amplified signals. However, the downside is that the amplifier is in no way shielded - it reacts just as wildly to small variations in the physical environment. Furthermore, the gain seems quite underdamped - a tiny impulse signal injected into the amplifier results in prolonged and vacillating output. Finally, the X/Y positioning of the output is rather small, even when the temperature swung wildly right over the flame, it really didn't seem to vary much at all just offset an inch or so.

As I see it, this has at least two negative consequences for Beth. First, if she is exhibiting a real, but very effect, it may very well be swamped by the noise in the system. Second, it makes it much harder to prove that any results are the result of her mind, because, despite the randomization of trials, she will very likely have ways to physically affect the flame unless a much more expensive apparatus is used. In short, with the wide and rapid fluctuations that I was seeing, I could never really tell if what I was doing was affecting the flame, or whether I was seeing the normal variations caused by the flickering flame.

3) Dave' thermocouple idea: I was basically thinking the same thing when I got out my thermocouple last night. If for example, you had a thermocouple in the North and South position, and instructed Beth to either move the flame N, S, or not at all, you might be able to record say 10 seconds worth of readings and average them. I would be impressed with, for example, if by thumping on the table, blowing at the flame, etc, you could never make the N sensor exceed 280F, but when Beth used her mind, she repeatedly was able to reach 300F or better. But if the data is essentially the same as the control runs, and requires significant statistical analysis to extract the results, then I would fear and assume that small physical forces have not been accounted for.

--------------
I'm musing about a different setup altogether. It just came to me, so even if it does work it'll need refinement. But it occurs to me that burning is a typical way to compute the caloric content of an item. So I'm wondering if it would be possible to say put two candles very close together in a chamber used for measuring caloric heat. Only one candle would be lit. Beth's task would be to move the candle towards the unlit candle, so it is partially vaporized and thus increased the caloric energy produced.

Thinking about it, I think it is still too sensitive to initial conditions - each candle will have a different caloric content within a certain rainge, and so Beth would have to exceed that output. I'm guessing her effect, if any, would be swamped by the normal variances of the candle.

[Prolix]

Beth, You said,

I did read about the torsion balance, but at this point, it is the influencing of a flame that I wish to establish.

I'm curious, and it might give you and others insight as to what this is all about, if you were to introspect a little and then reveal why your intuition leads you to test only flames. Of course intuition is an valid basis for making scientific hypotheses and experimental proposals, but the best experiments can be designed if the intuiter can describe her concepts in terms that she or someone else can operationalize.

I apologize if you have already done this in this thread.

[69dodge]

Originally posted by Beth Clarkson
While it's true, I don't require the use of four controls and four tests for the binomiial analysis, using the means gives me with more confidence in the results. I feel I'm less likely to get a false positive.Hmm ... I don't see it. On the null hypothesis, the probability of a success would seem to be 50%, either way. (And if not, we shouldn't be using a binomial test with p = 1/2, to begin with.)Means will always follow a normal distribution if the underlying data is even approximately bell-shaped.Well, almost always. A Cauchy distribution is bell-shaped too. But, yeah, it's usually true, in the limit, if we average enough values. (The underlying distribution doesn't even have to be bell-shaped; it can be anything at all, provided it has finite variance. Cauchy distributions have infinite variance.) However, the important question is, how many values are enough?

Trying to answer that question brings us back to the question of the "right" way to average angles.

If we have a uniform distribution on the interval [-180, 180) of the real line, and we average together a bunch of values taken from it, the distribution of the mean will approach a normal distribution centered on 0. On the other hand, if the wax drops are uniformly distributed around a circle, the average of a bunch of drops---for any reasonable definition of "average"---should likewise be uniformly distributed around the circle; it should not favor 0 degrees, nor any other position. Right? What else could the distribution of the mean be? If no position is special, then no position is special; averaging isn't magically going to pick out one position from all the other equivalent ones.

But how can this be? What happened to the central limit theorem? Averaging is supposed to reduce the variance and bring us closer to a normal distribution; yet here we end up with the same distribution we started with, no matter how many values we average!

The same thing happens if we average a bunch of Cauchy random variables; we end up with an identically distributed Cauchy random variable. There, we can explain it away by appealing to the infinite variance; reducing infinity by a factor of ten, or a hundred, or a thousand, leaves us still with infinity. But, here? Does a uniform distribution on a circle have infinite variance??

In a sense, yes. Variance measures how spread out a distribution is. Infinite variance means it's really, really spread out. A distribution on the real line that's uniform on the interval [-180, 180) can be further spread out, e.g., to [-360, 360). But a uniform distribution on a circle is the most spread out of any distribution on the circle. How could it possibly be further spread out? It already covers the whole circle.

Ok, so how can we define the "average" of wax drops in such a way that it will have the properties we want it to? We can think of the positions of the drops as two-dimensional vectors in a plane, and average them that way, instead of as one-dimensional positions on a circle. Then, the mean of many drops has a distribution that approaches a 2D Gaussian about the center of the circle---a real bell, if you will, instead of a cross section of one. The more drops we average, the narrower the Gaussian becomes. So, if we do things this way, the (2D) variance does decrease as expected.

Although each drop lies on the circle, the average of many drops usually won't. If, for some reason, we want it to, we can project outward from the center, back onto the circle. This procedure yields a uniform distribution of angles, as we earlier decided it should. And now we can see why the variance doesn't decrease if we want the average to lie on the circle. It did decrease, but then we somewhat artificially increased it again when we moved the average out to the circumference of the circle, farther from the center than it really was.

So far, I've been supposing that the drops are distributed uniformly around the circle of wax. If the flame is not precisely centered, so that the drop distribution around the circle is not precisely uniform, then the mean of a bunch of drop position vectors will approach a 2D normal distribution that's somewhat off-center. In this case, projecting outward from the center, back onto the circle, will yield a distribution of angles that approaches normality as the 2D distribution narrows; however, the better centered the flame is, the more drops need to be averaged, to approximate normality to any specified degree of closeness. Four drops might not be nearly enough.

Accurately centering the flame increases the sensitivity of the experiment; an infinitesimally small psychic deflection of a perfectly centered flame is sufficient to move the drop from one side of the ring completely to the opposite side, 180 degrees away. But, by the same token, a centered flame makes it easier for chance, too, to move the drop by a large angle, and we need to take this fact into account in the statistical analysis.

We take it into account by not assuming a normal distribution for the angle of the drops (or even, for the angle of the average of four drops) unless we're sure the distribution really is nearly normal. If the true distribution is more uniform than a normal distribution is, the true probability of getting a value far from the mean is higher than for the normal distribution, so using the normal distribution instead of the true distribution will yield p-values that are too significant.Trial to trial data seems to follow a uniform distribution (as would be expected). Deviations from the mean within a trial (looking just at control data to establish what would be typical) have a nice bell-shaped distribution. [...] The within trial variance is not constant from one trial to the next.I do not understand. Can you give more details? What is uniform and what is bell-shaped?

The four points of any single trial are just four points; that's not enough to figure out a shape.

If different trials have different distributions, as the different variances appear to indicate, of what use is combining data from different trials? What could such a combined distribution tell us about any single trial's distribution?

---------

To summarize, I think it's best to treat this whole thing as a 2D problem. It seems rather shaky to compute means and variances by treating angle measurements as if they were measurements of position on an infinitely long straight line, ignoring the difference in topology between a circle and a line.

[Beth]

Originally posted by Prolix
Beth, You said,



I'm curious, and it might give you and others insight as to what this is all about, if you were to introspect a little and then reveal why your intuition leads you to test only flames. Of course intuition is an valid basis for making scientific hypotheses and experimental proposals, but the best experiments can be designed if the intuiter can describe her concepts in terms that she or someone else can operationalize.

I apologize if you have already done this in this thread.

I am a rank beginner at this type of thing. A candle flame was recommended by my teacher as well as some books as being one the best things for a novice to start with and practice on.

In addition, candle making has been a hobby of mine for years, so I had lots of candles and different types of wax to try things out with. I actually started out with a substantially different set-up and have gradually evolved it into what I'm using now. I hope to continue improving it. I do not wish to attempt anything else until I have established with more certainty whether or not something is happening with the flame.

[Beth]

Originally posted by roger
Okay, several things, mostly off the top of my head.

2) It's hard to opine about this setup without studying what it does. So last night at 1230am I had a candle set in a short glass, and measured the temperature with a very sensitive J-style thermocouple. My set up was NOT the same as Beth's as I believe the glass was shorter, and I didn't have the wax ring. Nonetheless, I believe I gathered some suggestive data.


Wow! This is very informative. Thank you. Do you have any suggestions, based on this data, how I could improve my experimental set-up?

[Beth]

A lot of thought went into this post and you've given me a great deal to consider. Thanks. I'll try to answer the questions you've posed.

Originally posted by 69dodge
\Can you give more details? What is uniform and what is bell-shaped?

I've done basic histograms of the data. The general picture looks roughly uniform for the means of different trials, which is what I was expecting. A histogram looking at deviations from the means for the control data has a nice bell-shaped distribution. I can email you my data if you'd like to take a closer look at it.

\The four points of any single trial are just four points; that's not enough to figure out a shape.[/QUOTE]

Agree. That's why I was combining the data from different trials and looking at deviations from the mean.

\If different trials have different distributions, as the different variances appear to indicate, of what use is combining data from different trials? What could such a combined distribution tell us about any single trial's distribution?[/QUOTE]

It's simply a way to examine the data. Concerns about the validity of combining such data is why I am not relying on statistical tests that assume an underlying bell-shaped distribution.
---------

\To summarize, I think it's best to treat this whole thing as a 2D problem. It seems rather shaky to compute means and variances by treating angle measurements as if they were measurements of position on an infinitely long straight line, ignoring the difference in topology between a circle and a line. [/QUOTE]

I agree. I like the vector approach and will take a close look at implementing it.

[69dodge]

Originally posted by Beth Clarkson
I can email you my data if you'd like to take a closer look at it.Sure.

Drown me with data. Numb me with numbers. Stupify me with statistics.

Send stuff to yasakov%verizon*net.

(The strange symbols, and a superfluous 's', are there simply to snag silly spreaders of spam who strenuously strive to submerge me in ... something else that starts with 's'. [Hint: it's smelly.] Strike the 's' and substitute the standard signs.)

[davefoc]

Just a few thoughts on this.

A cigarette lighter might be an inexpensive source of a pretty stable flame.

On the thermocouple:
First, Roger mentioned J style thermocouple. This refers to the temperature range of the thermocouple itself and not the meter part of the instrument. Meters can use different type thermocouples that are identified with a letter.

Secondly, thermocouple meters that don't have a computer interface are much cheaper than $200 and are often just volt meters that have a thermocouple function.

I found Roger's experiments interesting. It seems like there were at least two significant results. One is that a candle flame is highly random and two a candle flame can be moved with very small forces. This seems to be what Beth has found already, hence the difficulty with the experiment. The thermocouple experiments reinforced this idea for me.

If I understood the results correctly it does seem that testing with thermocouples might be a useful approach.

1. Create a setup where the candle flame is as isolated as much as possible from physical disturbances caused by the individual under test.

2. Place the thermocouples so as to be able to detect flame movement.

3. Characterize the test setup as to how well isolated from the testee the flame is. For instance what effect does gentle tapping have? what effect does gentle blowing have?

4. Then during the test through observation validate that the testee is physically impacting the setup less than was done by the physical tests above so that simple mechanical forces originating from the testee can be ruled out as a source of force on the candle flame.

5. Look for a thermocouple variation greater than what was observed during step 3 above while psychic force is being applied. Alternatively look for more frequent variation when psychic force is being applied.

[roger]

Originally posted by Beth Clarkson
Wow! This is very informative. Thank you. Do you have any suggestions, based on this data, how I could improve my experimental set-up? Not exactly. I'm a big believer in the emperical method, at least for myself, as I find my predictions of how something will work is rarely reflected in how they actually do work. I.e., I'm not so good at guessing. :) I guess if I was setting up the experiment, I would do much as Dave suggests - set up the apparatus, then apply known forces (puffs of air, holding glass with hand, tapping table or floor, etc), and working to eliminate their effect on the flame.

By the way, does this still work if you are, say, 10 ft away? A 10 or 20 ft separation of you from the candle would do alot to allay concerns that you are using phsical means. 30 or 40 feet would be even better. I understand that with the current set up you need to see the wax fall, so you may be limited in how far you can be away from it.

If this ever gets tested by Randi, expect lots of very light and low density objects to be strewn all about the candle, to detect any sudden gusts of air. We tend to get arcane with our tests; Randi cuts to the chase. So sprinkle some styrofoam or such around the glass - if any of it moves during a test, you did something. Discard the results, even if you feel confident it didn't affect the flames. If at all possible, but styrofoam right on top of the wax disk.

The thermocouple that I referred to did in fact come with a multimeter. I'm set up here to do accurate temperature readings for an entirely different reason, but you can get this for, I don't know, $90 or so. Mine's an Extech from Radio Shack. You may very well find a cheaper source. You may also know somebody that you can borrow one from.


I would suggest 1 addition to your apparatus, if you are staying with the wax disks. Make a hood to go over the top of the glass. It should function like a baffle, and make it harder for gusts of air to get into the chamber.


edited to add:
I would also offer the additional requirements while testing yourself. You may already do this. Sit far enough away so you cannot touch the table. Sit in a heavy, soft, stable cushioned chair, and sit with your feet off the floor and tucked in the chair (so you can't tap your feet, or otherwise cause vibrations). If you need to be close, put on a paper mask so you are less likely to affect things with your breath. Get four soft foam squares, and put the squares on the table, set a tray on the squares, and then the candle/glass on the tray. All this, along with the hood, and strewn about styrofoam, should take you far towards eliminating unwanted physical inputs into the system.

[Ashles]

Hi Beth,

Welcome to the forum. It is so much better if we can get the chance to ask you these questions directly and not have to try and assume the answers (which we are often going to get wrong:) ).

I am certainly not the expert in statistics that many of the other posters here appear to be (and you obviously are) but I do know a little about psychology and have a couple of questions more on that front.

Firstly, you appear to admit that you don't have any evidence or reason to think you actually do have this ability.
You say:
I am a rank beginner at this type of thing. A candle flame was recommended by my teacher as well as some books as being one the best things for a novice to start with and practice on.
I am assuming that you have recently developed an interest in the paranormal or have come to encounter some people who have opened you to these types of possibilities.
What I am wondering about is the reasons behind you suddenly wanting to develop a paranormal ability (obviously I do not expect you to reveal any personal reasons on an internet forum). My concern about all of this is that you have never had reason to believe you have any of these abilities, but you seem to be being told you DO have these abilities by other people.
As a result you are expending a lot of your own time, energy, and no doubt money in pursuing this.

I am wondering if you have been in a certain state of mind which would be 'receptive' to this kind of influence or suggestion. Many people find themselves talking to mediums, sensitives, healers, spiritual teachers etc. at times in their life which are stressful or traumatic. Often these people try to help in the sense of being a shoulder to cry on, and they act in the position of amateur councillor or analyst and allow you to unburden yourself and hopefully to a certain extent come to terms with the issues.
But sometimes they try and make the person feel better about themselves by claiming they have "the gift" or some paranormal ability.)
On the surface this has two benefits (as far as the amateur councillor sees it) - firstly, of displacing the real issues that the person is dealing with and replacing them with a new and very involved interest - the paranormal.
Secondly it raises self-esteem.
Suddenly the person may have a special ability or insight. (It might even be suggested that this is why they have had difficulty with interpersonal relationships before.
It all gets very complicated. The real problems, of course, are still there and have not been dealt with by a trained councillor/therapist. An over-involvement in the displacement activity may help keep the problems at bay, but probably not indefinitely. Plus some sense of the person's identity and self-esteem is now tied up with an ability that may well not exist.
What if they can't really do what they have been told they can? There is often by this point a very high level of refusal to accept negative results. Surely more trials will prove it. Surely if I test in a different way?
Otherwise, of course, a fragile belief and coping system will subsequently collapse.

Regarding the ability itself leads me to my second question.
As a statistician, what would you say the odds were of developing a paranormal ability that can only be detected by statistical analysis, your very field?
You have been told that moving a candle flame is a good place to start - this would imply that this has been successful before. Have people with this ability always used statistical analysis to analyse their results? (If so none of us have ever read that research)
And if the effect is big enough to be detected without statistical analysis, again, why have we never heard of this? Why has no-one else ever applied to this challenge with this ability. Is your teacher not implying that it is at least reasonably common. Can he not do it?
What is your teacher's background and what abilities does he himself have that can demonstrate these abilities? What reasons does he himself have to believe in the paranormal.

Having said all of that these are just illustrative examples Beth and I hope none of it applies to you. For all we know you may well have this ability.
But do you feel it is really worth all this time to try and convince yourself of an ability you have zero evidence to believe exists except that someone else tells you so?

I am just asking that you have a brief step back and look a the whole situation calmly and logically, taking my questions into account. You certainly appear like a very intelligent, articulate and thoughtful person, and I hope you don't think I am saying these things apply to you, merely pointing out that they can sometimes apply in these situations and it's best to try and ensure that they definitely don't apply to you before going further.

If you have a good think about this whole test (a deep breath and a short pause from formulating the detailed and elaborate analysis), and believe that you are not trying to displace other problems, that you believe in your teacher, that you believe that paranormal activity can exist at a decetable level, then I wish you all the best.

Good luck with the challenge and please keep us updated as to how it is all going.

[IXP]

Welcome Beth,

I have followed the discussion of your claim from the start and have been fascinated by it because of the difference from all other claims I have seen. Unlike the others, who generally cannot even express what "ability" they have in a clear way, you are quite specific and you understand what it takes to validate such a claim.

Most of the people who have responded to your request here have concentrated on controlling the physical properties and on the statistical analysis of your tests.

Despite the concerns that others have expressed here, I seriously doubt that you have, either consciously or unconsciously, come up with a method to move the flame in an arbitrary direction by either physical or paranormal means.

I would like to explore more plausible source of error, which is unconscious bias in the setup of the experiment and the interpretation of the results. I will suggest a method that should reduce though not eliminate possible error from these sources. This method is barely diffent than yours and does not require any more instruments or controlled conditions, so it is something you can try immediatly with no extra expense.

Like others, I not like the idea of measuring and averaging angles. A simple change to the method can turn it into an equal probability binary test. Divide the disk into two semicircles by a diameter at a randomly chosen angle. Then the only observation required is to determine which side the drop fell from, the target side or not, and simple binomial statistics apply. Note that no averaging and no control trials are needed with this method so you get 8 times as many trials for the same effort as with your current method.

It would be much better if you can have an independent observer set up the apparatus, choose the random targets and interpret the results. But if you must do all this yourself, I would suggest the following procedure:

Before starting, mark the glass by dividing the circle of the top into 6 equal numbered sections. The marks should be on the side of the glass so that they can be seen when the wax ring is in place. Also draw a circle on a piece of paper the size of the glass bottom and mark and number six equal sized sections on it as well. Place this paper where the glass still stand.

For each trial:

1. Set the candle in the glass and light it.
2. Place the wax ring on the glass.
3. Roll a die and line up the number 1 section on the glass with the section on the paper selected by the die roll. Do not touch any of the apparatus after this step.
4. Roll a die to determine the number of the target section.
5. Perform your mental exercise to push the flame toward the target.
6. When the first drop falls it is a success if it falls in the target section or either adjacent section, a failure if it falls from any of the 3 sections furthest from the target. Bias can enter here, so try very hard to be objective. This is where an observer blinded to the target direction would be a great help.
7. Extinguish and remove the candle and allow everything to cool for a while between trials. Ideally use a new candle for each trial.

If you can affect the flame about 1 in 3 times as you have stated, you should get a 2/3 success rate. If you get 62 or better out of 93 trials this establishes your ability to a p value of < 0.001, which is required to pass a preliminary test.

[Beth]

Originally posted by davefoc



Just a few thoughts on this.

A cigarette lighter might be an inexpensive source of a pretty stable flame.



Thank you for taking the time to consider my problem. I am considering a cigaratte lighter as an inexpensive source of a stable flame. I need something that will allow me to uniformly position the flame. Depending on the set-up I end up using, a lighter may well be the best choice. They are certainly available in a wide variety of the shapes and sizes.

I like the idea of thermocouples as well, but I will need to research what's available and the cost before I make a decision.

Thanks for your suggestions.

Beth Clarkson

[Beth]

Originally posted by roger
Not exactly. I'm a big believer in the emperical method, at least for myself, as I find my predictions of how something will work is rarely reflected in how they actually do work. I.e., I'm not so good at guessing.

Ah, a man after my own heart! I'm not so good at guessing or predictions either and find actual experimentation to be essential.

By the way, does this still work if you are, say, 10 ft away?

I'm afraid not.

If this ever gets tested by Randi, expect lots of very light and low density objects to be strewn all about the candle, to detect any sudden gusts of air. We tend to get arcane with our tests; Randi cuts to the chase. So sprinkle some styrofoam or such around the glass - if any of it moves during a test, you did something. Discard the results, even if you feel confident it didn't affect the flames. If at all possible, but styrofoam right on top of the wax disk.


This is an interesting idea. I'll have to give it a try when I start up my experiments again.

I would suggest 1 addition to your apparatus, if you are staying with the wax disks. Make a hood to go over the top of the glass. It should function like a baffle, and make it harder for gusts of air to get into the chamber.

Such a set up has occurred to me, but unfortunately, I haven't come up with any way to implement such a hood. Any ideas on how to construct such a thing inexpensively?

THanks for all your good suggestions.

Beth Clarkson

[Beth]

Originally posted by Ashles
Hi Beth,

Welcome to the forum. It is so much better if we can get the chance to ask you these questions directly and not have to try and assume the answers (which we are often going to get wrong:) ).

Thank you for the welcome. I was quite shy about entering into the discussion and it’s been gratifying to be welcomed this way by so many different people. I will try to answer the main points you have brought up here.

I am certainly not the expert in statistics that many of the other posters here appear to be (and you obviously are) but I do know a little about psychology and have a couple of questions more on that front.

Firstly, you appear to admit that you don't have any evidence or reason to think you actually do have this ability.


If you accept the premise that such an ability can exist, then (and this is according to several different sources) the ability is not limited to a few people who were born with it, but is a skill that can be developed by anyone who is willing to spend the time and effort to do so. I suspect that like any other skill, such as throwing a ball through a hoop or hitting the right keys on a piano, some people are born with natural talents that make it much easier for them to develop such skills, but basically anyone who wants to can do so.
I am wondering if you have been in a certain state of mind which would be 'receptive' to this kind of influence or suggestion.

You raise a valid point in regard to background issues. However, I do not feel this is the case nor is it something I wish to discuss in a public forum.


Regarding the ability itself leads me to my second question.
As a statistician, what would you say the odds were of developing a paranormal ability that can only be detected by statistical analysis, your very field?
You have been told that moving a candle flame is a good place to start - this would imply that this has been successful before. Have people with this ability always used statistical analysis to analyse their results? (If so none of us have ever read that research)
And if the effect is big enough to be detected without statistical analysis, again, why have we never heard of this? Why has no-one else ever applied to this challenge with this ability.


Actually, I have never heard of anyone conducting a test of the type I have devised. Most people who attempt it judge their success by the visual appearance of the flames. I don’t consider that adequate.

Further, I would expect that, especially in the beginning, any effect I might have would be small. So how could I objectively detect success if, indeed, there were any to detect? That question is what led me to start setting up tests. My initial tests were quite crude by comparison to the complex protocol we are now discussing, but my early results were consistent with hypothesis that I was having an effect. If those tests had been negative, I would have ended my experimentation then. Instead, I tried to devise a better test. With every improvement to my tests, I have continued to have results consistent with having a small effect.

Now, there are certain issues that I cannot resolve alone – most obvious being the fact that I cannot blind myself to my own experiment. Thus, I was faced with the dilemma of either seeking help from others or abandoning my tests. I chose to seek help. I appreciate the comments and questions I have received here, from yourself and others, in order to better determine what is actually occurring.

But do you feel it is really worth all this time to try and convince yourself of an ability you have zero evidence to believe exists except that someone else tells you so?

While you may find it hard to believe, I have not been trying to convince myself of any particular belief. I am trying to objectively determine for myself what is actually occurring. I have, at various times in my past, been both a believer and a skeptic. At the moment, I am undecided and my tests are an attempt to make a more informed choice regarding which is most consistent with reality as I experience it.

I decided to test this particular aspect of psychic phenomenon because it is more suitable to testing than many others and because I did not require the assistance of any other individuals to conduct preliminary tests. To the best of my ability to set up and conduct an objective test alone in my home, the data I have gathered is consistent with the hypothesis of the existence of such an ability. That doesn’t mean it actually exists, but I intend to continue testing until I have more certainty one way or the other.

Good luck with the challenge and please keep us updated as to how it is all going.

Thank you for your good wishes.

Beth

[Beth]

[QUOTE]Originally posted by IXP
Welcome Beth,

I have followed the discussion of your claim from the start and have been fascinated by it because of the difference from all other claims I have seen. Unlike the others, who generally cannot even express what "ability" they have in a clear way, you are quite specific and you understand what it takes to validate such a claim.

Thank you for taking the time to consider my problem and make some suggestions.

I would like to explore more plausible source of error, which is unconscious bias in the setup of the experiment and the interpretation of the results. I will suggest a method that should reduce though not eliminate possible error from these sources. This method is barely diffent than yours and does not require any more instruments or controlled conditions, so it is something you can try immediatly with no extra expense.

Like others, I not like the idea of measuring and averaging angles. A simple change to the method can turn it into an equal probability binary test. Divide the disk into two semicircles by a diameter at a randomly chosen angle. Then the only observation required is to determine which side the drop fell from, the target side or not, and simple binomial statistics apply. Note that no averaging and no control trials are needed with this method so you get 8 times as many trials for the same effort as with your current method.



The problem here is that within a set-up (i.e. a particular candle) where melting first occurs seems to be specific to that candle and not entirely random. The wicks are not perfectly centered within the candles. It does seem to be random from one candle to the next. Hence, I find it works much better to take both control and test measurements with each candle and make a comparison of the two.

I prefer to take several measurement with each candle and find the mean rather than a single measurement. There are a number of reasons for this, but primarily if I can work with means rather than single data values, I have the advantage of dealing with a smaller variance and a known distribution for the means. Further, because I randomize the order of test and control measurements for each trial, this equalizes most of the uncontrolled factors and allows me to have more confidence in the comparison.

However, I did like your suggestions. They seem quite reasonable and appropriate and I appreciate your taking the time to write then down for me.

Thank you.

Beth

[roger]

Originally posted by Beth Clarkson
Such a set up has occurred to me, but unfortunately, I haven't come up with any way to implement such a hood. Any ideas on how to construct such a thing inexpensively? I'd look around for what I had on hand. I'd probably take a larger glass, and then rig up some kind of stand to hold it upside down. Just some coat hanger wire, or maybe 3 wood blocks set around the glass with the candle on it, and the larger glass lowered over the top with the rim setting on the blocks. You can easily do this for under $5, I would think.

[Stereolab]

Beth,

Just wanted to mention that I'm rooting for ya! :)

[Ashles]

Beth, it does appear that you have arrived at the flame test quite arbitrarily. You want to test your abilities in some way so why pick the flame test?
If you are testing for paranormal ability in general and have just picked the flame one almost at random then would it not make more sense to pick another instead?
If you are generating some form of miniscule force then this would be much more straightforward to test (as mentioned by other posters).
Tests could be designed quite easily for far fainter applications of force or air movement than the candle flame.

Why must it be a candle? Why could you not try to demonstrate an ability that would be easier to measure, but would require less actual force to display?

I am really confused, especally as you are approaching this whole field in the complete opposite way to most scientific investigations (an effect is observed or described, THEN the research to analyse it is conducted).
Why are you so tied into the flame ability when you are investigating a broader description of paranormal abilities?

For example at school we had to make a microbalance out of a drinking straw, a small screw and 2 pins. It was unbelievably sensitive. Sealed in a box it would be a very impressive and unequivocable demonstration. (Diagram attached)
And it would have a clear binary choice for movement. Up or down.
The statistics for that one would be dead simple.

And if you are only willing to be tested on flames can we ask why? If you have no evidence that you have this ability in particular over any other ability, we are you tied into it so firmly?

Regards,

Ash.

(Please have a try of my Microbalance - maybe it'll work for you and simplify all this testing)

[Ashles]

Okay, I know it looks a bit crap, but it really is amazingly sensitive.

[Beth]

[QUOTE]Originally posted by Ashles
Beth, it does appear that you have arrived at the flame test quite arbitrarily. You want to test your abilities in some way so why pick the flame test?

Well, I wouldn't call my reasons for selecting the flame arbitrary, but certainly I could have chosen something else. However, my reasons for choosing the flame in the first place are not the same as the reasons for not switching to something else at this point. Those reasons are not arbitrary, but they are personal and I do not wish to discuss them in a public forum like this.

Thank you very much for your suggestion. If, at some point in the future, I decide to extend my experimentation to include something other than a flame, I will certainly keep it in mind.

Beth

[KRAMER]

Welcome, Beth! It's great to see you here, and it's equally great to see such wide discourse with an applicant, which is so rare as to be practically non-existent.

Kudos to all forum members who welcomed Beth with open arms, as I'd promised her.

[Ashles]

It's just because my diagram's rubbish isn't it?:(

But seriously, nobody expects or even wants you to provide any personal information that you aren't totally comfortable with so if there are other reasons that you don't wish to change abilities then that's obviously totally your right.

I was only asking because as it stands it may be an over elaborate test for the JREF prize.

But we will certainly try to help you with your experiment for your own interest. Maybe you could try contacting a sceptical group in your own area and seeing if any of them have any ideas, or are willing to devote some time and equipment to the study.
Or a university might be interested.
But it is obviously important to have impartial (or actively sceptical) observers as there is going to have to be an element of judgement involved in the test, no matter how small.

[Dr Adequate]

Originally posted by Ashles
But it is obviously important to have impartial (or actively sceptical) observers as there is going to have to be an element of judgement involved in the test, no matter how small.
I think we have to eliminate it altogether. Having sceptical observers wouldn't help accuracy --- there's a reason why the sheep-and-goats effect is called "the sheep-and-goats effect" and not "the sheep effect".

[Dr Adequate]

Beth --- welcome to the forums, we got fun and games. Do have a look about on the other JREF forums, there's a lot of smart, funny people round here. Oh, and a guy called Interesting Ian...

About the candle flame --- I never tire of referring people to the story of N-rays (http://skepdic.com/blondlot.html). Everyone should be taught this at school. Have a little read of the article. (There's quite a lot of other stuff to be found on the web about this, if you google.) These were very smart people, they were professional scientists, and in some cases distinguished scientists whose other contributions to science were important and valuable --- who were using their own judgement to see if a screen in a darkened room was getting slightly darker or slightly brighter (the only way of detecting N-rays).. They saw what they expected to see. (It's a myth that scientists are trained to be "objective observers". It would be great if you could, but no-one knows how.) There was no real effect --- it was entirely subjective. N-rays don't exist. Being sceptical about stuff, I suspect that something of the kind may be going on here, but obviously only a rigorous test will find out.

[Ashles]

What I meant was if you are testing yourself then on some of the drops you might start thinking "Oh that was near the line, we'll call that a hit" whereas what you really need is others judging who will stick to the rules.

I know this sounds sort of obvious, but it seems like Beth might be testing herself or possibly having some rather uncritical judges and that, as you so Adequately say, needs to be eliminated.

[Dr Adequate]

Originally posted by Ashles
What I meant was if you are testing yourself then on some of the drops you might start thinking "Oh that was near the line, we'll call that a hit" whereas what you really need is others judging who will stick to the rules.
Well that would be nice. But as I said, it's called the sheep-and-goats effect. Sceptics are equally prone to errors of judgement in trials like these --- just in the opposite direction. Everyone's prone to this. Half the scientific method is there just to make up for the fact that you can't trust scientists to see what's there or to say what they see. Well, that goes for sceptics too --- scepticism doesn't come along with some magic power of being objective, just with a realisation that judgement is subjective and that you need to control for this. So you need someone judging who makes up their mind how the drop of wax fell who doesn't know in what direction Beth was trying to influence the flame (which means that that person should not be present at the test).

[Skeptical Greg]

Wouldn't candle flames tend to be more chaotic, due to variation in the composition of the candles?


Wouldn't it seem reasonable to ask yourself " Why should this effect have more influence over a candle flame than the flame of an alcohol lamp? " ; which will burn more steadily than a candle..

[Beth]

I just wanted to say thanks to everyone who's been participating in this thread, donating their time and thought to helping me improve my experiments. You've given me some excellent ideas and a lot of things to consider. I won't be able to implement all the ideas, but I'll be giving a lot of thought to what I've learned here as I make my choices about the design of my next set of experiments.

If I continue to have interesting results, I'll post there here. I anticipate that it will take several weeks to several months of testing to generate adequate results. If I don't post anything, you can presume that the improved experimental techniques led to the expected results.

Beth Clarkson

[Ashles]

Your welcome Beth. Good luck with your experimentation.

But if the results do come back negative please don't let your 'teacher' make out it was somehow your fault, or telling you that you have to start trying a new ability instead - you could waste years going down that road, and it never seems to lead anywhere productive.

Try and be as sceptical of this teacher as you would be of anything else that was claimed but there was no evidence for.


Of course if you have positive results then I'm sure we'll be hearing from you again. :)

It'd be great to hear from you either way.

All the best.

[Taed]

Originally posted by B.S
I can't imagine a more useless ability.

That's funny because the comic book _Rising Stars_ has a character with exactly that ability -- she can only affect very small objects in a very small way from a distance. She ends up becoming an assassin for her government; she would "pinch" the vein that goes to the target's brain, and the target would soon die. No murder weapon, it appears to be a natural death, and she's far away from the target.

I guess it's not the size, but how you use it...

[B.S]

Well, thats definitely an interesting ability. I can think of other things that could be pinched from a distance ... but maybe thats not a good idea.

Seriously though, My intent was to point out that if this minor ability takes this much time and work to even just barely see a tiny result, what would it take to move an object that just weighed a pound? I couldn't imagine the work that would require. We would die of old age before we could train any really useful ability, and might as well spend our time doing something worthwhile. (Like writing comics).

[Sujay]

Well, well... I had a look at the protocol,and I must say, I am thoroughly IMPRESSED by what Ms. Clarkson offered.
After reading it carefully , I understood that the protocol is not trying to allow chance to come into play. It is trying to completely ELIMINATE it!!
Yes! If the test is executed according to the protocol suggested by Ms. Clarkson, and if she succeeds, the probability that it is a fluke is close to ZERO, nada, nil!!!

Mr. Kramer, there seems to be some confusion on what she said. She has indicated she is able to 'influence the flame', one in three times. This does not indicate her success rate.

Let me try to explain. Now suppose I believe that I have some 'powers' which can 'influnce' the spinning of the coin, and make it fall on the ground with HEADS facing upwards, ALWAYS.

Suppose I devise a test where in I spin a coin SIX times. Because I can 'influence' the spinning of the coin, I believe the coin should fall 'heads' facing upwards all the time. I spin the coin TWICE, three times.

Try 1 :

Heads and Tails.

In this mini-test its one Heads and one Tails i.e. 50-50, and not 100% Heads, so I concur that I could not 'influence' the coin.

Try 2 :

Heads and Tails

Again, it is WITH chance, so I concur my powers did not work again.

Try 3 :

Heads and Heads.

This time its Heads both times (100%), so from my perspective I was able to 'influence' the coin, into BREAKING the laws of chance. But I was able to do so just ONE in three tries. Objectively though, the 'hit rate' was higher than chance, at 66%. However, this could well be a 'fluke', and I could get a hit-rate of 33% next time.

However, assuming I do the test a HUNDRED times. And I find that ALL hundred times, the strike-rate is ALWAYS 66%. Is it still chance? Assuming I do it for a THOUSAND times, and it is ALWAYS 66% ALL the thousand times, is it still chance? NO it is not. But yet, according to me, I am able to influence the coin just ONE in three times.

Is it still chance, if my strike rate is ALWAYS (and I mean ALWAYS) more than 50%? May be not much more, may be JUST above chance, but still, ALWAYS more? No it is not.

Similarly, Ms. Clarkson claims that she claims JUST above chance (actually higher : between 66% and 75%?), but ALWAYS does so.
And in order to prove for SURE that she ALWAYS does so, she has to repeat the test MANY times.

To Ms. Clarkson, how could you reduce the number of tests?
Hmm its tough to say, but you try to beat the chance of a fluke by nearly 99.999%, and I'd say that is too high. I am sure the JREF folk would settle for a much lower number (Even 95% maybe? :o)

Also, as she suggested, if ONE or TWO more volunteers could participate, there would be more statisitcs to compare, and it would happen quicker.
Even in the absence of volunteers. Just use similar apparatus. Three identical glass tumblers, three indenctical candles, and identical wax rings, and set them all up in the same way, but let her apply her 'powers' on just ONE of the three tumblers. Then compare the statisitics across the the the three tumblers, and compare how 'unique' the statistics of her tumblers are, to the point of identifying whether they are nearly impossible.

[KRAMER]

I have to say that I'm with Randi on this. The suggested protocols are WAY too complex for such a simple claim.

The claim is that she can influence the movement of a candle flame. Simple claim, simply tested.

Here is Randi's response:

==============================================

The use of wax rings is unrealistic. I suggest a video, in which she is given commands to make the flame move in a specific direction, all decided by random means. The instructions go on the audio track.

Independent persons then view the tape without the audio, and decide where and when she was instructed to move the flame.

-JAMES RANDI