Beth Clarkson's application

[roger]

As far as the position of the flame vs. heat convection: influencing either by telekenesis would be equally valid as a paranormal claim. The claim should be changed from "I can influence the position of a flame," to "I can influence the position of the first drip of wax from a ring of wax postioned over a candle flame."

Yes, but this is a terrible, terrible setup for testing for heat convection. Unfortunately, that means it's also terrible for testing flame movement.

If you want to test for movement of a flame, you need to measure the movement, or control for all the confounding variables associated with the various, and rather complex thermal behavior of this apparatus. Frankly, I doubt that it can be controlled for, except with very high sigma QC of the candles, rings, etc., and then a heck of a lot of control studies to show that the remaining variables are not influencing the results.

Now, one can argue that so long as you compare control vs actual trials, and they are randomized and double blind, and that you can show a statistically significant result, the biases don't matter. I have a problem with that - that her control runs show a normal distribution rather than a uniform distribution is *not* what you would expect from a properly set up experiment. We know something is wrong, but what? Until you know that, it's pretty hard to say that because you randomized control vs real runs, etc., that your experiment is sound. I agree she is showing an effect, but what kind of effect. Physical, I'd wager. We need to know.

 

[IXP]

Yes, but this is a terrible, terrible setup for testing for heat convection. Unfortunately, that means it's also terrible for testing flame movement.

know.

But you are assuming the mechanism is by flame movement or heat convection. Perhaps her telekenetic effect is on the molecules in the wax ring, increasing their energy to the melting point before the rest of the ring. Perhaps the candle is even extraneous. The JREF states that it is not interested in mechanisms, only in the effect to be demonstrated.

Don't get me wrong, I do not think there is any paranormal effect here. And I was the one who pointed out the the normal distribution is wrong for the controls. But the bias can be controlled for by using random targets and still be statistically evaluated. Has there been ANY preliminary test where all biases, no matter how small have been accounted for? I doubt it.

I too would favor a better experiment. I suggested one earlier, that at least make the result clear without subjective judging, but it is subject to some of the same problems you have mentioned, such as the uniformity of the ring, air currents, and centering of the flame.

What I expect to happen is that some protocol very different than hers will be adopted and then she will say she was tested for something she could not do instead of her original claim, even though she had agreed to the new protocol.

 

[roger]

But the bias can be controlled for by using random targets and still be statistically evaluated. Has there been ANY preliminary test where all biases, no matter how small have been accounted for? I doubt it.

I disagree that randomizing the trials is sufficient to account for the biases. Some biases, yes, and that is probably what you are referring to. Not all of them. For example, suppose that this experiment works because she unconsciously taps her foot on the floor, and this is enough to jiggle the flame. Furthermore, suppose that she *always* taps her foot, just differently, thus providing the two different normal distributions for the control and actual trials.

Now the vibration issue is something that I trust the JREF to deal with. If it is something obvious like this, than her 'ability' will disapear as soon as Kramer and Randi tighten up the protocol. This is what I think happens with all the other tests that aren't purposeful cheating - it's a relatively simple physical process, and a so-so experimental design is enough to stop the claimant from doing what they claim.

However, if they didn't detect that, than the statistical analysis is completely irrelevant. All the statistics show is that there is a statistically significant difference between control and actual trials; it does not show that there was a paranormal effect.

I'm sure that I'm not telling you something you don't already know or think ; I'm just fleshing out the ideas. We're clearly in agreement and on the same page on this.

Anyway, I certainly agree that most of the JREF trials are not spectacular examples of perfect scientific design. They make reasoned guesses of what is causing the effect, make a decent protocol that controls for it, and sees what happens. If the claimant can't produce results, as is normal, everyone goes home. If they can, then it bares a bit more investigation, and attempts to control more factors. And that's fine, because they are not doing science (trying to find the exact mechanism), but trying to determine if the claimant can do what they say.

I'm just concerned about this apparatus because there are *so many* ways for it to be biased, but the reality is probably she is exploiting just one of them, and, as usual Randi et. al., will quickly discover it and move on.

Actually, what I really hope is that Beth Clarkson is lurking on the forum and reading this. You out there, Beth? Unlike so many of the applicants, she is clearly both intelligent and lucid. I'm hoping that she will consider some of the many possible flaws in her current set up, and that she will rethink her tenative conclusion that something is happening here. She's risking tenure, her marriage, etc., over a disk of wax that may be off center 1 degree, or a candle that isn't standing straight, etc. It's perhaps an indictment of our educational system that you can get into grad school in the math and sciences without being able to set up a good experiment, or at least recognize that it is lacking. Maybe she can, normally, but her beliefs are clouding her judgement? I dunno.

 

[drkitten]

It's perhaps an indictment of our educational system that you can get into grad school in the math and sciences without being able to set up a good experiment, or at least recognize that it is lacking.

Why should a graduate student in mathematics be expected to set up a good experiment? The last time I checked, mathematics wasn't an experimental science. Do you also expect your physics graduate students to play the bagpipes well?

 

[IXP]

Why should a graduate student in mathematics be expected to set up a good experiment? The last time I checked, mathematics wasn't an experimental science. Do you also expect your physics graduate students to play the bagpipes well?

What, you didn't learn experimental mathematics in school?

Proposition: All positive number are prime.

1, yep
2, yep
3, yep
4, must be an anomoly, we'll discard it
5, yep

Confirmed, all numbers are prime.

 

[SpaceFluffer]

Ahh...proof by lazy induction.

 

[IXP]

I disagree that randomizing the trials is sufficient to account for the biases. Some biases, yes, and that is probably what you are referring to. Not all of them. For example, suppose that this experiment works because she unconsciously taps her foot on the floor, and this is enough to jiggle the flame. Furthermore, suppose that she *always* taps her foot, just differently, thus providing the two different normal distributions for the control and actual trials.

The subject tapping her foot is not a bias in the experimental setup. I did mention that she must be isolated from the apparatus, preferably by being in another room.

I guess where we disagree is in how far the protocol has to go to control everything. I am fairly confident that as long as the subject does not set up the apparatus and is isolated enough to eliminate any air movements she might cause to affect the flame, both the controls and trials will come out consistant with chance.

I think it would be helpful in a case like this if the challenge provided for a demonstration given by the claimant on their own terms. This should provide enough info to put the right safeguards in place for the preliminary trial. I would love to see a videotape of such a demonstration.

 

[roger]

Why should a graduate student in mathematics be expected to set up a good experiment? The last time I checked, mathematics wasn't an experimental science. Do you also expect your physics graduate students to play the bagpipes well?

I was not able to get out of high school without doing a number of experiments in chemistry, physics, etc. I can't think of any way I could have gotten into college as a math major, if I wanted to, without taking those courses. Now, I freely admit that a lot of what we did was dictated by the teacher, but we were also expected to figure out why things weren't going the way there were supposed to (they never did), and we did have to do some of our own design.

Now, I wouldn't expect that anyone getting through HS and college to be *competent* at setting up any random experiment; but it is my fond wish (note I said 'perhaps' in my post; I added that after reflecting on the fact that my statement was overstated) that schools would teach critical thought, and that she and everyone would get enough out of their science classes to at least have the thought "perhaps this experiment doesn't have enough controls", even if they can't figure out how to do it.

A fond wish, but not one I think will be fulfilled!

However, I have higher expectations for a statistics student. Now, one can certainly spend their lives in academia, never really applying the statistics. However, if you are responsible for teaching students, they WILL be using statistics for physical processes. They need to understand the limits of what the statistics are telling them, and will have a large helping hand in industrial design. We do it here all the time (I'm just tangentially involved, but we do a lot of machine vision here, and it's all about measurement and statistics). To that end, yes, I expect a math major to have some knowledge of experimental control.

More to the point, for my first job I calculated cancer statistics for NIH (I worked for a private company contracted to them). That's grandly stated - I wrote software that did calculations; the heavy duty stats were done by math PhDs to whom I groveled . You better believe these people needed to have an understanding of experimental design. You can't just throw student-T tests at data and think you've proved something.

Unfortunately, not all of them did have that understanding. I had occasion to go crying up the chain of command that Doctor so-and-so was doing bad science. Unfortunately he was working under a grant from an outside source and the NIH couldn't stop him. At least I was able to get some honesty in the reporting of the results.

The point here is that I was not an expert at biology nor stats, but had passing familiarity with both; our stats people were not experts at biology, experimental design or software math packages, but had passing familiarity; and the medical doctors we worked for were not experts at stats, but had passing familarity. There was little call for somebody who only knew their particular subject, and couldn't understand what the other was saying, and perhaps improve on it. The doctor's gave me useful suggestions, and vice versa.

I recognize the other side of this argument - there are plenty of math careers where you don't really need this knowledge. I feel strongly, however, that schools should prepare us for a range of career options. Heck, when I graduated I wanted to get into chip design. Here I am writing 3D software. Careers change.

So, "perhaps".

 

[UnrepentantSinner]

North Texas Skeptics had a telekinetic claim and came up with a very novel way of handling it.

http://www.ntskeptics.org/2003/2003may/may2003.htm#mind

 

[roger]

Now *that's* a good test. Thanks, US.

 

[KRAMER]

NTS and RUSSELL SHIPP?

North Texas Skeptics had a telekinetic claim and came up with a very novel way of handling it.

http://www.ntskeptics.org/2003/2003may/may2003.htm#mind

Wanna see the correspondence between myself and this telekinetic claimant, Russell Shipp? He applied for the JREF Challenge in April 2004. I referred him to NTS for testing.
Although NTS had extensive dealings with Mr. Shipp (who had been tested by them on several ocassions), Mr. Shipp for some reason did not see fit to make me aware of any previous dealings he'd had with them. Although the relationship he shared with NTS seemed cordial and friendly, his emails to me (in which he referred to NTS associates as "biased knaves") revealed a strong distaste for NTS and their sincere attempts to show him that his "powers" did not exist. In the end, Mr. Shipp had become very nasty-nasty indeed, threatening legal action and what-not.

So go to the Challenge Application section to see what Russell Shipp was all about.

 

[KRAMER]

NTS and RUSSELL SHIPP?

North Texas Skeptics had a telekinetic claim and came up with a very novel way of handling it.

http://www.ntskeptics.org/2003/2003may/may2003.htm#mind

Wanna see the correspondence between myself and this telekinetic claimant, Russell Shipp? He applied for the JREF Challenge in April 2004. I referred him to NTS for testing.
Although NTS had extensive dealings with Mr. Shipp (who had been tested by them on several ocassions), Mr. Shipp for some reason did not see fit to make me aware of any previous dealings he'd had with them. Although the relationship he shared with NTS seemed cordial and friendly, his emails to me (in which he referred to NTS associates as "biased knaves") revealed a strong distaste for NTS and their sincere attempts to show him that his "powers" did not exist. In the end, Mr. Shipp had become very nasty-nasty indeed, threatening legal action and what-not.

So go to the Challenge Application section to see what Russell Shipp was all about.

 

[IXP]

Re: NTS and RUSSELL SHIPP?

So go to the Challenge Application section to see what Russell Shipp was all about.

I don't see any reference to Russell Shipp there.

 

[richardm]

Re: No "judging"...

There can be no "judging".

The test protocol must be of such a design that there is no room for interpretation or "judging". No other protocol is acceptable.

The results must be entirely self-evident. Period.

But there have been setups in the past which have involved a degree of judgement. I'm thinking here partly about the proposed test for Sylvia Brown, but mostly about the former "Cloud Busting" test. This involved people watching a video of the sky while cloud busting was going on, and seeing if they could tell which cloud was being busted (they couldn't).

Could a similar scenario be used for your claimant? Have the candle set up with quadrants as previously described. Film the candle while Beth does her thing. Show the video to a panel of individuals. Do this a few times. If for each video all (or most of) the individuals agree that the quadrant the flame moves towards is the same one, and if that's the one Beth was supposed to be directing it to, then the million dollars is won.

Seems quite simple, and removes the need for anything overcomplicated with respect to software or other measuring devices.

 

[KRAMER]

man vs. machine?

I'd greatly prefer one computer program to "judge" the results than a half dozen observors.

 

[KRAMER]

More from Ms. Clarkson

Another communication from Ms. Clarkson reveals a lot about the mechanics of self-deception. I have posted it in the Challenge section.

 

[drkitten]

Re: More from Ms. Clarkson

Another communication from Ms. Clarkson reveals a lot about the mechanics of self-deception. I have posted it in the Challenge section.

The relevant quote, being, of course :

One crucial ingredient in successful trials is how well I can convince myself that it works beforehand.

Be fair, Kramer. "Performance anxiety" is a well-studied and well-known phenomenon, and covers everything from stage fright to "choking" in a sports competition through impotence. Lack of confidence at almost anything is well-known to be a performance killer; I could give you at least a half-dozen peer reviewed articles about how lack of confidence in one's ability to do mathematics reduces scores on standardized tests, and there's certainly nothing paranormal about taking the SAT (with the possible exception of students hoping for divine intervention).

 

[Ashles]

But this doesn't sound exactly like Performance Anxiety.
That relates to the importance of the particular test and the pressure from the audience/examinors/etc.

But Ms Clarksons statements seem to relate to tests carried out with a lack of pressure or particular importance (I am assuming these are the tests she has carried out for her own research) (Edited to add: - I know she says that all her tests create anxiety; then why does she perform differently in different tests? And why does she question whether this is the reason other people have failed? She sounds like a strong believer in the first place, already creating excuses for herself and others.
Testing on your own to confirm your ability should not cause the levels of anxiety she seems to be talking about. Again I am concerned about her mental stability and whether she actually should perform the test)

This doesn't sound like performance anxiety as there would be nothing to be anxious about - but it does sound like she gets better results in a better state of mind which could quite possibly mean that she is seeing positive results where there are none. This could well be self-deception.
Plus we have to factor in the involvement of the 'partners' and the affect their influence may have.

This all creates problems either way in that, if her (already highly difficult to detect) ability is so sensitive to moods then being tested for a million dollar claim as a result of which she may be putting her career and relationship in jepoardy is about the least likely circumstances imaginable for her powers to work.

This could involve a lot of work and effort to test someone who is
A) Self -deluded
B) In posession of a paranormal ability so sensitive and difficult to measure that it will fail to be detected in any test where there is any pressure to prove the ability

Still I hope the test is carried out and that Beth accepts the result of the test so she can move forward with her life.

I worry whether she is giving her students her full concentration with all this going on.

 

[DinoJ]

I have followed this Beth Clarkson challenge a little closer than most of the others, because I live in the same city. Seems to me the part where she must convince herself that the ability will work is also the out she will use when it fails,

Kramer hopes that since she is an intelligent woman that hopefully she will realize her ability doesn't stand up to scientific scrutiny. I think if she fails she will say that she did not get herself in the positive frame of mind and that is the reason for a failure. She has set the table for her own win-win situation if and when a test should take place.

If I can help in any way with the testing if it is observing or whatever is needed please feel free to ask, I am willing to help if I can. I live in a suburb of Wichita, not more than 3 miles from WSU.


Dinoj

 

[Marian]

Re: Protocol

What we really need is a test protocol that leaves nothing open to interpretation.

I really don't know if such a protocol is possible with this claim.
I hope that something can be worked out.

I am bewildered by this intelligent woman's inability to see that the flame is behaving no differently when she attempts to influence it as it does when she is not doing anything at all.

Much like the music of THE SHAGGS, it brings my mind to a complete and utter halt.

Perhaps using a burner of some kind that emits a wind-resistant flame. Since she's using a paranormal ability, a wind-resistant burner should be of no issue to her. And additional measures could be used to ensure no strong drafts, no one blowing near the flame (though they have some burners that can withstand extremely high winds. See, being a smoker CAN come in handy!

If she is creating wind (and therefore such a burner wouldn't work), then she doesn't have to manipulate flame. That would be a different claim and one that could be tested in a sealed room with instrumentation where she just creates air movement of a specific type.