The "Process" of John Edward

[Thanz]

Originally posted by BillHoyt

Thanz,

Perhaps you've forgotten that Kerberos' analysis rejected his null hypothesis? My analysis excludes data his included, and my altered analysis rejected my null hypothesis.
I have not forgotton that at all. Kerberos was focussed at the other end of the spectrum - less common initials.

What I did was take his raw counting data, and applied YOUR hypothesis to it. And what I found was that YOUR null hypothesis could not be rejected if we used his counts.

It could also not be rejected if we used my counting methods.

The only analysis which could be used to reject your null hypothesis is the one that you did based on a flawed counting methodology.

I would like to know why you think it is appropriate to count "And they're also talking about somebody who would be known as either Richard or Rich, because a big R-connection that comes up connected to you" as more than one guess. In my count, I counted this as one R guess. From your description, I gather you counted it as 3 R guesses.

[Lurker]

Still exploring the topic of Poisson.

Bill, lets take a step back and try a simple example. Let's say in the census that the letter "J" started in 90% of the population. Now let us say in the LKL transcript JE had a total of 85 guesses of which 80 were "J" guesses.

Apply Poisson just like you did previously arrives at this table, a portion of which I have excerpted:

Count CDF
75 0.461976422
76 0.507613296
77 0.552953827
78 0.597422424
79 0.640483788
80 0.681661216
81 0.72055101
82 0.756832342
83 0.790272365
84 0.820726672
85 0.848135548
86 0.872516699
87 0.893955297
88 0.912592261
89 0.928611673


AM I reading this incorrectly but my interpretation has the Poisson saying there is only a 87% of getting less that 86 "J" guesses in 85 tries. That there is a 13% of getting MORE than 85 "J" guesses in 85 tries.

Is my interpreation correct? IF not, where is it wrong? Thanks for your time.

BTW, results from EXCEL. Try it yourself. It is remarkably easy to set up.

Lurker

[BillHoyt]

Originally posted by Lurker
AM I reading this incorrectly but my interpretation has the Poisson saying there is only a 87% of getting less that 86 "J" guesses in 85 tries. That there is a 13% of getting MORE than 85 "J" guesses in 85 tries.

Your error is in blue. Sadly, I have pointed this out so many times I am now blue in the face.

[BillHoyt]

Let me put you out of your misery, Lurker. There is a 13% chance that you would count 85 "J"s. Period. No qualification on "number of tries". Are you with me yet?

[Walter Wayne]

Lurker with prob of J at 90%, it doesn't make sense, but for N=85 and prob=0.1336 we get the following for binomial and poisson distribution. Note the p=0.05 is at 18 for both.

Walt

[Lurker]

Bill:

Thanks for the response. Recall in our example we took a 85 person sample, just like you did for the "J" analysis. What is the probability according to a Poisson analysis that we would get 85 or less "J"s using p=0.9?

Cheers,

Lurker

[Lurker]

Originally posted by Walter Wayne
Lurker with prob of J at 90%, it doesn't make sense, but for N=85 and prob=0.1336 we get the following for binomial and poisson distribution. Note the p=0.05 is at 18 for both.

Walt

Walt, thanks! This is what I was interested in.

BTW, why doesn't it make sense when p=0.9 in my example provided? Could it be because Poisson is nto a good approximation at that level of p? THAT is what I was trying to get through to Bill which he refused to acknowledge.

Lurker

[Walter Wayne]

Originally posted by BillHoyt
Let me put you out of your misery, Lurker. There is a 13% chance that you would count 85 "J"s. Period. No qualification on "number of tries". Are you with me yet? What?

For poisson distribution, m = pN, and thus the tail moves out as N increases. It would be a horrible tool if it didn't, because if p=0.05 occurs at lets say 18, then one could just count guesses until one gets to 18 and declare success.

Walt

[Lurker]

Originally posted by Walter Wayne
What?

For poisson distribution, m = pN, and thus the tail moves out as N increases. It would be a horrible tool if it didn't, because if p=0.05 occurs at lets say 18, then one could just count guesses until one gets to 18 and declare success.

Walt

Why is it Tai Chi, Lurker, and Wayne all see that N is part of the Poisson distribution yet the self-styled expert, Bill Hoyt, does not?

This seems fairly selfevident.

Lurker

[BillHoyt]

Originally posted by Walter Wayne
What?

For poisson distribution, m = pN, and thus the tail moves out as N increases. It would be a horrible tool if it didn't, because if p=0.05 occurs at lets say 18, then one could just count guesses until one gets to 18 and declare success.

Walt
Walt,

Ask yourself this: what was N for the crows? (See my post to Lurker.)

Cheers,

[BillHoyt]

Originally posted by Walter Wayne
What?

For poisson distribution, m = pN, and thus the tail moves out as N increases. It would be a horrible tool if it didn't, because if p=0.05 occurs at lets say 18, then one could just count guesses until one gets to 18 and declare success.

Walt
Walt,

Ask yourself this: what was N for the crows? (See my post to Lurker.)

Cheers,

[Lurker]

Originally posted by BillHoyt

Walt,

Ask yourself this: what was N for the crows? (See my post to Lurker.)

Cheers,

Ah, here we go with the questions answering question routine...

Good luck Walt, you are going to need it.

Lurker

[BillHoyt]

Originally posted by Lurker


Why is it Tai Chi, Lurker, and Wayne all see that N is part of the Poisson distribution yet the self-styled expert, Bill Hoyt, does not?

This seems fairly selfevident.

Lurker

Argumentum ad populem.

What is N for the crow example?

[BillHoyt]

Originally posted by Lurker


Ah, here we go with the questions answering question routine...

Good luck Walt, you are going to need it.

Lurker

I've already given my answer: you are wrong. I have already given explanations. They are still not getting through. Is my crow example wrong? No. Then ask yourself why it works when we have no N Ask yourself what really moved Walt's tail. (And, no, I'm not getting personal about Walt.:D )

[BillHoyt]

Flash from my Kreskin's Krystal Ball(tm): Walt will get it first.

[Lurker]

Well, if Walt gets it and we are in error I will be the first to apologize.

Lurker

[Lurker]

Originally posted by BillHoyt


I've already given my answer: you are wrong. I have already given explanations. They are still not getting through. Is my crow example wrong? No. Then ask yourself why it works when we have no N Ask yourself what really moved Walt's tail. (And, no, I'm not getting personal about Walt.:D )

Yes, your crow example is wrong. Dreadfully wrong.

Lurker

[BillHoyt]

Originally posted by Lurker


Yes, your crow example is wrong. Dreadfully wrong.

Lurker

Well, then, there you have it! Thanks for contributing to the Tottles.

[T'ai Chi]

Originally posted by Lurker

Why is it Tai Chi, Lurker, and Wayne all see that N is part of the Poisson distribution yet the self-styled expert, Bill Hoyt, does not?

This seems fairly selfevident.

Lurker

Hold on a sec here. :)

The sample size drops out of the probability function. In the Poisson, the sample size, N, is not a parameter.

N is used to estimate lambda, and lambda is used for the Poisson distribution, but this estimation of lambda can be done outside of the Poisson distribution.

So, I guess you can say you can use N in the Poisson as part of calculating lambda, but you can't say that N is a parameter in the Poisson.

But anyway, I doubt the applicability of the Poisson here. If JE just used J's, that would be OK. However, why no one is interested in the other high frequency letters JE uses is beyond me.

[BillHoyt]

Originally posted by T'ai Chi


Hold on a sec here. :)

The sample size drops out of the probability function. In the Poisson, the sample size, N, is not a parameter.

N is used to estimate lambda, and lambda is used for the Poisson distribution, but this estimation of lambda can be done outside of the Poisson distribution.

So, I guess you can say you can use N in the Poisson as part of calculating lambda, but you can't say that N is a parameter in the Poisson.

D*** crystal ball. Another wrong prediction!

Walter, do you see what's really going on with your tail? (No, you don't need a mirror for that. :D )

Cheers,

[Thanz]

Originally posted by T'ai Chi
Hold on a sec here. :)

The sample size drops out of the probability function. In the Poisson, the sample size, N, is not a parameter.

N is used to estimate lambda, and lambda is used for the Poisson distribution, but this estimation of lambda can be done outside of the Poisson distribution.

So, I guess you can say you can use N in the Poisson as part of calculating lambda, but you can't say that N is a parameter in the Poisson.
I had a post that said something like this in the works, but now I can drop it. I think that Lurker is just saying that BillHoyt couldn't calculate the expected number of J guesses without knowing the total number of guesses.

In his crow example, if he knew that the expected number of crows were 5 per acre, and he was shown a field with 9 crows, he would not be able to do his calculations if he also did not know the size of the field with 9 crows.

But anyway, I doubt the applicability of the Poisson here. If JE just used J's, that would be OK. However, why no one is interested in the other high frequency letters JE uses is beyond me.
Hey - I am interested in other letters!

[Walter Wayne]

Originally posted by BillHoyt
Let us say the average number of dead crows per acre is known to be 5. Let us go to Montana and set up 1 acre grids and count in each grid. We see one grid with 9 dead crows and wonder what is the one-tailed probability of that high (or higher) a count. We use a mean of 5, and look at the cdf for >=9. And we get .03.Here you are not doing a binomial approximation (success=1, failure=0), just a count. However, one this is still parameterized by area and acres=1 in this case. If acres=2 then one would use a mean of 10.Now let's go to counting initials. We pick an initial that has a frequency of .5. We count 9 such initials in a field of 10. We use a mean of 5 and look at the cdf for >= 9. And we get .03.

Now let's pick an initial that has a frequency of .05. We count 9 such initials in a field of 100. We use a mean of 5 and look at the cdf for >= 9. We get .03.The mean contains information on poth p and N. N may affect mean, but the mean is not enough to determine N.N, sir, was incidental. Poisson is not affected by it one iota.Would you argue that if y=f(ab), that a is "incidental" to y and that y is not affected one iota. If y=f(ab) then it is sufficient to now ab, but neither a nor b or incidental.

In our case p(x)=f(pN), knowing pN is sufficient, but neither p nor N are incidental.

Walt

Edit: Looks like T'ai Chi answered this before me. However, I disagree with the wording that N "drops out" of the relationship. In our particular case we have set p=0.1336. N is still in the equation. If p = 5/N then N would indeed drop out.

[BillHoyt]

Originally posted by Thanz

I had a post that said something like this in the works, but now I can drop it. I think that Lurker is just saying that BillHoyt couldn't calculate the expected number of J guesses without knowing the total number of guesses.
No, I think the mistake is deeper than that. Otherwise, he would not be construing the Poisson cdfs as "errors" when erroneously contrasting them with his Ns.

[Lurker]

Yes, that is one of the points I am trying to make. Bill could not calculate the mu without n.

And bill, if calling me a totle makes you happy then go right ahead.

And your crow example is still wrong. I suggest you examine it more closely to see your error. With some thought on your part it should become painfully obvious.

Lurker

[BillHoyt]

Originally posted by Walter Wayne
If y=f(ab) then it is sufficient to now ab, but neither a nor b or incidental.

In our case p(x)=f(pN), knowing pN is sufficient, but neither p nor N are incidental.

Walt

If y=f(ab) then y=f(x), and the a and b are incidental. If y=f(a,b), then you have a different story. For Poisson, p(x)=f(mu). Mu is the only thing that is important. Poisson doesn't care how you got the mu.

[Thanz]

Originally posted by BillHoyt

No, I think the mistake is deeper than that. Otherwise, he would not be construing the Poisson cdfs as "errors" when erroneously contrasting them with his Ns.
Regardless, is the rest of my post correct?

You could no more do the analysis of the J guesses without knowing the total number of guesses than you could do the analysis of crows without knowing the size of the field. Correct?

[Lurker]

All right Bill, you keep avoiding this example.

When p=0.9 in the census fo r"J" names. And JE uses 85 initial guesses. What is the probability that in those 85 guesses, that JE will get 85 or less "J" names? 84 or less?

Lurker

[Walter Wayne]

Originally posted by BillHoyt


D*** crystal ball. Another wrong prediction!

Walter, do you see what's really going on with your tail? (No, you don't need a mirror for that. :D )

Cheers, My tail? Are you refering to the fact that the poisson approximation does not yield a 0 pdf for values of x>N. That is why approximations are so fun :)

[Lurker]

Originally posted by Walter Wayne
My tail? Are you refering to the fact that the poisson approximation does not yield a 0 pdf for values of x>N. That is why approximations are so fun :)

And we run full circle to why I wondered about the accuracy of the Poisson Distribution model being valid. I really just questioned it by attempting to show its limitations which Bill refused to acknowledge. Bill seems to defend it with a religious fervor I see only at a Benny Hinn revival.

Since then, others have shown that for the original problem it was fairly accurate which I agreed to. What I find humorous is Bill still insists my questions and demonstrations of the limitations of Poisson are somehow invalid.

Interesting the mind of Bill the Believer.

Lurker

:D :D :D :roll: :roll: :)

[BillHoyt]

Originally posted by Lurker


And we run full circle to why I wondered about the accuracy of the Poisson Distribution model being valid. I really just questioned it by attempting to show its limitations which Bill refused to acknowledge. Bill seems to defend it with a religious fervor I see only at a Benny Hinn revival.

Since then, others have shown that for the original problem it was fairly accurate which I agreed to. What I find humorous is Bill still insists my questions and demonstrations of the limitations of Poisson are somehow invalid.

Interesting the mind of Bill the Believer.

Lurker

:D :D :D :roll: :roll: :)

You and Walt are running in circles. I still hold out hope Walt will get it.

[Lurker]

From http://www.math.uiuc.edu/~hildebr/361/feb14.html

Math 361 at Univ Illinois

"I introduced two important approximations to the binomial distribution, the Poisson approximation and the normal approximation. The Poisson approximation is best suited in the case of rare events and a large number of trials. while the normal approximation works well when the number of trials is More precisely, in order for the Poisson approximation to be accurate, p should be very small, n should be large (ideally of order 1/n), and k should be small (which it almost always is, since the cases of interest are those when k is 0, 1, 2, or some other very small number). If these conditions are not satisfied, the Poisson approximation should not be applied, and it may be completely off."


Bill, what was I questioning about number of trials, probability all effecting the accuracy of Poisson?

Lurker

[Thanz]

Mr. Hoyt -

You have done a whole lot of crowing in this thread, but you have still neglected to answer my question which I believe goes to the heart of the validity of your analysis. Here it is again, in case you missed it:I would like to know why you think it is appropriate to count "And they're also talking about somebody who would be known as either Richard or Rich, because a big R-connection that comes up connected to you" as more than one guess. In my count, I counted this as one R guess. From your description, I gather you counted it as 3 R guesses.

[Lurker]

Thanz:

Bill sure is a good dancer. It is hard to keep hopping about refusing to answer reasonable questions but he seems to have mastered it.

Lurker

[BillHoyt]

Originally posted by Lurker
From http://www.math.uiuc.edu/~hildebr/361/feb14.html

Math 361 at Univ Illinois

"I introduced two important approximations to the binomial distribution, the Poisson approximation and the normal approximation. The Poisson approximation is best suited in the case of rare events and a large number of trials. while the normal approximation works well when the number of trials is More precisely, in order for the Poisson approximation to be accurate, p should be very small, n should be large (ideally of order 1/n), and k should be small (which it almost always is, since the cases of interest are those when k is 0, 1, 2, or some other very small number). If these conditions are not satisfied, the Poisson approximation should not be applied, and it may be completely off."


Bill, what was I questioning about number of trials, probability all effecting the accuracy of Poisson?

Lurker

You really don't understand that paragraph, do you? Go back over it. The Normal distribution is also inaccurate? Huh?

The discussion there, sir, is about using Poisson to approximate binomial. It does not say Poisson is a lesser distribution any more than it says the Normal is a lesser distribution because of the inaccuracies when the normal is used to approximate binomial!

Wow! We have been through this before. Poisson is not "binomial lite". Normal is not "binomial lite". These sources discuss how accurately Poisson and Normal model Binomial.

Now you get my monkey...
http://www.baltobluegrass.com/bbggraph/monkey.gif

Please stop. You are merely demonstrating how militant you can be in your ignorance. Even the monkey is shaking his head.

[Instig8R]

Originally posted by neofight


Concerning anything substantive? No. The only thing that I conceded was that it appeared JE might have been wrong in a part of his interpretation about why the "Malibu Shrimp" recipe was kept "secret".

He interpreted it as meaning that it was secretly based upon one of Deborah's mother's recipes, which he believed was what he was being shown him by her mom, and Deborah said the real reason it was kept secret was because of the questionable clams that she and her friend used when making it, although she did admit that the recipe was based, at least loosely, upon her mom's recipe.

Instig8R feels JE pressured Deborah into admitting that fact, and considers this is a monumental matter. I feel that it could indeed be true, that the recipe was loosely based on one of Deborah's mother's recipes, since most daughters do tend to learn some cooking from their moms.

In any case, I also feel that this whole matter was blown way out of proportion by Instig8R, since the balance of that reading stands, and it was a good reading, with a lot of excellent, accurate hits, and in no way does any of that hinge upon this one, rather irrelevant point.......neo

Hey, neo-- Perhaps the following will help jog your memory... It is our dialogue on 6/25/02 (before you saw the edited reading on television):

(Instig8R)
"Now, I wonder why he didn't ask in what way the sitter's mom contributed to that recipe..."

(neofight):
"Why is that even important? Basically it's been said already. I think it's clear from the context that the woman and her friend simply took one of her mother's recipes that she probably really liked, and improved upon it, adding various ingredients and truly making it a culinary delight. They were probably very proud of the results, and were loath to admit it was not entirely theirs. I would be surprised if there was anything more to the story than that, but it would be nice to find that out for sure....neo"

You seem to have changed your opinion since then. You argued quite vigorously as to the meaning of that hit in June 2002. The "hit" sure seemed relevant in June, 2002. Funny how viewing the edited reading affected your memory and your interpretation of the live reading, isn't it? Memory is a funny thing, isn't it? :)

The point is that JE said that he had Deborah's friend (Helen) and Deborah's mother, together. Mom showed up, specifically to call attention to that stolen recipe. Remember? The recipe was based on something that Mom made, and she never got the credit that she deserved. The point is that JE pressured Deborah into a false validation. She never stole a recipe from her mother, and all those false accusations were edited out of the final reading.

You were so sure, in June 2002, that Deborah would NOT have acknowledged a lie, and you seemed very impressed that JE was let in on that little secret. Yeah, Deborah was busted by the spirits of her mom and her friend. If not for editing, this reading would not have been fit for broadcast.

[Lurker]

Originally posted by BillHoyt


Wow! We have been through this before. Poisson is not "binomial lite". Normal is not "binomial lite". These sources discuss how accurately Poisson and Normal model Binomial.

Now you get my monkey...
http://www.baltobluegrass.com/bbggraph/monkey.gif[/url]

Please stop. You are merely demonstrating how militant you can be in your ignorance. Even the monkey is shaking his head.

Have you derived the Poisson Distribution, Bill? I could be wrong but isn't it derived by taking the limit of the Binomial Distribution as n=>infinity? What implication does this have?

Lurker

[CFLarsen]

Originally posted by Instig8R
The point is that JE pressured Deborah into a false validation.
...
If not for editing, this reading would not have been fit for broadcast.
Two points, impossible to argue against.

[TLN]

Originally posted by CFLarsen
Two points, impossible to argue against.

We'll see... ;)

[Walter Wayne]

Bill,

I think I see what you are getting at. Based on 1 trial, we expect JE to guess J 0.1336 times. If on average over our tests JE guesses J 0.25 times/trial then the p value will be identical.

You are normalizing the units of the average and sample. Instead of comparing the expected 0.1336 J's/trial compared to found 25 J's in 100 trials, you are comparing expected 0.1336 J's/trial to 0.25 J's/trial.

Walt

Edit: My post has some redunancies and repeats itself.

[T'ai Chi]

Originally posted by Thanz

Hey - I am interested in other letters!

Sorry Thanz!!

I should have specified that only Bill is not interested in analyzing the other high frequency letters (at least according to his analysis that only focuses on the high frequency letter J).

[Walter Wayne]

Originally posted by T'ai Chi


Sorry Thanz!!

I should have specified that only Bill is not interested in analyzing the other high frequency letters (at least according to his analysis that only focuses on the high frequency letter J). Actually, somewhere in this thread he mentions that one could also check the theory by seeing if low frequency letters were under represented. I'd quote it, but it is a long thread.

Walt

[T'ai Chi]

I agree with that WW.

My point is that Bill's analyses only consider one letter at a time. He knows that doing the same test multiple times for the letters of interest introduces more error than doing one test that tests several letters at once. If he was interested in testing multiple letters, I'd think he'd find a more appropriate analysis.

[69dodge]

Originally posted by BillHoyt
The discussion there, sir, is about using Poisson to approximate binomial.Certainly.

It is my impression that the discussion here is about the same thing.

[69dodge]

Originally posted by Walter Wayne
I think I see what you are getting at. Based on 1 trial, we expect JE to guess J 0.1336 times. If on average over our tests JE guesses J 0.25 times/trial then the p value will be identical.

You are normalizing the units of the average and sample. Instead of comparing the expected 0.1336 J's/trial compared to found 25 J's in 100 trials, you are comparing expected 0.1336 J's/trial to 0.25 J's/trial.Yes.

The sum of a bunch of independent random variables, all of which have Poisson distributions, itself has a Poisson distribution. The mean of the sum is the sum of the means, which don't have to be identical, though here they are.

Edited to add: I'm not sure what I was thinking when I wrote the above. It's correct, but it's got nothing to do with what you wrote.

I guess I don't quite understand what you wrote, actually. Can you elaborate?

[69dodge]

Originally posted by BillHoyt
Poisson doesn't care about n. If we look at an acre of land and find 12 dead crows, what was the N? Who knows, who cares?The n is the number of crows in the world that could possibly have ended up dead on that acre of land. And p is the probability that any given one of them actually would do so.

We don't care about the exact value of n, but only because we do know that it is much larger than 12.Poisson is not parameterized on N.Quite true. However, if we're considering using a Poisson distribution with parameter np to approximate a binomial distribution with parameters n and p, we should ensure that n is sufficiently large and p is sufficiently small. Otherwise, the approximation will be very poor.

In all the cases that I know about where a Poisson distribution is used, the theoretical justification is that it is an approximation of an underlying binomial distribution with large n and small p. Does anyone know of a different reason why we should expect a physical system to have a Poisson distribution?

[69dodge]

Originally posted by BillHoyt
Now let's go to counting initials. We pick an initial that has a frequency of .5. We count 9 such initials in a field of 10. We use a mean of 5 and look at the cdf for >= 9. And we get .03.Based on your figure of 0.03, I believe you meant '>', not '>='. I will use '>' in the rest of my reply.

I'm not sure what you are claiming. You give an answer of 0.03. But what is the question? I see two possibilities. Let X be a random variable with a Poisson distribution whose mean is 5. What is the probability that X > 9 ?

The answer to this question is 0.0318.

I pick ten letters at random. Each has, independently, a probability of 0.5 of being a J. Let Y be the number of J's that I pick. What is the probability that Y > 9 ?

The answer to this question is 0.000977.

That the answers differ can mean only one thing: Y does not have a Poisson distribution with mean 5. In fact, it has a binomial distribution with n = 10, p = 0.5.Now let's pick an initial that has a frequency of .05. We count 9 such initials in a field of 100. We use a mean of 5 and look at the cdf for >= 9. We get .03.Again, I see two possibilities for what you are asking. same question as before.

same answer as before.

I pick 100 letters at random. Each has, independently, a probability of 0.05 of being a J. Let Z be the number of J's that I pick. What is the probability that Z > 9 ?

The answer to this question is 0.0282.

Again, the two answers differ. And again, this is because Z does not have a Poisson distribution with mean 5. It has a binomial distribution with n = 100, p = 0.05.

However, the two answers are closer than they were before. This is an illustration of the fact that, as n increases and p decreases, with np remaining constant, the binomial distribution with parameters n and p approaches the Poisson distribution with mean np.

[Lurker]

Originally posted by 69dodge
The n is the number of crows in the world that could possibly have ended up dead on that acre of land. And p is the probability that any given one of them actually would do so.

We don't care about the exact value of n, but only because we do know that it is much larger than 12.Quite true. However, if we're considering using a Poisson distribution with parameter np to approximate a binomial distribution with parameters n and p, we should ensure that n is sufficiently large and p is sufficiently small. Otherwise, the approximation will be very poor.

In all the cases that I know about where a Poisson distribution is used, the theoretical justification is that it is an approximation of an underlying binomial distribution with large n and small p. Does anyone know of a different reason why we should expect a physical system to have a Poisson distribution?

You write much clearer than I did but that is what I was trying to get across in my posts which a certain other poster continually argued around.

Lurker

[Thanz]

Originally posted by T'ai Chi


Sorry Thanz!!

I should have specified that only Bill is not interested in analyzing the other high frequency letters (at least according to his analysis that only focuses on the high frequency letter J).
That could be because we don't have enough data to do a more comprehensive test on multiple letters. Of course, Mr.Hoyt won't admit this, and will stick to his flawed J analysis, as it may mean admitting that I was correct when I said the sample size was too small.

[BillHoyt]

Originally posted by 69dodge
In all the cases that I know about where a Poisson distribution is used, the theoretical justification is that it is an approximation of an underlying binomial distribution with large n and small p. Does anyone know of a different reason why we should expect a physical system to have a Poisson distribution?
We are not trying to approximate binomial with Poisson here. We are simply using Poisson. Period. Yes, radioactive decay is a physical system that has a Poisson distribution.

[Thanz]

Originally posted by BillHoyt

We are not trying to approximate binomial with Poisson here. We are simply using Poisson. Period.
Why? What makes Poisson the appropriate tool here?

[Lurker]

Originally posted by BillHoyt

We are not trying to approximate binomial with Poisson here. We are simply using Poisson. Period. Yes, radioactive decay is a physical system that has a Poisson distribution.

And radioactive decay has a near infinite N (atoms) and a small p. I wonder why Poisson is applicable?

Bill, did you find out how Poisson was derived yet? Have you addressed dodge's binomial comparison yet?

Lurker

[Thanz]

Hey! New page - 21!

Since Mr. Hoyt still hasn't answered my question regarding his counting methods, I'll repost it on the new page so he doesn't have to go looking for it.

I would like to know why you think it is appropriate to count "And they're also talking about somebody who would be known as either Richard or Rich, because a big R-connection that comes up connected to you" as more than one guess. In my count, I counted this as one R guess. From your description, I gather you counted it as 3 R guesses.

[SteveGrenard]

I once saw a list of all the names thrown out by a medium and tabulated by Michael Shermer. There were something like 40 names and he said these all happened during the course of a seminar several hours long. This was a misleading statement since half of them were the same name and about a 1/4 were phonetically alike. This exercise proves nothing in such cases.

Richard = Ricky = Rick = Dick = Dicky = Rich = Ritchie

Isnt it possible someone with the name Richard has been called by different people at different times all of those nicks?

[BillHoyt]

I'll repeat my previous post, and add some clarification:
I re-worked the transcripts and came up with different results and a different method. Here it is, in a nutshell:

1. I used the census data figures orginally presented, although these may need tweaking.
2. I excluded the CO show data, and concentrated solely on the available, unedited transcripts from LKL, etc.
3. I looked at JE's style and adjusted the counting procedure as follows:
o I counted all of his name guesses
o Whether he stated them as names or initials, I counted them
o I excluded impossible-to-deal-with things such as "a B softened by a vowel," and chalked that up to a "B" guess.
o I included even bizarre names such as "pepper", "salt", "brooklyn" and other nickname guesses, except that
o I only counted "Liz", "Elizabeth" type guesses as the full given name, and did not also count an "L". but
o When JE recited a littany of names, I counted each one, whether they had the same initial or differing initials (again excluding the "Liz/Elizabeth, Ronny/Ronald, and Bill/William" type guesses, where I only counted the intial of the full given name.

Sound a bit complicated? You should read the transcripts. I could not see another way to approach things fairly given that sometimes he was all over the board. My hypothesis was, that, if there is a JE mediumship process, i should honor as much of it as I could figure in making the counting rules.
Regarding the multiples, I realize this post didn't full explain my reasoning. Sometimes, JE calls out a name or an initial for a person who passed, sometimes for somebody else still living, sometimes he uses the same initial for possibly two different people. Sometimes the names he tosses out are impossible to categorize. "Helen , Ellen or L-name" is a prime example. What the heck do you do there? The first one was H, the second E, the third goes to L. Does he mean L preceded by something? Or L followed? He does the same thing with the "B softened by a vowel" thing. Sometimes he mentions simply a vowel in front. Other times he precedes that vowel with a consonant.

There are three choices as I see them: drop such guesses entirely and thereby lose most of the data, pick one letter and run with it (but which one?) or make a uniform rule allowing him as much leeway as you can. I said I tried to honor his process as much as possible, and I decided to grant him the leeway. The specific leeway I did not grant was "Rich or Richard", where one is clearly a nickname for the other. In those cases, I counted one guess. If he said "Rich or Richard or R" I counted two guesses.

With 26 available letters, and with a focus on just "J", the major distortion I expected was a blossoming denominator. (Clearly, this did happen.) The overall effect I expected was the "J"s would either barely hold onto their significance or that they would be washed out by all the multiple Ds and Rs and Ms and Bs, etc.

[Lurker]

Bill,

Have you had a chance to look into the derivation of the Poisson yet?

For fun, try calculating the odds for getting a heads by flipping a coin three times. List the odds for each total heads as 0 times, 1 time, 2 times and 3 times.

Here is what I got via a probability tree which has zero error:

0 0.125
1 0.375
2 0.375
3 0.125

Making it a cumulative we get

0 0.125
1 0.5
2 0.875
3 1.0

Now I plug this problem into Poisson using mu=np=3*0.5=1.5

0 0.22313016
1 0.5578254
2 0.808846831
3 0.934357546


But wait, Bill, the numbers are different. Why is that? I have provided the EXACT solution and good old Poisson is not very close. Why is that?

Lurker

[BillHoyt]

Originally posted by Lurker
But wait, Bill, the numbers are different. Why is that? I have provided the EXACT solution and good old Poisson is not very close. Why is that?

Lurker

http://www.baltobluegrass.com/bbggraph/monkey.gif
Surprising that you misapply Poisson and see an error?

[Thanz]

Originally posted by BillHoyt
Regarding the multiples, I realize this post didn't full explain my reasoning. Sometimes, JE calls out a name or an initial for a person who passed, sometimes for somebody else still living, sometimes he uses the same initial for possibly two different people. Sometimes the names he tosses out are impossible to categorize. "Helen , Ellen or L-name" is a prime example. What the heck do you do there?
I focussed on how wide a net he was casting. We have already debated whether it is appropriate to consider a guess of simply "John" as equivalent to "J". For the purposes of my count, I considered them equal, although I think a strong argument exists that they are not.

For this specific example, I considered it to be 3 guesses - as he would accept any similar H, E, or L name as a "hit". In this instance, he is using 3 letters to guess one person, so I felt it appropriate to count all three.

The specific leeway I did not grant was "Rich or Richard", where one is clearly a nickname for the other. In those cases, I counted one guess. If he said "Rich or Richard or R" I counted two guesses.
This is where we differ. I consider that to be one guess - as only one intitial letter is involved. He will accept any R name, and he is hoping to be close with Richard. But he is not making two guesses - the net stays only as wide as "R". As we are considering only the letters involved, I considered this to be one guess of the letter R. He is NOT casting 2 R nets here, which is what your count would seem to say.

With 26 available letters, and with a focus on just "J", the major distortion I expected was a blossoming denominator. (Clearly, this did happen.) The overall effect I expected was the "J"s would either barely hold onto their significance or that they would be washed out by all the multiple Ds and Rs and Ms and Bs, etc.
However, that clearly did not happen. Adding in extra "J" guesses has more of an effect than adding in extra letters to the denominator. We have radically different counts. I don't know exactly what transcripts you used, but I used the LKL transcripts posted by Renata and got 9 J guesses out of a total of 43, compared to 18 and 85 for you.

When Kerberos did his count, he counted 78 guesses and 14 Js. The difference between his and yours are 4 Js and 7 total guesses. But the 7 extra total only move the expected by about .6, but the extra 4 Js makes all the difference in whether we can reject the null hypothesis.

[Lurker]

Originally posted by BillHoyt


Surprising that you misapply Poisson and see an error?

Hey, now we are getting somewhere! Finally! Now, how about you tell us when Poisson will be accurate and when it will not.

Thanks!

Lurker

[CFLarsen]

Originally posted by SteveGrenard
I once saw a list of all the names thrown out by a medium and tabulated by Michael Shermer. There were something like 40 names and he said these all happened during the course of a seminar several hours long. This was a misleading statement since half of them were the same name and about a 1/4 were phonetically alike. This exercise proves nothing in such cases.

Richard = Ricky = Rick = Dick = Dicky = Rich = Ritchie

Isnt it possible someone with the name Richard has been called by different people at different times all of those nicks?

Can we see this list, Steve?

[Lurker]

Bill:

I would have thought a math genius like yourself would have been happy as a clam to dive into the derivation of the Poisson Distribution. I wonder why you haven't?:wink:

Could it be because of its relation to the binomial...:D

Lurker

[BillHoyt]

Originally posted by Lurker
Bill:

I would have thought a math genius like yourself would have been happy as a clam to dive into the derivation of the Poisson Distribution. I wonder why you haven't?:wink:

Could it be because of its relation to the binomial...:D

Lurker

Lurker,

You are an absolutely waste of time on this issue. Poisson can be derived from binomial. It can also be derived from geometric. It can also be derived by solving the differential equation:

f(t) = q*(1 - f(t))

The last derivation is from basic radioactive decay.

So what if they can be derived from one another? Don't you understand the Central Limit Theorem?

Now go away.

[T'ai Chi]

I guess we should all use the same readings, and we all should count the J-name occurances, and see what we all get.

Oh, in my opinion Bill probably thinks that JE just does cold readings with the letter J or something.

[Lurker]

Originally posted by BillHoyt


Lurker,

You are an absolutely waste of time on this issue. Poisson can be derived from binomial. It can also be derived from geometric. It can also be derived by solving the differential equation:

f(t) = q*(1 - f(t))

The last derivation is from basic radioactive decay.

So what if they can be derived from one another? Don't you understand the Central Limit Theorem?

Now go away.

My, touchy, aren't you. :) And what does one assume when one derives it from those equations? Hmm? You always seem to avoid that question. Why is that, Bill?

This is getting humorous as Bill frantically tries to avoid giving me credit for any shred of knowledge. It is so much easier for him to debate with ad homs.

Lurker

[Lurker]

Come on, Bill. Enlighten us as to when it is applicable to use Poisson? Others have asked. Seems pretty simple. Why don't you answer?

Is it valid for all values of p?

Is it valid for any population size?

Simple questions for someone as knowledgable as you clearly are. And I do mean that. You clearly understand more about statistics than I do. I fully grant you that. So tell me when Poisson is applicable and when it is not. Two simple questions above.

Lurker

[CFLarsen]

Lurker,

Do you think it would be prudent for people to have a chance to be able to get back to you, before you chastise them for not answering?

Not everyone is on JREF 24/7, you know.

[T'ai Chi]

CF, what do you think of just looking at the J's?

[CFLarsen]

Originally posted by T'ai Chi
CF, what do you think of just looking at the J's?

I'll await the statistical experts.

[TLN]

Originally posted by CFLarsen
I'll await the statistical experts.

See if Girl 6 will help out.

[CFLarsen]

Originally posted by TLN
See if Girl 6 will help out.

I said "statistical", not...oh. OK. :)

[Thanz]

Originally posted by CFLarsen


I'll await the statistical experts.
Why the cop out, Mr. Larsen? You were all over me with stats questions earlier in the thread.

You don't have to be a stats expert to answer some of the basic questions here.

1. Do you think that an analysis of only one letter is meaningful to ascertain whether or not JE is cold reading? If someone were to ask you what you thought was a good test for cold reading comparison, would it be test of one letter or multiple letters?

2. What do you see as the proper counting technique for the following JE (style not a quote) guess: "I am getting a John or Joe or some other J, J-o connection here". How many J guesses would you count out of that?

[TLN]

Originally posted by Thanz
Why the cop out, Mr. Larsen?

Because he's not a statistician and neither are you. Therefore, neither of you are qualified to say what would be statistically significant in this example.

When you've got nothing left but bashing folks based on admissions of ignorance you've got nothing left at all.

[CFLarsen]

Originally posted by Thanz
Why the cop out, Mr. Larsen? You were all over me with stats questions earlier in the thread.

Not a cop out. But I can spot a fake a mile away. I spotted you. And I've watched you, "grilling" BillHoyt on statistcal matters. You're not fooling anyone, you know. You have no idea what you are talking about.

Originally posted by Thanz
You don't have to be a stats expert to answer some of the basic questions here.

I prefer to leave the expert question to the experts. What's wrong with that? You should try it, you know...

[Lurker]

Originally posted by CFLarsen
Lurker,

Do you think it would be prudent for people to have a chance to be able to get back to you, before you chastise them for not answering?

Not everyone is on JREF 24/7, you know.

Claus,

I have been asking the same questions, or variations for over a week now with no answers from Bill. He refuses to answer direct questions. You would know that if you had bothered to read the thread.

Lurker

[Thanz]

Originally posted by TLN
Because he's not a statistician and neither are you. Therefore, neither of you are qualified to say what would be statistically significant in this example.
I am not asking him to give me the derivation of Poisson, or how we should set significance levels, or anything else. I am NOT asking him what is "statistically significant".

I am asking about simple test design. If he doesn't think he is qualified to comment on any of this, why did he attack me earlier? He was asking some very pointed statistically based questions earlier. Much more statistically based than my questions. Shouldn't he have left that to the experts?

Also, since when does someone have to be a statistician to make a comment here? Is Mr. Larsen an expert on fraud? On cold reading? On speaking to the dead? On scientific design? He comments on all of these things.

The questions I have asked are extremely basic, and if Mr. Larsen is as serious as he claims to be about paranormal research he should be able to answer them easily. I chalk his refusal up to simple avoidance. Since when does one need be a statistician to have an opinion on how to count J guesses?

[Thanz]

Originally posted by CFLarsen
Not a cop out. But I can spot a fake a mile away. I spotted you. And I've watched you, "grilling" BillHoyt on statistcal matters. You're not fooling anyone, you know. You have no idea what you are talking about.
How can you know this? You are no "expert" by your own admission..... Oh, I know. Maybe it is because I admitted that I was not an expert. I am not trying to fool anyone, doof.

I prefer to leave the expert question to the experts. What's wrong with that? You should try it, you know...
I am not asking expert questions. I am asking basic test design questions. Is it more meaningful to test all letters or one letter? How should we count a guess of "J"?

You seem to think that you are qualified to comment on the validity of Schwarz's experiments and to design and conduct your own experiments into whether JE is talking to the dead. What expertise allows you to do that, but not answer my simple questions?

[CFLarsen]

Thanz,

Let's see. You are only arguing basic test design questions and not asking expert questions?

Originally posted by Thanz
Needless to say, I get numbers substantially different than BillHoyt's. In the LKL readings posted by Renata, I counted 43 guesses, of which 9 were J. If I plug this into the Poisson calculator, I get a probability of >= 9 of .128, which means that we cannot reject the null hypothesis.
...
One more thing - Kerberos' count of hits was 14 J hits out of 78 guesses. If I pop those numbers into the ol' Poisson calculator, I get a probaility of >= 14 of .168, which also means that we cannot reject the null hypothesis.
...
I am not ignoring the fact that the method was applied equally to all initials. I am saying that the method is incorrect, and leads to incorrect counts for both the J value AND the denominator. Having both numbers wrong leads to a wrong analysis. It just means that ALL of your data is incorrect. I was not accusing you of bias in the data - in that you would count only extra "J" guesses. I am saying that your methodology was flawed from the beginning.
...
That could be because we don't have enough data to do a more comprehensive test on multiple letters. Of course, Mr.Hoyt won't admit this, and will stick to his flawed J analysis, as it may mean admitting that I was correct when I said the sample size was too small.
...
Why? What makes Poisson the appropriate tool here?


These questions are not about test design but definitely expert questions about statistics. And you are definitely a fake, when it comes to statistics. When reading your posts here, it is obvious you take your leads from other posters. Did you know what Poisson was before this thread started?

My "expertise" in commenting on Schwartz' abominations and design my own experiments comes from my studying what has been done, where the flaws are, and how it can be improved.

In other words, I read and learn.

What questions of yours have I not answered?

[T'ai Chi]

Originally posted by CFLarsen

I'll await the statistical experts.

I'll give you an answer: readings are done with more than one letter. Therefore our statistical analysis should reflect that, and we should look at more than one letter. It seems sensible to analyze all the high frequency letters together (there are about 5 or 6 of them).

Bill's counting of the J's and focusing solely on the J, is probably consciously or unconsciously a wish for a rejection of a null hypothesis. I wonder if any statistician would say the analysis is appropriate.

[Thanz]

Originally posted by CFLarsen
Thanz,

Let's see. You are only arguing basic test design questions and not asking expert questions?
Can't you "read and learn" in this thread? Nothing in those posts requires any massive amounts of statistical understanding. The counting method I used to get the counts in the first quote was explained in that post. I thought it was simple and easy to understand. As to the numbers from the Poisson calculation, I used a calculator available on the web and reported the results. Nothing magical here either.

The third quoted portion deals with the counting method, and I am coming to the stunning conclusion that if your counting method is flawed, your data will also be flawed, and you will not get an accurate analysis from flawed data. Is that too hard for your non-expert brain to comprehend?

Finally, asking why he used a certain distribution (Poisson) instead of others is a simple question for someone who understands stats (as BillHoyt claims to) to answer. He picked hte test. I am just asking why.

These questions are not about test design but definitely expert questions about statistics. And you are definitely a fake, when it comes to statistics. When reading your posts here, it is obvious you take your leads from other posters. Did you know what Poisson was before this thread started?
It is called "reading and learning". Of course I take leads from other people in this thread that understand stats better than I do. I'd be an idiot if I didn't. Also, none of those questions were aimed at you. Nice try at diversion. I didn't ask YOU whay Poisson was appropriate, I asked BIllHoyt. I am asking not to trap him, but because I don't know why he chose Poisson. Other people here seem to disagree with him on his chosen test.

As for your specific question, no I did not know what a Poisson distribution was before this thread. I may have heard the term back in my undergrad days, so it was a little familiar, but if pressed I would not be able to explain it. But I ask you, so what? I have learned in this thread how to use the Poisson calculator and apply it to data as presented. I have a limited understanding of what the output of the calculator tells me (I can apply it to the test Mr. Hoyt did here).

My "expertise" in commenting on Schwartz' abominations and design my own experiments comes from my studying what has been done, where the flaws are, and how it can be improved.

In other words, I read and learn.
Just as I have done here.

What questions of yours have I not answered?
Do you really not know? They were specifically directed to you. Here they are again:

1. Do you think that an analysis of only one letter is meaningful to ascertain whether or not JE is cold reading? If someone were to ask you what you thought was a good test for cold reading comparison, would it be a test of one letter or multiple letters?

2. What do you see as the proper counting technique for the following JE (style not a quote) guess: "I am getting a John or Joe or some other J, J-o connection here". How many J guesses would you count out of that?

[BillHoyt]

Originally posted by Lurker
And what does one assume when one derives it from those equations? Hmm? You always seem to avoid that question. Why is that, Bill?


The definition of a Poisson distribution is based on that of a Poisson process. A Poisson process is one that satisfies the following:


1. The changes (or events) that result from the process can be grouped into nonoverlapping intervals.

2. The numbers of changes (or events) in the nonoverlapping intervals are independent from one another.

3. This independence holds for all intervals.

4. The probability of exactly one change (or event) in a sufficiently small interval, h = 1/n equals n*p , where p is the probability of one change (or event) and n is the number of trials.

5. The probability of more than one change (or event) in a sufficiently small interval, h, is essentially 0.

The Poisson distribution results when such a process occurs over n trials.

There are no hard and fast definitions to it, other than the above.

Lurker, you continue to insist that there is a question for me to answer about what one assumes when one derives Poisson from other distributions. This question makes no sense. The assumptions are grounded in the terms of those other distributions to derive Poisson, and are meaningless once you get to Poisson.

Poisson does not make an error when it predicts a p(x) beyond your n. The "n" is not in the Poisson distribution. Look closely at the definition above. "n" comes into play when you design your "bins" (the intervals). It affects your expected mean. Once you decide your expected mean, you have identified everything in Poisson. The "n" is now gone.

[CFLarsen]

Thanz,

There's a difference between learning and parroting. Because you can punch in numbers on a calculator does not mean you understand how the method works.

I have replied to the questions you refer to. I await the experts. Do you have a problem with understanding what people tell you?

[Thanz]

Originally posted by CFLarsen
Thanz,

There's a difference between learning and parroting. Because you can punch in numbers on a calculator does not mean you understand how the method works.
I guess that you are quite familiar with the difference between learning and parroting. In this thread, for instance, the questions that you asked me earlier were clearly "parroting". It is also obvious that you haven't been able to learn anything in this thread.

As for whether I understand how the method works, well, no, I can't derive the Poisson distribution, nor do I know the specific formula. I understand enough, however, to be able to do the various calculations and determine whether we can reject the null hypothesis. Which is apparently more than you have been able to pick up.

You see, at least I am trying to understand. You, however, are simply parroting Mr. Hoyt and avoiding questions that require no expertise in statistics at all. I guess that avoiding questions and giving non-responsive "answers" is your true expertise.

I have replied to the questions you refer to. I await the experts. Do you have a problem with understanding what people tell you?
I have a problem figuring out what part of "What do you see as the proper counting technique for the following JE (style not a quote) guess: "I am getting a John or Joe or some other J, J-o connection here". How many J guesses would you count out of that?" requires any expertise in statistics.

You are familiar with mediums. You are familiar with the guesses they make. Do you not know how to count? Is that it? Do you need a statistician to count for you?

[CFLarsen]

Thanz,

I see you do have a problem with what people tell you. Especially if you don't like it. I notice that you merely dismiss me, while admitting that you have, indeed, no idea what you are talking about. Thanks for proving my point.

You need to understand one thing, though: You do not control the life of other people. You do not have the right to command other people to do what you want them to do. And you have no right to point your finger at them, either.

Do you understand this simple thing, or do we have to go 2589 rounds before you get it?

Have a nice day.

[Thanz]

Originally posted by CFLarsen
I notice that you merely dismiss me, while admitting that you have, indeed, no idea what you are talking about.
I dismiss you as you seem completely unable to come up with any thoughts on the matter on your own. I have by no stretch admitted that I "have no idea what [I am] talking about". You, sir, are a liar. All I have done is admitted the limits of my knowledge, which is more than you have done.

You have simply parroted Mr. Hoyt while asking me questions, giving the impression that you have a clue about what the questions mean. Then, when asked some simple questions that do not require anything close to any expertise in statistics, you say "I'll wait for the experts". Which, in reality, I suspect means "I'll wait and say whatever BillHoyt tells me to say".

You need to understand one thing, though: You do not control the life of other people. You do not have the right to command other people to do what you want them to do. And you have no right to point your finger at them, either.
It will be a long, long time before I get over the irony of being told this by Mr. Larsen, who consistently hounds others and even starts threads with lists of questions that he demands other people answer.

Mr. Larsen, you dish it but you can't take it. You revel in making the lives of "believers" miserable on this forum (and probably other forums). For you to turn around and tell me that I "do not have the right to command other people to do what I want them to do" or "to point the finger at them" is quite simply the pinnacle of hypocrisy.

I trust that this point has been lost on you, however.

I notice, as well, that you can't even explain why you are waiting for the "experts" or what expertise my questions would require.

[CFLarsen]

Thanz,

Have I made claims about my statistical knowledge? Far from it, I believe. Do you understand the difference between making a claim and not making a claim? Here, on this board, claims are questioned. When people don't make claims, they should not be challenged to defend themselves. That's what you are trying to do here.

You do your best to cast doubt about me. You fail miserably, though. Why are you falling back on nasty innuendo about my behavior on other boards? Is that because you have run out of arguments?

You clearly don't like my answers. It is because they don't satisfy you. So, you feel entitled to call me a liar and start imagining my objectives.

Go on, Thanz. Imagine. It's what you do best.

[Lucianarchy]

Originally posted by Thanz

I dismiss you as you seem completely unable to come up with any thoughts on the matter on your own. I have by no stretch admitted that I "have no idea what [I am] talking about". You, sir, are a liar. All I have done is admitted the limits of my knowledge, which is more than you have done.

You have simply parroted Mr. Hoyt while asking me questions, giving the impression that you have a clue about what the questions mean. Then, when asked some simple questions that do not require anything close to any expertise in statistics, you say "I'll wait for the experts". Which, in reality, I suspect means "I'll wait and say whatever BillHoyt tells me to say".

It will be a long, long time before I get over the irony of being told this by Mr. Larsen, who consistently hounds others and even starts threads with lists of questions that he demands other people answer.

Mr. Larsen, you dish it but you can't take it. You revel in making the lives of "believers" miserable on this forum (and probably other forums). For you to turn around and tell me that I "do not have the right to command other people to do what I want them to do" or "to point the finger at them" is quite simply the pinnacle of hypocrisy.



That must be what they call a 'Hot read' then is it?

:roll:

[Thanz]

Originally posted by CFLarsen
Have I made claims about my statistical knowledge? Far from it, I believe.
Some level of statistical knowledge was implied when you were asking me specific statistical questions. Or are you admitting that you had no idea what those questions meant and you were just parroting BillHoyt?

Do you understand the difference between making a claim and not making a claim? Here, on this board, claims are questioned. When people don't make claims, they should not be challenged to defend themselves. That's what you are trying to do here.
Dude, I was just asking for your opinion on a couple of questions. Questions that do not require any expertise in statistics, just some logical and critical thinking skills. Since you can't answer them, I guess you don't have any of those skills.

You clearly don't like my answers. It is because they don't satisfy you.
I don't like your answers because they provide nothing of value to the discussion. Have you provided ANYTHING of value to this entire discussion? Or did you just want to jump in to ask me stats questions that you don't even understand?

Look, if you lack the logical and critical thinking skills to answer my questions, that's fine. Be a dolt. I don't care. Continue on your mission to fight against the "believers" no matter what they actually say.

[T'ai Chi]

Has anyone looked at other letters besides the letter J yet?

[CFLarsen]

Originally posted by Thanz
Some level of statistical knowledge was implied when you were asking me specific statistical questions. Or are you admitting that you had no idea what those questions meant and you were just parroting BillHoyt?

Oh, I had an idea, that's all. That idea was not unfounded. Why do you want to shift the focus from you to me?

Originally posted by Thanz
Dude, I was just asking for your opinion on a couple of questions. Questions that do not require any expertise in statistics, just some logical and critical thinking skills. Since you can't answer them, I guess you don't have any of those skills.

Did I make a claim? If so, what was it? If not, what right have you to demand an answer?

Originally posted by Thanz
I don't like your answers because they provide nothing of value to the discussion. Have you provided ANYTHING of value to this entire discussion? Or did you just want to jump in to ask me stats questions that you don't even understand?

Oh, come off it! You were cornered on statistical issues, and now it's all my fault. Pluhease...!

Originally posted by Thanz
Look, if you lack the logical and critical thinking skills to answer my questions, that's fine. Be a dolt. I don't care. Continue on your mission to fight against the "believers" no matter what they actually say.

Sure, go ahead with the personal attack, in the (vain) hope that your lacking statistical knowledge will not be noticed.

[Walter Wayne]

Originally posted by T'ai Chi
Has anyone looked at other letters besides the letter J yet? Yes. I looked at the three transcripts from Renata's thread. Counting Robbie, Rob or Robert as a single R, I found the following doing very rudimentry stats stuff.

6 J's of 28 is statistically high though not within p=0.05. The next question becomes, what letters where "sacrificed" in order to give more J answers? Were low percentage letters neglected more than the remaining high percentage letters?

Of the remaining 22
2 M's: about right
3 R's : slightly higher than expected
2 from the bottom 14 percentile (P,W,H,N,F,V,I,O,Y,Z,Q,U and X): slightly below expectation

The reason I lumped the bottom ones together is that each of them individual has an expected value less than 1. I chose 14% to lump together because that group of 13 letter the same probability as J alone.

The above are consistant with guessing only high percentage values in order to improve hit rates. J, M and R are guessed often at the expence of the unlikely letters. But nothing came anywhere near a p=0.05.

However, continuing down the high probability letters D, C, A, and S each have 1 guess. Then middling probality letters L and B have 4 and 3 guesses respectively. This appears contradictory to our original conclusion. Why guess L and B so frequently at the expense of some higher probability letters.

So, based on a very small number of readings, JE has affinities for J, M and R as expected and aversion to the low probability letters, also expected. However, he seems to have an "unnatural fetish" for L and B.

Admittedly, I used a less then perfect method for doing analysis on a group of related bins but for my curiousity purposes it was sufficient.

The reason I didn't post this earlier was:
1. I have a probablem with using census statistics. After all he is trying to contact/emulate dead people. So the statistic we should look at are for older people, not the population at large.
2. The number was so small from those three transcript, I didn't think the statistics meant anything anyways.
3. This thread had fallen off the front page, and I had hoped laid to rest. :p

My count from the transcripts, in order from most likely letters to least, was:
28 easily binned guesses
6 J
2 M
3 R
1 D
1 C
1 A
1 S
4 L
3 B
1 E
2 T
1 K
- G
1 P
- W
- H
1 N
- F, V, I, O, Y, Z, Q, U , X

8 guesses in "other"
2 J or G
1 C or K
1 H or E
1 E or L
1 vowel then B
1 vowel then L
1 "spice" name

[neofight]

Originally posted by Thanz
It will be a long, long time before I get over the irony of being told this by Mr. Larsen, who consistently hounds others and even starts threads with lists of questions that he demands other people answer.

Yes, Thanz. The irony is exquisite indeed. And "pinnacle of hypocrisy" does fit Claus quite well. :)

I trust that this point has been lost on you, however.


No doubt.......neo

[neofight]

Originally posted by Walter Wayne
The reason I didn't post this earlier was:
1. I have a probablem with using census statistics. After all he is trying to contact/emulate dead people. So the statistic we should look at are for older people, not the population at large.


Hi there! I'm not sure there is a good basis for saying this, Walter, since a good percentage of the spirit energies that come through belong to children and young people. :) .....neo

[Lurker]

Ah, finally Bill starts to answer some of the salient questiuons I have been asking for over two weeks!

Originally posted by BillHoyt


The definition of a Poisson distribution is based on that of a Poisson process. A Poisson process is one that satisfies the following:


1. The changes (or events) that result from the process can be grouped into nonoverlapping intervals.

2. The numbers of changes (or events) in the nonoverlapping intervals are independent from one another.

3. This independence holds for all intervals.

4. The probability of exactly one change (or event) in a sufficiently small interval, h = 1/n equals n*p , where p is the probability of one change (or event) and n is the number of trials.

5. The probability of more than one change (or event) in a sufficiently small interval, h, is essentially 0.

The Poisson distribution results when such a process occurs over n trials.

There are no hard and fast definitions to it, other than the above.

Lurker, you continue to insist that there is a question for me to answer about what one assumes when one derives Poisson from other distributions. This question makes no sense. The assumptions are grounded in the terms of those other distributions to derive Poisson, and are meaningless once you get to Poisson.

Poisson does not make an error when it predicts a p(x) beyond your n. The "n" is not in the Poisson distribution. Look closely at the definition above. "n" comes into play when you design your "bins" (the intervals). It affects your expected mean. Once you decide your expected mean, you have identified everything in Poisson. The "n" is now gone.

Bill, look at #1 through #5 again and pay particular attention to #4 and #5. Think about your sample size of 85. Consider the case where your sample size was 5 and the probability was 0.8.

Think about it Bill, Integrate your pdf to get your cdf for the range (0:5). When you complete this exercise, then the light may go on for you.

Lurker

[CFLarsen]

Originally posted by neofight
Hi there! I'm not sure there is a good basis for saying this, Walter, since a good percentage of the spirit energies that come through belong to children and young people. :) .....neo

Please point to those statistical analyses that show the age groups of "spirit energies" coming through.

Please either:

address the question, providing either a retraction or evidence of your claim, or
state that you refuse to answer.

[Thanz]

Originally posted by CFLarsen
Oh, I had an idea, that's all. That idea was not unfounded. Why do you want to shift the focus from you to me?
You did not have an idea. You simply copied BillHoyt. You do not actually KNOW what the questions you were asking meant. You simply wanted to harass me. Also, why do you want to shift the focus from you to me? We are discussing your lack of any ability to answer simple questions. I don't know why you keep bringing up my level of stats knowledge.

Did I make a claim? If so, what was it? If not, what right have you to demand an answer?
I asked because I thought you might have something to say about the topic, given your interest in the area of mediumship. You claim to know something about experimental design. You seem to claim that you are qualified to test JE to determine if he is a genuine medium. I am not sure why you think that you are qualified to do this, but have no idea how to count the number of J guesses in a transcript.

However, it appears I am mistaken. You do not want to answer my questions, probably because you do not want to go against BillHoyt and need to wait for him to tell you what to think. That's fine.

Oh, come off it! You were cornered on statistical issues, and now it's all my fault. Pluhease...!
No, I am simply pointing out the obvious:
1. You have not provided anything of value to this discussion.

2. You asked me a set of questions that you do not understand. Your motive seems to have been to harass me, as you would not understand the answers to the questions anyway.

Sure, go ahead with the personal attack, in the (vain) hope that your lacking statistical knowledge will not be noticed.
I am really not sure how my "lacking statistical knowledge" could NOT be noticed, considering that I have admitted the limits to that knowledge. I have, however, contributed positively to the discussion of this topic. I have performed the Poisson analysis to the count provided by Kerberos, and showed that on his count we could not reject the null hypothesis. I have done my own count of the J hits in Renata's JE LKL transcripts, and done the Poisson calculation on that count. On my count, we also cannot reject the null hypothesis according to the parameters set out by BillHoyt.

The only count that we can reject the null hypothesis is the count from BillHoyt. I have pointed out that Mr. Hoyt's counting procedure is not logical for the purposes of his test, and therefore his data is flawed.

What have you done? Nothing but harass me. When asked simple questions of logic, you hide. You are a poor skeptic, Mr. Larsen.

[BillHoyt]

Originally posted by Lurker
Ah, finally Bill starts to answer some of the salient questiuons I have been asking for over two weeks!
I have been answering. You still refuse to listen.
Bill, look at #1 through #5 again and pay particular attention to #4 and #5. Think about your sample size of 85. Consider the case where your sample size was 5 and the probability was 0.8.

Think about it Bill, Integrate your pdf to get your cdf for the range (0:5). When you complete this exercise, then the light may go on for you.

Lurker
N is not part of the distribution. Period. You have not found an error by finding probability beyond N. You simply make a laughable error. I am tired of saying it. YOU read my description and try to understand it this time.

[CFLarsen]

Thanz,

I did not want to "harrass" you. Please stop attributing motives to me I don't have.

I merely pointed out that you had no idea what you were asking BH about. Which was not all that far from the truth, was it?

If you feel I have "harrassed" you, I urge you to report me to the moderators, so we can clear this thing up. I, for one, would rather not have this serious accusation hanging over my head.

So, either report me, or stop crying about harrassment. Be true to your word or be a crybaby.

[Lurker]

Originally posted by BillHoyt

I have been answering. You still refuse to listen.

N is not part of the distribution. Period. You have not found an error by finding probability beyond N. You simply make a laughable error. I am tired of saying it. YOU read my description and try to understand it this time.

Integration will set you free, Bill. You DO know how to integrate, don't you?

Lurker

[BillHoyt]

Originally posted by Lurker


Integration will set you free, Bill. You DO know how to integrate, don't you?

Lurker

Laughingstock.

[Thanz]

Originally posted by CFLarsen
Thanz,

I did not want to "harrass" you. Please stop attributing motives to me I don't have.
Then what was your motive? The answers to the questions you were asking me would mean nothing to you. Actually getting the answers could therefore not be your motive.

[CFLarsen]

Originally posted by Thanz
Then what was your motive? The answers to the questions you were asking me would mean nothing to you. Actually getting the answers could therefore not be your motive.

I answered this already. I pointed out that you did not know what you were talking about. If you consider this "harrassment", well....:rolleyes:

I answered your question, please answer mine:

Have you reported me to the moderators for harrassing you?

[Thanz]

Originally posted by CFLarsen


I answered this already. I pointed out that you did not know what you were talking about.
You did not know anything about statistics yourself. Wasn't BillHoyt already posting in a similar vein? Doesn't he actually know something about statistics? Why did you not leave this for the "experts" as well? Why is it that you hide when asked to actually think for yourself?

I answered your question, please answer mine:

Have you reported me to the moderators for harrassing you?
No, and I'm not going to run and tell mommy either. :rolleyes:

[CFLarsen]

Originally posted by Thanz
You did not know anything about statistics yourself.

Not correct. I am not an expert on statistics, that's all.

Originally posted by Thanz
Wasn't BillHoyt already posting in a similar vein? Doesn't he actually know something about statistics? Why did you not leave this for the "experts" as well? Why is it that you hide when asked to actually think for yourself?

I don't know why BH does things. You really have to ask him about that. However, I did not detect anything wrong with what he was saying.

Which does not mean that he - or I - could be wrong. Which I have made clear before, too.

Originally posted by Thanz
No, and I'm not going to run and tell mommy either. :rolleyes:

Fine. Then stop complaining about being harrassed.

[Thanz]

Originally posted by CFLarsen

Not correct. I am not an expert on statistics, that's all.
Oh, so you DO know something about statistics? What do you know? Have you taken any courses? Did you understand the questions you were asking me? Would you have understood the answers to those questions? Can you do any statistical analysis yourself? Can you make any informed decisions about what statistical tools to use for a given situation?

I don't know why BH does things. You really have to ask him about that. However, I did not detect anything wrong with what he was saying.

Which does not mean that he - or I - could be wrong. Which I have made clear before, too.
But you were simply asking me questions that BillHoyt was already asking. Why bother repeating them? If you don't care about the answers (as you would not have understood them) why bother asking?

Fine. Then stop complaining about being harrassed.
No. If I think you are harrasing me, I'll say so.

[CFLarsen]

Thanz,

Do you have a problem regarding getting attention? Or do you have a Claus-fetish? You seem to want to provoke discussions with me, no matter what, even if no subject exists.

I've said what I want to say here. I haven't made any claims here (other that I am not an expert on statistics - how do I prove that? Easy...!).

Find another playmate to pick fights with. I prefer discussions of some intellectual merit. You merely bore me.

[Thanz]

Originally posted by CFLarsen
Do you have a problem regarding getting attention? Or do you have a Claus-fetish? You seem to want to provoke discussions with me, no matter what, even if no subject exists.
No, it is not about you, it is about discussing claims and the paranormal and research design. Things that you claim to be interested in. But refuse to actually answer questions about. If you won't share your knowledge, why are you posting here?

I've said what I want to say here. I haven't made any claims here (other that I am not an expert on statistics - how do I prove that? Easy...!).
No, you have claimed some positive level of knowledge of statistics. I am asking you to back that up with some details, so I know in the future what questions I can ask for your opinion on and what weight to give to that opinion. So, are you going to answer my questions about your claim that you know something about statistics?

Find another playmate to pick fights with. I prefer discussions of some intellectual merit. You merely bore me.
I tried to engage you in a conversation with intellectual merit. I asked for your contribution on test design and for a proper method of counting the JE guesses. You, despite what you say above, were not interested. Instead, you want to make long lists of questions to harass others with. You also need time to follow Clancie around the board so that you can disagree or pick apart anything she says.

Claus, I know that you feel that I am somehow not being fair with you. But all I did was ask for your input on the substantive content of the thread. It was something that you seemed perfectly willing to do when it consisted of buggin me about stats. But when asked to actually contribute something worthwhile, you clam up. You haven't even told me why my questions require "expertise" in the field of statistics.

You claim to be a skeptic. But you seem to be focussed on things other than an actual inquiry into the paranormal. If you can't "embarrass" some "believer", it seems that you lose interest rather quickly. I question your true motives.

[Lurker]

Bill:

I apologize. I seem to have been talking over your head. I will no longer mention integrals to you. It seems you fail to understand them.

Lurker

[BillHoyt]

Originally posted by Lurker
Bill:

I apologize. I seem to have been talking over your head. I will no longer mention integrals to you. It seems you fail to understand them.

Lurker

http://www.baltobluegrass.com/bbggraph/monkey.gif
Laughingstock.

[Thanz]

Originally posted by BillHoyt


The definition of a Poisson distribution is based on that of a Poisson process. A Poisson process is one that satisfies the following:


1. The changes (or events) that result from the process can be grouped into nonoverlapping intervals.

2. The numbers of changes (or events) in the nonoverlapping intervals are independent from one another.

3. This independence holds for all intervals.

4. The probability of exactly one change (or event) in a sufficiently small interval, h = 1/n equals n*p , where p is the probability of one change (or event) and n is the number of trials.

5. The probability of more than one change (or event) in a sufficiently small interval, h, is essentially 0.

The Poisson distribution results when such a process occurs over n trials.

There are no hard and fast definitions to it, other than the above.


Mr Hoyt -

I am still fuzzy as to why Poisson is appropriate for the analysis we are doing here. Could you perhaps specifically apply your general description of Poisson (quoted above) to the specific analysis we are doing here? What about the JE guesses makes you think it is a "Poisson Process"?

Much appreciated.

[Lurker]

Originally posted by BillHoyt

Laughingstock.

Hmm, when one looks around the room and notices that nobody else is laughing, a clueless person would continue laughing.

Lurker

[Lurker]

Show me the math, Bill. It is a rather simple request, is it not?

Lurker

[BillHoyt]

Originally posted by Lurker


Hmm, when one looks around the room and notices that nobody else is laughing, a clueless person would continue laughing.

Lurker

If you don't hear the derisive laughter, Lurker, I suggest Q-tips.

[T'ai Chi]

Around 20 pages after statistics were being talked about, and still no appropriate statistical analysis has been done of the JE counts.

Think some of the JREF board participants focus a little too much on rhetoric?

Nah!:k:

[BillHoyt]

Originally posted by T'ai Chi
Around 20 pages after statistics were being talked about, and still no appropriate statistical analysis has been done of the JE counts.

Think some of the JREF board participants focus a little too much on rhetoric?

Nah!:k:

This from one who insists on exhaustive induction? Righhhhht.

[Lurker]

Bill,

Yes, I hear you laughing. Well done! Very good response. Now, when you finish laughing at my ineptitude, could you kindly show me your calculations on the problem I proposed?

Thanks!

Lurker

[Thanz]

Mr. Hoyt -

Will you be answering my question regarding the applicability of Poisson to the present analysis?

[BillHoyt]

Originally posted by Thanz
Mr. Hoyt -

Will you be answering my question regarding the applicability of Poisson to the present analysis?

I tire of answering the same question over and over again. Please have somebody explain my answer to you.

[Lurker]

Bill:

For me, all you have to do is a simple math exercise. As per your limitations, I promise there will be no integrals.

Lurker

[Thanz]

Originally posted by BillHoyt


I tire of answering the same question over and over again. Please have somebody explain my answer to you.
You know, I would ask someone else, but you are the only one who is advocating the use of Poisson. It was you who used Poisson for your analysis. I am trying to understand why.

I do not think that you have actually answered this question once, let alone "over and over again". I know that you don't think Poisson is "Binomial lite", and I know that you don't think that there is enough data for a chi square.

What I don't know is why JE guesses would be a Poisson process. Since it is your claim that Poisson is appropriate in this analysis, I must ask you to back up that claim and explain why.

I am sincerely trying to understand why you have used this particular analysis.

[BillHoyt]

Originally posted by Lurker
Bill:

For me, all you have to do is a simple math exercise. As per your limitations, I promise there will be no integrals.

Lurker

I told you the error. Poisson is not parameterized by N. The error is yours. Have somebody else explain this to you. Take a course. Whatever you need to get up to speed.

[Lurker]

Bill, take the same problem you did in your "J" analysis. Change the p to 1.0. Apply that to the census data. Your expected value for the 85 is 85 right? Does Poisson predict this? Show the CDF.

Try again for p=0.9. Again for p=0.8.

Remember, same problem. All we have changed is the p from the census.

Do the math, Bill. Why are you afraid to do the math?

Lurker

[BillHoyt]

Originally posted by Lurker
Bill, take the same problem you did in your "J" analysis. Change the p to 1.0. Apply that to the census data. Your expected value for the 85 is 85 right? Does Poisson predict this? Show the CDF.

Try again for p=0.9. Again for p=0.8.

Remember, same problem. All we have changed is the p from the census.

Do the math, Bill. Why are you afraid to do the math?

Lurker

I told you the error. Poisson is not parameterized by N. The error is yours. Have somebody else explain this to you. Take a course. Whatever you need to get up to speed.

[neofight]

Originally posted by CFLarsen
Thanz,

Do you have a problem regarding getting attention? Or do you have a Claus-fetish? You seem to want to provoke discussions with me, no matter what, even if no subject exists.

I've said what I want to say here. I haven't made any claims here (other that I am not an expert on statistics - how do I prove that? Easy...!).

Find another playmate to pick fights with. I prefer discussions of some intellectual merit. You merely bore me.

:roll: OMG, this never-ending pot calling the kettle black nonsense is too much to take at times! rofl Oh the irony of it all! Claus, sometimes you are sooooo funny. Guffaw! Especially when you are not at all aware of it. :roll: .....neo

[Lurker]

Bil, Bil, Bill:

Still afraid to do the math? Show me the math. You continue to avoid the question. Do the math. p=1.0, 0.9, 0.8.

Do the exact same "J" analysis and use p as above and apply to your expected mean and your census data (which is how you got the p in the first place).

Do science a favor and do the math and quit ducking. You are giving us science people a bad name here.

Lurker

[BillHoyt]

Originally posted by Lurker
Bil, Bil, Bill:

Still afraid to do the math? Show me the math. You continue to avoid the question. Do the math. p=1.0, 0.9, 0.8.

Do the exact same "J" analysis and use p as above and apply to your expected mean and your census data (which is how you got the p in the first place).

Do science a favor and do the math and quit ducking. You are giving us science people a bad name here.

Lurker

:bs:

[Lurker]

Originally posted by BillHoyt


:bs:

I am trying to teach you something here, Bill. Why don't you just do the math. Or explain how the problem is ANY different than what you did.

Thanks,

Lurker

[Lurker]

There are two sides here. One side wants to repeat the same problem with a different value of p. The other side refuses to try the exercise.

You tell me which you think represents the skeptical philosophy better?

In my opinion, Bill, you are starting to look a lot like Gary Schwartz.

Lurker

[Thanz]

Originally posted by Lurker
Bill, take the same problem you did in your "J" analysis. Change the p to 1.0. Apply that to the census data. Your expected value for the 85 is 85 right? Does Poisson predict this? Show the CDF.

Try again for p=0.9. Again for p=0.8.

Remember, same problem. All we have changed is the p from the census.

Do the math, Bill. Why are you afraid to do the math?

Lurker
I'll do some math!

I can calculate the expected number for our sample of 85, but I don't know what number you want me to use for the observed number of events. Are we using 17 still?

[Lurker]

Originally posted by Thanz

I'll do some math!

I can calculate the expected number for our sample of 85, but I don't know what number you want me to use for the observed number of events. Are we using 17 still?

Thanz:

I must admit I am a bit flummoxed by Bill. I have tried being nice. I have tried being snide. I have tried being damn annoying. For someone who seems to pride himself on being a skeptic he sure dodges simple questions.

As to the problem, the 17 count is immaterial. For p=0.9, and using the 85 sample size, what is the probability of getting 85 or less?

Simple. Why won't Bill show the math?

Lurker

[Thanz]

Originally posted by Lurker
Thanz:

I must admit I am a bit flummoxed by Bill. I have tried being nice. I have tried being snide. I have tried being damn annoying. For someone who seems to pride himself on being a skeptic he sure dodges simple questions.

As to the problem, the 17 count is immaterial. For p=0.9, and using the 85 sample size, what is the probability of getting 85 or less?

Simple. Why won't Bill show the math?

Lurker
Well, I have done the math. I'm prepared to post it if you are interested or if you think it will move the discussion along.

[T'ai Chi]

Say JE uses J more than expected. BUT, what if JE also uses some of the other high frequency letters less than expected?

Could we say anything meaningful here?

[BillHoyt]

Originally posted by T'ai Chi
Say JE uses J more than expected. BUT, what if JE also uses some of the other high frequency letters less than expected?

Could we say anything meaningful here?

I thought you were a statistician. An astounding question. How about you spin some hypotheses. You might even answer one of the naive questions that has haunted some of the woos on this thread.

[Thanz]

Originally posted by BillHoyt
There are three choices as I see them: drop such guesses entirely and thereby lose most of the data, pick one letter and run with it (but which one?) or make a uniform rule allowing him as much leeway as you can. I said I tried to honor his process as much as possible, and I decided to grant him the leeway. The specific leeway I did not grant was "Rich or Richard", where one is clearly a nickname for the other. In those cases, I counted one guess. If he said "Rich or Richard or R" I counted two guesses.

I am quoting this again because I do not think that you have provided any logical basis for your counting method. You have adequately explained what the counting method was, but not the logical reason behind it.

Specifically, WHY is it appropriate to count "Rich or Richard or R" as two separate guesses of "R"? He is clearly guessing for ONE person, not two, and hoping that he gets a hit close to richard. If they say "Randy" he will still accept it as a hit. But either way, he is just casting ONE R net here for ONE person.

The comparison we are doing is to the census data, which obviously counts people. Should we not also be counting people in the guesses, rather than just names? Isn't it obvious in this example that JE is referring to only one person?

Your count is inaccurate and inappropriate to be used in a comparison to the census data. Unless you think the census counts people more than once.

[BillHoyt]

Originally posted by Thanz
The comparison we are doing is to the census data, which obviously counts people. Should we not also be counting people in the guesses, rather than just names? Isn't it obvious in this example that JE is referring to only one person?

What is obvious is that JE calls for a person with a spice name and accepts a dog's name as an answer. What is obvious is that JE sometimes calls out the same initial and states there are two people by that name. What is also obvious is that his multiple guesses are difficult to tally in a regularized way. I chose a rule that would cover the most cases: count all names cited unless one is clearly a nickname of another. This rule covers JE in cases where he guesses "Ellen, Helen, or some 'L' name" and it covers "Richard or some 'R'" name. You may disagree with the method, but I have explained it. Again.

[Clancie]

So, Bill, if JE said to you, "I'm getting someone with a 'B' name, like Bob, Bill, Ben" you would count that as four guesses for 'B'.

And that counting method makes sense to you. :eek:

(Also, as I've said before, "Ginger" shouldn't count at all, btw, since JE got the name symbolically, not phonetically).

Oh well. Just another example of a "skeptic" who is unwilling to admit when he's wrong. :rolleyes:

[Thanz]

Originally posted by BillHoyt


What is obvious is that JE calls for a person with a spice name and accepts a dog's name as an answer.
So what? We shouldn't count nicknames anyway as they are not part of the census.
What is obvious is that JE sometimes calls out the same initial and states there are two people by that name.
And in those cases we count two guesses. Not hard.
What is also obvious is that his multiple guesses are difficult to tally in a regularized way.
Difficult, maybe. But not impossible. Re-read my method and tell me where the flaws are, if you can.
I chose a rule that would cover the most cases: count all names cited unless one is clearly a nickname of another. This rule covers JE in cases where he guesses "Ellen, Helen, or some 'L' name" and it covers "Richard or some 'R'" name. You may disagree with the method, but I have explained it. Again.
The only reason you picked it is because it is easy? Even though you know that it is incorrect?

Let's look at it from another angle. What are we trying to determine? We are trying to determine whether or not JE is cold reading. We have theorized that JE will guess higher frequency letters more often as they increase the chance of a hit. What we are concerned with is how much JE's methods seem directed to increasing his chance of a hit. The way that he does this is by casting as wide a net as possible.

One way to cast a wide net is to guess high frequency letters. Another is to guess multiple names. When we look at the data, we see that he rarely guesses multiple names that are dissimilar from one another. He does not say "Bob or John or Peter". he will say "Peter or some P name". In this instance, saying "Peter" does nothing to increase his chances of a hit (a hit of Peter will already be covered in "P"), it only makes him seem better if the P name IS Peter. If we are simply looking at frequency of letters, it makes no sense to count "Peter" and "P" as two separate and distinct guesses of the letter "P". They are not. It is only one "P" guess for one person.

Frequently this occurs when JE will guess a name like Peter, get nothing, and expand his guess to any P. It is still one P guess, not two, as he is still guessing for one person.

As for the Ellen/Helen, or J or G name, I counted each of those as 2, as he is attempting to widen the guess net beyond just one letter.

My method is also easy to apply, and it yields, IMO, a more accurate count for our purposes. Can you make any argument, beyond "Ease" for your method? Specifically, can you make any argument that it is more accurate?

[BillHoyt]

Thanz,

First post out of the box in reply to me and already you distort my post? Where did I say "ease" or "easy"? Would you care to try again or do you just wish to push an agenda?

[CFLarsen]

Originally posted by Clancie
So, Bill, if JE said to you, "I'm getting someone with a 'B' name, like Bob, Bill, Ben" you would count that as four guesses for 'B'.

And that counting method makes sense to you. :eek:

Do you have a better solution?

Originally posted by Clancie
(Also, as I've said before, "Ginger" shouldn't count at all, btw, since JE got the name symbolically, not phonetically).

Why not? Ginger is a name, such as Mary and Deborah. How do you know when JE gets a name symbolically? What's the symbol for "Eula"??

Originally posted by Clancie
Oh well. Just another example of a "skeptic" who is unwilling to admit when he's wrong. :rolleyes:

Perhaps you could spare us your personal dislikes and instead try focusing on the issues.

[Thanz]

Originally posted by BillHoyt
Thanz,

First post out of the box in reply to me and already you distort my post? Where did I say "ease" or "easy"? Would you care to try again or do you just wish to push an agenda?
Well, if the reason is not ease of applicability, then what is it? That is what I got out of your post. If it is