Actually, to correct an earlier mistake on my part, Kramer said “I hope he get’s bone cancer” not “I hope his mother gets bone cancer”. For what it’s worth.
True. For a conclusive test it would have to be blind, that is, the people being “healed” would have to not be able to see or in any way sense the people attempting to “heal” them. Which means that if part of the “healing” involved laying on of physical hands, direct eye-to-eye contact, etc., it would be a LOT harder to test, bordering on impossible.
Humanist, it is not very polite to drag in a quotation from a separate thread to rebut that poster when that poster has chosen to stay out of this thread.
I think we can agree that this Kramer guy is abrasive, snide, dismissive, etc. I agree that JREF might be better served by someone more diplomatic. But that doesn’t invalidate the test protocols or prove that paranormal ability exists.
I also bear in mind that Kramer has to deal with a large number of kooks and it must get pretty taxing after a while.
There’s a message from Kramer on Nov. 16th stating that the 27th has been chosen. A few minutes later Mr. Carey says its ‘no go’. I find it difficult to believe that up until that point Mr. Carey had no input regarding the date. Certainly if any dates were ‘no go’ I would take pains to inform people of this fact ahead of time.
You have no evidence of that whatsoever.
Cheap hyperbole wins you nothing. A date was announced some 11 days in advance of the test date, and 6 days before the ‘ultimatim’ was issued. (Perhaps had Mr. Carey not cancelled the date already, there would not have been an ultimatim).
If Mr. Carey had the events you listed above then he should have made those scheduling aware of that fact. This is most certainly his fault for not bringing any such conflicts up. (You may notice that Mr. Carey lists no excuse, Peter has to make them up for him to blame this on Randi.)
I still have little reason to believe your assertion that they kept Mr. Carey out of the date choosing loop. Princhester is correct, there is nothing that Randi can do that will satisfy you.
Diogenes, Kramer goes well beyond abrasive, snide etc. He’s flat out not competent. Go over to the JREF and read the challenge application and other threads concerning TC Albin, Beth Clarkson and Paul Carey. Kramer’s understanding of statistics and experimental design is beyond piss poor, his ability to comprehend what people are explicitly saying to him is highly dubious and his ability to understand nuance is set to “uberdefensive”. These latter abilities are quite simply fatal to the ability to negotiate, which is what his job is.
I understand that applicants are often very difficult people, I understand that Kramer may be a really nice guy etc, but his job requires certain skills and he doesn’t have them.
Incorrect. The medicine argument is one I’ve made for the first time in this thread. The very first. I challenge you to find it elsewhere; I’ll eat crow right here if you can.
Preaching requires a basic level of accuracy. Seriously.
I certainly agree that someone more suitable could be found to do the job. The more gracious, attentive and cooperative the negotiations are, the better it would be for JREF’s credibility. They should endeavor never give the impression that they are avoiding testing an applicant if the applicant is making any sort of falsifiable claim. I also agree that Kramer doesn’t seem to understand statistics and given that an awful ot of these claims depend on beating statistical chance you would thing that the contact person should have a grasp of the numbers.
What I am talking about is your insistence that you can’t conduct a test of a phenomenon that doesn’t have a 100% success rate to a high level of statistical confidence.
If that wasn’t clear from my post, I apologise.
Diogenes I don’t think even gracious, attentive and co-operative are necessary. Abrupt, accurate and matter-of-fact would be fine.
Hey, that’s objectivity, much appreciated. I think this gets back to what I said first in my first post in this thread: It’s primarily about the philosophy of testing, secondarily about the “paranormal.”
We are at cross-purposes here. What you say is quite true: you can test such things to a high level of statistical confidence.
But that’s not what I’m saying. I’m saying that the odds that Randi demands are such that anything without a 100% success rate would be deemed invalid. *He’s * the one that demands (at least on occasion) 1 in 10[sup]10[/sup] in order to proclaim a winner.
As you are well aware, in the social sciences a big ol’ fat p-score is just fine. FWIW, phenomena are considered “demonstrated” by stats that would earn the virulentest scorn of skeptics were the object of study the “paranormal.”
But I still don’t see how you can continue to think this. There is no difficulty with getting a 1 in 10[sup]10[/sup] on something that you can’t do with a 100% success rate, if you just do a fair few repititions.
I don’t think we’re really disagreeing.
The 100% success rate is a red herring. Randi just happened to stipulate in one test that the sucess rate be 100% (ten out of ten), with odds such that chance was 1 in 10 billion. It’s the high odds that are the problem, not the 100% success rate per se (after all, 100% success in guessing the outcome of one coin toss would be next to meaningless; 75% success in guessing 1,000 times would be mind-blowing and represent similarly astronomical odds).
The point stands that in the social sciences and medicine such odds against chance are never demanded, and the actual stats are scornworthy by the standards of the hard sciences.
The p-score comprises such astronomical odds. What the p-score means is that the probability of those medical or social results departing from chance as widely as they did is astronomically small.
Yes they are: that is the very basis of statistical significance.
It seems the item being latched on to this particular iteration of the same old argument is “what is statistically significant”?
Perhaps we can take a look at specific examples one by one instead of hand-waving about how such strict standards aren’t ever followed in the social and medical sciences.
Aeschines, you, the departed Snakespirit, and Peter “JREF killed my dog” Morris have been equivocating about standards of evidence on the topic of the paranormal for years now; but what will it accomplish? Even if you establish that JREF screwed up on a couple times (Kramer sure sounds like a weird guy to have on board) that won’t automatically render the applicants endowed with paranormal abilities and win them the prize.
Here in this lengthy wrangle, you questioned the fact that anecdotal evidence is unacceptable in science, but that in the social and medical sciences anecdotal evidence is actually acceptable. This is a position surprisngly similar to the one you are expressing in the present thread. I responded that though some sciences make use of anecdotal evidence, it is only systematic evidence that may be used scientifically. I eventually provided this example of the difference:
Are you sure you are not similarly confusing your terms in this discussion? It would seem so. The FDA, for example, routinely fails millions of dollars’ worth of experimental drugs every year on the basis of unacceptable statistical significance in its trials. We wouldn’t have developed a clue on how to treat serious diseases over the last century if we were truly unable to establish which treatments proved successful in a statistically meaningful way.
The reason that Randi demands a p less than 0.05 is that there is a million bucks at stake. If I could get him to agree to a p of 0.05, I just have to go to the trouble of being tested several times before I’m likely to win by chance instead of skill.
If someone really has these powers, demonstration to an arbitrarily small p would be no problem.
Oh come now. A p-score of .05 would mean that there is a 1/20 chance that the results occurred simply due to chance. Those aren’t astronomical odds against chance. In the social sciences they work with p’s a lot bigger than that. In medicine, smaller, but certainly not 1 in 10 billion. No drug would ever be approved.
You just restated a key part of my argument.
You just rejected a key part of my argument without actually responding to it. Reread, rethink.
Curt is exactly right. If I have the ability to predict a coin flip before it happens approximately 7 out of every 10 times, that is DEFINITELY paranormal. If I go and say to Randi “I can predict a coin flip 7 out of every 10 times”, and he says “you must get 10 out of 10”, then he’s either a fraud or an idiot. If, on the other hand, he says “well, then you should have NO problem getting, say, 65 out or 100, or 130 out of 200, or 195 out of 300”, and we make those numbers large enough, then it will still be pretty easy for me, but will be only 1 chance in 10 billion (or whatever). The only real problem there is if I genuinely can do 7 out of 10, consistently, but doing so is an enormous strain on me, and I can only do that 10-coin-flip test, after which I need to rest for a week to regain my psychic powers. In which case, the test would take years to run. In which case, my power probably wouldn’t be applicable to the JREF challenge.