This comes up sometimes in these debates, and it’s evidence of extreme statistical ignorance.
Here’s the deal: There are odds associated with each event you describe above. The odds of calling 20% (or more) 10-sided-die rolls right in 10 throws can easily be calculated:
The odds of getting 0 right: 0.9[sup]10[/sup] = 34.9%
The odds of getting 1 right: 0.9[sup]9[/sup] * 0.1[sup]1[/sup] * 10 (combinations) = 38.7%
The odds of getting 0 or 1 are therefore 73.6%. Hence, the odds of getting two or more out of ten are 26.4%
The odds of getting 20% of 200 (i.e., 40) are nowhere close. (Brute-forcing the calcs on Excel), the odds of getting 40 or more out of 200 are 0.0017%, or about 1 in 58,798. Still, with these odds Randi would stand to lose $1,000,000/58,798, or just about $17 even.
I’d be willing to call someone “successful” with those odds, but Randi would still be a fool to take the bet. This is a perfect illustration of what’s wrong with the math behind the challenge.
The odds of getting 20% of 200,000 rolls would be truly astronomical.
Got it?
No disagreement here.
Too bad you didn’t read my first big post. This is a key part of my argument: The odds at which we would commonsensically dub someone a “success” and the odds Randi requires to call someone a success are widely divergent.
OK, let’s start up a “Bad Science” page, too: The number nature of tests required to prove a claim is based upon the nature of the claim, not subjective opinions of how “weak” or “strong” it is. If you think my claim to fly in a machine is “weak,” then all I have to do is fly by you once to prove it. Or would you prefer more?
Oodles, huh? Glad to see you’ve got the jargon down.
But you had time to post a post in which redundancy is the least of your worries.
“Just be done”?! So if a drug’s efficacy versus a placebo had to have a p score associated with it of 0.00000001, no drug would be inappropriately rejected?
C’mon man, you’re saying stuff that just can’t be taken seriously.
Sorry, Peter, you’re wrong here too. 99.999% means a 0.001% chance of success, or 1/100,000. 1,000,000/100,000 is 10. Randi would still stand to lose $10.
Does this qualify as the train wrecking? If so, my prediction that it’d be on page five was correct, and I’m therefore psychic.
Anyway, I’ve also studied statistics, and while Aeschines is (mostly) correct, he’s undeservedly smug about it.
Interestingly, a number of JREF respondents to the Beth Clarkson thread were persuasive and rational, which I guess makes them wild exceptions to the foaming mad-dog backflipping image of “Randi fans” cited so often by Peter.
You weren’t doing very well in this thread, as regards myself, even before I got here. I didn’t join in till page four but on page three you’d already said “even Princhester saw how unreasonable JREF was being” which can only be parsed as meaning that Princhester is not someone that one would normally be capable of detecting when JREF is being unreasonable. And then later on the same page, you go on to say “even to a fanatic as obsessive as Princhester” which can only be parsed as saying that I am a fanatic and an obsessive.
But off you flounce in your little huff, Peter: it may cause you appear hypocritical, but at least you won’t have to be confronted by the numerous unanswered questions currently hanging over your head.
Any sign of weakness in these threads is taken by the ignorant to mean that one’s point is a matter of opinion or speculation. One must then train them in simple facts while they contumeliously assume their own rectitude and sneer until the denoument of tail-between-legs embarrassed slinking off.
In short, whip the dog before he starts to gnaw on you.
Actually, unfounded arrogance is a sign of poor technique and an attempt at argument by intimidation. Fortunately, the intelligent among us (myself included) can parse the facts from the bullshit.
As for the signs of weakness, I haven’t seen a strong anti-JREF claim in this thread (i.e. one with any real evidence behind it) beyond the fact that this “Kramer” guy is a jerk.
As near as I can tell from all the fumbling around, the main problem you have here with JREF, Aeschines, is that they strictly demand repeatable evidence; being the sponsors of the prize and the designers of the trials, it is entirely within their mandate to set the guidelines for the experiments, much as the FDA (to use your example) sets guidelines in cost/benefit terms when testing drugs.
Forgive me for bringing up one of the past discussions similar to this one where you and a few others tried to argue in favour of a scientific view of the paranormal. I’m sure there are those who found it useful (or at least amusing), even if you didn’t.
The evidence required to win the million dollar prize is stringent, yes – and well it should be, particularly for experiments where a psychic might get a hit by pure chance. Obviously a lesser degree of strictly repeatable evidence is required for successful feats that may not be attributed to chance, such as levitation. Do you argue that the testing rules ought to be relaxed when attempting to prove that which has systematically failed to be proved in the past, and which, whenever previously investigated, has proved to be the product of trickery?
I don’t care about Randi’s org. I even haven’t said a single word about Randi himself. I’ve been talking about the Challenge and the problems with it.
Man, you sure know how to miss a point. Several points. All my points. Don’t tell me what you think my problems with this or that are, especially if you’re wrong. Speak to the arguments. Reread, rethink.
No disagreement. But the conclusions one may draw from whether or not a particular person has passed the Challenge, or whether or not a particular phenomenon has been verified by the Challenge, are limited to the extent that the scope of the Challenge is limited.
I don’t disavow what I said in that thread; I merely indicated that it had no pertinence in this particular discussion.
No, really? [/sarcasm] That’s been one of my major points, and (must I state it again), therein lies the rub. Or one of the big rubs.
You phrase it wrong–it’s not a “lesser degree”–but yeah. One may guess that the Challenge would deem one obvious instance of levitation enough to take the dough. You might call this category “grossly atypical physical phenomena” (i.e., amazing shit). The trouble is few psychics claim to be able to do such things in the first place; I think mediums and ectoplasm, etc., would be just about the only thing regularly (which is not to say commonly) claimed.
This is a veritable knot of assumptions. I made my points clearly about the Challenge’s underlying principles and limitations in the posts where I made them. Speak to the arguments, sir, speak to the arguments.
So have I. And JREF owns, sponsors, and defines the challenge. The criticisms you post here seem to be primarily that the basic design of the challenges are somehow faulty. Design must be approved by JREF before the challenge can take place, so they’re essentially in charge and are therefore the people you are railing against. Plus, you have alluded to Randi as some kind of hero worshipped by blind and biased sceptics (somewhat oxymoronical), and you are arguing alongside the greatest Randi detractor on these message boards, a fellow with considerable experience in making strident claims and accusations.
On the other hand, you could rethink and restate those mangled arguments and state your position with greater clarity when you are so asked, especially given how our past discussions have turned out, hm? If you find yourself repeatedly uttering commands such as “reread, rethink” without any further attempt at clarification, the one who is being unclear could well be you. And it is, as I will demonstrate.
The same goes for every single test, experiment, trial, examination, etc. on the face of the planet: so what? Again, I ask you to restate your thesis instead of indirectly trying to impugn the challenge, JREF efforts, or whatever else it is you are doing.
I posted that extract because you were hand-waving about a discrepancy you perceive between analysis of experimental results in the paranormal and in medicine/social science (by the latter I assume you mean the various flavours of pyschology, as you did last time we discussed this). This is exactly the same argument you trotted out in the thread I linked, albeit spun somewhat differently because you are now talking about statistical significance in particular rather than the admissibility of anecdotal versus systematic evidence.
This self-assertive sentence falls short on meaningful content. Tell us about this “rub” in greater detail. Help us poor folk who know so very little about reasoning and science: why do you have a problem with the approach I summarized in the above quote? What exactly is the rub?
Kramer says “the great majority of applicants who actually agreed to a protocol (and were tested) were dowsers. I’d say about 50% if not more, with “remote readers”, psychics and ESP claims making up the remainder.”
Anyway, so when witnessing an effect that is almost unequivocable – such as a man levitating into the air under carefully controlled test conditions – it is all right to accept somewhat relaxed replicability of a paranormal feat, though I would never go so far to say that once is enough. Enimal ist keinmal, remember, and if a feat is not replicable then it is no good because chance or trickery may not be ruled out. The problem you then move towards is that psychics and similar claimants are time and time again unable to perform and replicate paranormal feats under agreed challenge conditions without recourse to trickery; this is routinely confounded into statements such as “I can’t do it in the presence of a sceptic” or “I can’t do it 100% of the time” or any number of others, when in fact the requirements are twofold: perform significantly better than chance, and at a rate at least as great as that which the psychic agreed to before the test.
I still don’t understand your objections and equivocations.
So you don’t like the probabilities used in a challenge, even when the applicant/claimant agreed to the guidelines. I don’t see the issue. The psychic claimed to be able to do something; he was tested and found wanting. What’s left?
The only discernible position I find you maintaining is this one from your first post in this thread:
I agree that the existence of the Million dollar challenge is not a catch-all remedy against ignorance such as paranormal belief. It is conceivable, after all, that true psychics with real powers exist but do not want to take the Challenge, win a million, or be put in the spotlight. But the fact that no one has ever won the Challenge, or even made it past the preliminary test, is quite a damning factor when it comes to the usual array of claptrap claims and mumbo-jumbo that has been formally and exhaustively investigated for well over a century and a half. With the Challenge we have, yet again, an absolute lack of evidence for any paranormal claim; personally, I prefer to cite proper scientific tests that have consistently failed to demonstrate support for the paranormal, however the Challenge really isn’t anywhere near as bad as you make it out to be.
Your conclusion above is built on very shaky ground. Pending clarifications and restatements of thesis on your part, let’s look at the most important problems, since you seem quite eager to have your earlier post addressed:
The challenge does not require judges for a very good reason: the results speak for themselves, given that the design of each challenge necessarily incorporates demonstration of paranormal abilities in a controlled setting. Remember, all parties contribute and agree to the structure of the challenge before carrying it out. This isn’t synchronized swimming or other discipline where a human performance/aesthetic judgement is required. You don’t need judges in an experimental setting, you simply need good design, appropriate conditions and control, an observer who influence the results as little as possible, and peer review (to scrutinize the entire exercise for errors, fraud, etc.).
Psychics who undergo scientific testing are not exempt from peer review of published results (which is a good thing, because some psi proponents resort to dubious statistical tools such as meta-analysis to identify effects so infinitesimal they are otherwise invisible). The Challenge (to my knowledge) skips the peer review part, which one would assume would make it even easier for psychics to be home safe – if only they could actually demonstrate anything.
Such as? This is a discussion about replicability of feats and minimizing the occurrence of success by chance. Again, I see nothing wrong with discussing with a psychic what he can and can’t do, and then designing an experiment to establish whether he can in fact do it consistently.
I would imagine the odds of success by chance vary with each challenge.
The whole point of the exercise is that the Challenge is indeed virtually impossible unless you truly do demonstrate paranormal abilities in accordance with mutually agreed guidelines. That is precisely why the Challenge is structured like that. If you have the goods, you win the million, otherwise you go home.
You do realize that the peculiarities of the Challenge you complain about (when you get them right) are there by design to make it as close to a proper scientific test as possible? You don’t have judges in such tests, and you certainly do your best to minimize or eliminate the effect of chance when you are testing psychics. I see nothing seriously wrong with JREF doing the same for the Challenge, in fact I think it is a primary requirement of the exercise. If you want to discuss individual items such as flawed procedures in a specific test, that is another matter; everyone makes mistakes every now and then of course, but the approach of the Challenge seems otherwise sound.
[Snip Massive verbiage completley unrelated to points I’ve made thus far.]
Yeah, you don’t need judging, just peer review, which is expert opinion, which is juding.
Oh, it skips the peer review part? How scientific!
Stop imagining and start citing.
You are simply asserting the thesis that is subject of the debate. Yes, even the great Abe is required to read the posts, quote them, and respond to particular arguments if he wishes actually to “participate” in the debate. Tough, I know.
The “science” of the whole thing is what is under debate. See previous comment.
Nope, it’s the approach that’s under fire. Try actually reading some of the posts.
Thanks, professor, for relieving me of my “extreme statistical ignorance” by explaining what I already know. My point was that if someone truly has the power to predict 10-sided die rolls 20% of the time, then doing 200,000 tests (an arbitrary large number) would prove beyond a stastical doubt that their power is legitimate. But on this point you obviously agree with me.
By “weak” I meant closer to the odds predicted by chance. I did not mean anything subjective. Obviously some claims are qualitative and not quantitative in nature.
So you’re going to needle me for using the word “oodles?” This from the man who used the word “commensensical” in the same post? Give me a motherf***ing break.
I could take this criticism far more seriously if you had understood my post. I’ll try to be more “commonsensical” in my writings next time so their meanings are more obvious.
