Right. Peter is concerned that skepticism is suffering dreadfully from the activities of Randi. Peter just wants us to steer clear of the man to help the Cause.
What a fine example of the concern troll in action.
There may be reason to chase down folk myths where they are persistent and have no other explanation such that it seems they must have a degree of truth. That is not the case here. Why dowsers think they can dowse is well understood. It’s a combination of ideomotor effect and lack of understanding about the ubiquity of underground water. There is no need to look further.
Also, your position on the statistics would make sense if (to continue the theme) Randi was throwing reindeer off the roof and only pointing out that they hit the ground but refusing to notice that they seem to fall unnaturally slowly. That simply isn’t the case. They hit with exactly the resounding thud you’d expect.
Dowsers have been tested plenty of times and not just by Randi. I don’t know if anyone has done a meta study but from what I have seen of published results they are null, null, null. It’s not as if people are persistently getting 3 out of ten (where chance would give one) and Randi is ignoring the results because the rules were that the claimant had to get nine. Over and over people don’t fail by a smidgen, they fail with results that looks precisely like chance.
There is no evidence of a moderate phenomenon hidden in tests looking for strong phenomenon.
If this is what I have seen, it is what Randi has seen 100 times more. There is no reason whatever to call the guy a quasi fraud for reaching the obvious conclusion and saying it out loud.
If I thought I had some ability, I’d want someone like Randi so I could prove to him what I knew was true. Also, what better way to shut him up or at least discredit him than to prove him wrong and take his million?
Just in case anyone is interested in the subject, but hasn’t read my posts in past years, I’d like to bring up three pro-dowsing books by the novelist Kenneth Roberts, Henry Gross and his Dowsing Rod, Water Unlimited, and The Seventh Sense.
Kenneth Roberts, late in his life, became interested – obsessed, even – with the possibility that the retired game warden, Henry Gross, had a valuable talent and was able to find water in a thirsty world. Believing that others should pay for this service, he formed a company to market Henry’s abilities world-wide. These books, preceded by a short magazine article, chronicle the experience. Roberts is an engaging writer and uses his own talents to document the story.
There were successes and there were failures. What’s interesting is the sincere attempt to find out what caused the phenomenon of dowsing, but that was ultimately a failure, too. It didn’t stop them from using it or believing it worked.
It’s a good exercise for skeptics to read these books, as many of the stories seem to support the dowsing theory, including finding water where there supposedly was none (in Bermuda and other places). Map dowsing was used as a preliminary tool.
I challenge any reader to find the flaws in Roberts’ reasoning. They are hard to find, but they are there.
And, further, FP as GIGObuster posted in the other GD thread, you really need to read this study:
Even where a very sympathetic study is conducted looking for any evidence of ability, the only way you can try to get a positive finding out of it is to cheat by mining the data for isolated examples of unrepeatable statistical blips.
Randi would know about that study. Any pronouncement he makes is just making the obvious joins between dots.
Fatheringay-Phipps makes a true if extremely pedantic point. You can’t conclusively infer that tests designed to detect a major effect of dowsing can’t necesarily rule out a very subtle one - although as you have more and more of these tests and a larger data set, the less likely it becomes probable.
But… practically, how likely is this to be an important distinction? People have been claiming for thousands of years to have magical powers, and the vast majority of people who have ever lived (including those alive today) believe that at least some of these magical claims is true. So you’d think the world would be absolutely swamped in magic and the supernatural, right? But of course it isn’t. Despite all this widespread belief, and all these claims, no one has ever demonstrated any of these magical powers in a scientifically verifiable way. Many have been exposed as frauds.
So can we say magic isn’t real from this? No. Nor can we conclusively say - on the most pedantic level - that there’s no pink unicorn under my bed, even if I have a thousand people look and they all agree. But practically it’s proven. There’s no unicorn under my bed, there’s no magic. This is the vastly likely nature of the world.
The JREF tests allow a way for someone with magic powers to demonstrate it. Everyone so far who has taken the test has been either a fraud trying to fool it, or someone who simply set themselves up to maintain a delusion of their power. This isn’t at all uncommon - again, we’ve all got all sorts of crazy cognitive biases, we all believe whatever we want to believe, and almost everyone thinks there’s magic in the world.
But realistically, it’s extremely unlikely anyone will ever beat the JREF test. Not because it’s unfair, but because there’s no fucking magic in the world. Sorry, it sucks, but it’s almost certainly true.
So that being said, why waste the effort putting a whole lot of more effort into tests to try to examine a very subtle magical effect? Especially when the claimants aren’t claiming a subtle effect, and they’re confident they can show a major effect. Yes, you could run a few hundred trials or a few thousand trials of a dowsing test and start having some confidence that people can’t score even subtly above chance, but that’s logistically a pain in the ass. And the odds of there being such a subtle effect are so small as to not be worth it.
So technically yes, you can’t say “the JREF testing proves there’s no dowsing effect”, in the same way that 100 people looking under my bed and not seeing a pink uniform doesn’t prove it isn’t there. But that’s pedantic and probably worthless. But even so, what you can say to anyone who claims to have clearly demonstrable powers (and there are thousands of frauds amongst this group) is - if you’re so sure of your abilities, why not take the million dollar challenge? And this is very much a valid critique against their claims, because the sort of claims they’re making are exactly what those tests are designed to handle.
Is anyone arguing to the contrary? I think FP may be attempting to suggest that Randi has said to the contrary but I don’t think that is correct.
(Bolding mine)
I know it’s a typo for unicorn, but …
If there’s no pink uniform under your bed, how can you look fabolous at parade?
Or did the pink unicorn eat your uniform? 
Someone, I think his name is Randi something, has commented that scientists are the most gullible people there are as it’s part of scientific thinking that people don’t cheat.
Randi has made some comments along those lines but whether he’s said exactly that I am not sure. However, it would be silly to say that. Any scientist who studies people knows they lie like hell in surveys etc.
Two different disciplines: scientists who study people = psychology, sociology know that people lie intentionally and un-intententionally.
But the scientists who study para-psychology often come from some kind of physics background - they are looking if there’s some explanation for the powers that makes sense, so they are thinking along the lines of waves or energy of some kind; not of cheating persons.
And even with those knowledgeable, there’s quite a gap between structuring surveys to get past unintentional fudging, and somebody who’s outright performing trickery at the level of a skilled magician. Scientists are still human, after all, and as such still can fall into the “seeing what you want to see instead of what’s really there” problem, esp. with a skilled misdirector/ sleight of hand person.
I’ve never read these books, but most of what (little) I know about dowsing is from the chapter on it, focusing on Roberts & Gross, in Gardner’s Fads and Fallacies in the Name of Science. (That book was a staple of my childhood, and I still have the tattered copy.)
I had that book, too… dunno what happeend to my copy, though. It’s probably jammed in the back of some bookcase. Gardner’s chapter on Dianetics (i.e. nascent Scientology) was quite interesting, as was the book overall, though I recall one amusing error - he cited a pseudo-scientific claim that transplanting goat gallbladders or something into humans would restore their health and vitality, and suggested this kind of glandular animal transplant had no scientific basis. Decades later, though, we can see some promising results from transplating animal glands such as pig livers into ailing humans.
This in no way excuses the charlatan who cheated and maimed people in the 1950s, though.
The point I was trying to make was that without a specific test and observed data, it is impossible to say that a true psychic 30% success rate was statistically significantly different from a 5% success rate. I think our disagreement also follows from our different application of statistics. I imagine that actuarial statistics is primarily concerned with model formulation and point estimation, while I may have more exposure to rigorous hypothesis testing and clinical trial design. In my line of work, the entire test must be defined from the beginning, and post hoc analysis of the data outside of the original defined parameters is seen as very bad practice. Thus my definition of statistically significant is exactly equivalent to passing Randi’s test. In my buisnessf the petitioner found that he got it right 30% of the time and thought that that was good enough to prove an effect the correct response would be to wait a year, and re-apply suggesting at test that was correctly sized so as to detect the less stringent alternative.
Of course.
This doesn’t follow. Because you could design the test at the outset to test for a lower level of correlation.
See above. But as an additional point, I would note that we’re not discussing whether the results of Randi’s tests prove a lower level of correlation. We’re discussing whether they rule out a lower level of correlation.
Note that Princhester claimed above that on Randi’s tests “Over and over people […] fail with results that looks precisely like chance” and other posters have made similar claims. If that’s the case, then Randi is entitled to make his claims, but I’ve not seen any documentation of this. My impression is that results have varied in Randi’s tests and I’m not aware of any meta-analysis.
Roberts had a feud going with Gardner (and Ray Hyman) about this. I am by no means in Roberts’ camp, but I find his writing perhaps the best example I’ve ever seen of flawed reasoning made plausible. So many woo books just make you :rolleyes: but Roberts’ stories are in a different class.
I was first introduced to the series decades ago, and made the mistake of reading the last book first. It was filled with statements like “Henry asked his rod for the depth, and it replied, 66 ft.” I found it incomprehensible as to just how this worked – assuming Roberts wasn’t pulling my leg or writing a Tolkien-type tale. Did the rod actually utter words?
I had to read the first book, Henry Gross and his Dowsing Rod, to find out. In the frontispiece is a photograph of Gross holding the rod and querying it, which I found out involved asking a question, either out loud or by thinking it, and waiting for the rod to dip. A dip meant “yes” if it was a yes or no question; it meant “keep going, you’re partly right” if it was a number question.
A sample session:
Henry: Is it 50 feet to water?
Rod: (dips)
Henry: 60 feet?
Rod: (no dip)
Henry: 65 feet?
Rod: (no dip)
Henry: 66 feet?
Rod: (dip)
So Henry announces it is 66 feet to the water. Then he begins another sequence to find out the depth to the bottom of the “vein” and how many gallons per minute it flows, whether it is good to drink (!), etc. You can see that this could be time consuming.
The very idea that a piece of wood can hear human speech, understand English, and provide its own motive force, even when coupled with the dowser (because one theory is that it’s the dowser’s mind that is part of the equation, the rod is merely an indicator) was a little too much for me to swallow. But Roberts skips over this, saying, “The ability to find water is such a valuable resource, we shouldn’t waste time testing it – let’s put it to work!”
Another interesting example…Henry was often called upon to roughly dowse, from hundreds of miles distant to the site, using a map. Roberts kept copious notes, but claimed to not show them to Henry. He tells of a trip months after a map dowse, when Henry goes to the site and dowses it more accurately, Henry comes up with the same numbers. Of course, we can all think of excuses for this; maybe Henry has a very good memory; maybe he peeked at the notes, maybe Roberts didn’t write them down until later, maybe it was a lucky guess, but Roberts uses examples like this to prove that map dowsing works!
Very educational. I am able to find the first book in Amazon for only a few dollars; the other two are rarer. There is also a Kenneth Roberts Reader, which contains the introduction to the whole crusade as one of several short articles/stories, and that, too, is worth reading.
Oh look! A formal warning. Moderators are getting sick of this and Peter has no idea how much they protect him rather than the other way round.
Sorry, the editing window closed before I caught my mistake in post 116. The dialogue should read:
Henry: Is it 50 feet to water?
Rod: (dips)
Henry: 60 feet?
Rod: (no dip)
Henry: 55 feet?
Rod: (dip)
Henry: 56 feet?
Rod: (no dip)
So Henry announces it is 55 feet to the water…
Well, you could, but do you think they want to run hundreds or thousands of trials for every applicant that comes their way? Why not just test to the claiM? IF a person says they can do something 100% of the time, set the bar at 80-90% and save yourself some effort.
Well, now we know why Peter Morris hasn’t shown up in this thread.
http://boards.straightdope.com/sdmb/showpost.php?p=13528265&postcount=129
Here’s an idea. In the future, all threads that are likely to contain the words “James” or “Randi” or any combination or permutation thereof, should be posted in the BBQ Pit. Problem solved.