Cold Reading Demos at TAM2

CFLarsen said:

Well, duuuuh....that was your claim that you knew, you know....


So you are asking me to explain exactly how someone uses their brain to select something? LOL. Whatever.


In other words, the statistician has the same autistic abilities as Dustin Hoffman's character in the movie "Rainman".


You need to read up more on the concept of 'bias' apparently. Here is an interesting link: http://www.nsf.gov/sbe/nuggets/026/nugget.htm.

Evaluators are like everyone else; they aren't immune from possible bias, but since they are directly involved with the experiment, they should be blinded from who got what treatment.
 
T'ai Chi said:
So you are asking me to explain exactly how someone uses their brain to select something? LOL. Whatever.

Yes, actually. Do you still want to try, or are you just going to repeat the question?

T'ai Chi said:
You need to read up more on the concept of 'bias' apparently. Here is an interesting link: http://www.nsf.gov/sbe/nuggets/026/nugget.htm.

Oh, I understand bias very well. What I also understand is that, for your explanation to work, your statistician need to be autistic. Not have supernatural powers, just autistic.

I get it.

:dl:

T'ai Chi said:
Evaluators are like everyone else; they aren't immune from possible bias, but since they are directly involved with the experiment, they should be blinded from who got what treatment.

Moving the goal posts again. Who said the statisticians are "directly involved" with the experiment? The statisticians get the data after the experiment is done.

Still dancing on hot coal, I see.
 
CFLarsen said:

Yes, actually. Do you still want to try, or are you just going to repeat the question?


Yawn. You need to read up on bias and understand it this time.


What I also understand is that, for your explanation to work, your statistician need to be autistic. Not have supernatural powers, just autistic.


Um, whatever. I don't hold jokes about autism very funny.


Moving the goal posts again.


Nope, despite your beliefs otherwise.


Who said the statisticians are "directly involved" with the experiment? The statisticians get the data after the experiment is done.


Statisticians, as I've mentioned, are often involved in more than 'number crunching'. If you don't know that by now...


Still dancing on hot coal, I see.

Whhoah, out of my area!! Me, I'm just a data monkey. Bill is the one who knows about dancing, ask him.
 
T'ai Chi said:
Yawn. You need to read up on bias and understand it this time.

Oh, I understand it.

T'ai Chi said:
Um, whatever. I don't hold jokes about autism very funny.

It is not a joke, I assure you. The statistician has to be autistic. How else (apart from the supernatural explanation) do you explain his ability to choose unconsciously the biased statistical method?

Is the statistician autistic or not?

T'ai Chi said:
Nope, despite your beliefs otherwise.

It is not my beliefs. Anyone can read your posts.

T'ai Chi said:
Statisticians, as I've mentioned, are often involved in more than 'number crunching'. If you don't know that by now...

But not in your own example. I keep pointing that out to you, but you keep moving the goal posts.

T'ai Chi said:
Whhoah, out of my area!! Me, I'm just a data monkey. Bill is the one who knows about dancing, ask him.

Let's leave it to the readers to decide who is dancing.
 
T'ai Chi said:


You first posted your silly demand on 1/31/04 at 12:58AM. Then you posted it again the same day at 7:45AM. You're an impatient one, aren't ya?

You played the authority card, zulu-time-zone boy. If you consulted for theses, you would have been acknowledged in those theses. Give us the student names, years and institutions.

You invited this by resorting to an authority claim because you were having your a** handed to you. Now be a good little tr'oll and choose: withdraw or give us the evidence to back up your claimed "authority." What's the problem, zulu-time-zone boy?
 
T'ai Chi said:


You and Claus are asking me to explain how someone uses their brain to select something? Um..

Look, if an evaluator knows deep in his mind that one method generally gives smaller p-values than another method, he might unconsciously tend towards that one (or another one) if he desires to unconsciously support his bias as much as he can.

No no no. Let's recap a bit here, since you're clearly getting lost in your own poppycock.

o You keep doing this side-slip of smearing the roles of evaluator and statistician.

o You said the test selection bias can occur before the data have been analyzed. Hence, the paranormal questions which you need to address, sir.

o If you wish to withdraw that paranormal claim, you are now stuck claiming the statistician looks at the data twice. The first time, he doesn't get the result he wants. He then would change tests at least once to get the result he wants. This is no longer bias, but outright scientific fraud.

o "Generally gives smaller p-values" is only true for certain test types over certain data ranges. If you maintain this an a priori situation, you again are making paranormal claims. If a posteriori you again have scientific fraud. More probably, the statistician has selected an inappropriate test, which would be caught in review torn apart by other researchers following publication.

o How much gandy-dancing must you do with your alpha levels to make all this magic happen? Why not just make your bullsh!t transparent and set alpha to .026 to make sure the test falls on the right side of the line? Why not add a few data so long as you're already being "unconsciously" fraudulent? Just to make sure p falls on the correct side of the line?
 
CFLarsen said:

It is not a joke, I assure you. The statistician has to be autistic. How else (apart from the supernatural explanation) do you explain his ability to choose unconsciously the biased statistical method?

Is the statistician autistic or not?


Bias can affect everyone. Autism has nothing to do with it, of course.
 
BillHoyt said:

You played the authority card, zulu-time-zone boy.


Sorry 3 dollar Bill. You should know that extraordinary claims require extraordinary evidence. Me claiming I've done consulting work ain't an extraordinary claim. You should really stick to the statistical issues. Hint: my work history is not a statistical issue.


If you consulted for theses, you would have been acknowledged in those theses. Give us the student names, years and institutions.


See above. To make a blanket statement that all consultants get acknowledged is amusing, as that is definitely not always the case.


You invited this by resorting to an authority claim because you were having your a** handed to you.


You are really being dense here. An argument by authority is when I say Mr. X is right because Mr. X has a Ph.D. There is nothing wrong with arguing from experience though, especially because I've backed it up with mathematics, the same mathematics that you seem to shy away from.


Now be a good little tr'oll and choose: withdraw or give us the evidence to back up your claimed "authority." What's the problem, zulu-time-zone boy?

:nope: Very boring.
 
BillHoyt said:

o You keep doing this side-slip of smearing the roles of evaluator and statistician.


No slipping. An evaluator can be people that are not technically statisticians. Triple blind applies to evaluators in general.


o You said the test selection bias can occur before the data have been analyzed. Hence, the paranormal questions which you need to address, sir.


Nothing paranormal about bias. I really can repeat this all day.


o If you wish to withdraw that paranormal claim,


I won't withdraw it because I never made this claim to begin with. Claus attributed it to me.


o How much gandy-dancing must you do with your alpha levels to make all this magic happen? Why not just make your bullsh!t transparent and set alpha to .026 to make sure the test falls on the right side of the line? Why not add a few data so long as you're already being "unconsciously" fraudulent? Just to make sure p falls on the correct side of the line?

Um. I just messed around with some numbers and I noticed the p-values were closest to .02, hence my examples. Would you like an example where alpha = .05, because I can create one.

You've failed to address why "triple blind" studies have and are being done by researchers. Gee, I wonder why they even consider something so "paranormal"? I think you should contact each and every one of them and tell them they are idiots.

:D
 
T'ai Chi said:
Bias can affect everyone. Autism has nothing to do with it, of course.

But this bias you are describing is by someone who unconsciously is able to choose the right (biased!) method before he even starts analyzing the data.

This is correct, right?

So, either he is fraudulent - is he? - or he is autistic - which you say he is not - or he has supernatural powers.

If you have a fourth explanation, please let us know.
 
T'ai Chi said:
Me claiming I've done consulting work ain't an extraordinary claim. You should really stick to the statistical issues. Hint: my work history is not a statistical issue.

See above. To make a blanket statement that all consultants get acknowledged is amusing, as that is definitely not always the case.
It was you who inserted the claim into the stream, sir. But it is good that you won't use this ploy again. then you are withdrawing your claim, and we won't be treated to another one in the future.
 
T'ai Chi said:
No slipping. An evaluator can be people that are not technically statisticians. Triple blind applies to evaluators in general.
No it doesn't. It applies to evaluators evaluating raw data where bias might creep in. Raw data, sir. Reading MRIs. X-Rays. Making "still sick" or "cured" calls for diseases or conditions without clear-cut numbers to rely on. After these decisions are made by the evaluator, then it goes onto statistical analysis which is now relatively cut & dry. Quit smearing the roles to recover from your errors.

Nothing paranormal about bias. I really can repeat this all day.

I won't withdraw it because I never made this claim to begin with. Claus attributed it to me.
What other conclusion is there when someone claims a statistician can a priori choose a model to force an outcome with unknown data sets? Hmmm? Conversely, if you claim our stat guy looked at the data first, didn't like the results, then ran it through another model to get what he wanted, you are now talking scientific fraud.

I just messed around with some numbers and I noticed the p-values were closest to .02, hence my examples. Would you like an example where alpha = .05, because I can create one.
That's right, T'ai, you can construct examples in the abstract. But the cut-offs are fairly standard in science. .05, .01, sometimes .001. With such a choice restriciton, your claims about biased a priori model choice become even more laughable.
You've failed to address why "triple blind" studies have and are being done by researchers. Gee, I wonder why they even consider something so "paranormal"? I think you should contact each and every one of them and tell them they are idiots.
I have very clearly addressed that issue on this and other threads. You've managed not to understand it. You, not I, also managed to construct the only example I've seen yet that contain this paranormal claim of a statistician being able to pre-select a statistical model that will favor one outcome over another.
 
CFLarsen said:

But this bias you are describing is by someone who unconsciously is able to choose the right (biased!) method before he even starts analyzing the data.


Um the evaluator is able to consciously or unconsciously exercise his bias if he is not blinded.


So, either he is fraudulent - is he? - or he is autistic - which you say he is not - or he has supernatural powers.


Nope. He's average person who is affected by bias like the rest of us.
 
BillHoyt said:

But it is good that you won't use this ploy again. then you are withdrawing your claim, and we won't be treated to another one in the future.

I'm not withdrawing the claim. I'm also noting your red herring, ie:

Me claiming I've done consulting work ain't an extraordinary claim. You should really stick to the statistical issues. Hint: my work history is not a statistical issue. To make a blanket statement that all consultants get acknowledged is amusing, as that is definitely not always the case.
 
BillHoyt said:

No it doesn't.


Yes, it does. Triple blind refers to blinding evaluators in general.


What other conclusion is there when someone claims a statistician can a priori choose a model to force an outcome with unknown data sets? Hmmm?


Um, he has the data with group membership, that is the whole point of blinding and removing the group membership, to help prevent bias.


That's right, T'ai, you can construct examples in the abstract. But the cut-offs are fairly standard in science. .05, .01, sometimes .001. With such a choice restriciton, your claims about biased a priori model choice become even more laughable.


But the main point still stands, obviously. I repeat, would you like me to construct an example where alpha = .05 (or another number of your choosing)? Let me know.


You, not I, also managed to construct the only example I've seen yet that contain this paranormal claim of a statistician being able to pre-select a statistical model that will favor one outcome over another.

Stupid.
 
T'ai Chi,

Let's take this step by step.

You are claiming that the statistician is able to unconsciously choose the right (biased!) method before he even starts analyzing the data.

Is this correct? Just yes or no.
 
T'ai Chi said:
Yes, it does. Triple blind refers to blinding evaluators in general.
Really? So a protocol can jump from single blind straight to triple, do not pass go, do not collect any consulting fees? What an easy-breezy and so transparently sleezy definition!

Um, he has the data with group membership, that is the whole point of blinding and removing the group membership, to help prevent bias.
Really? You said he was blind, you nincompoop. You have just contradicted yourself completely. He can't be blind to group membership and know membership in order to be biased.

Quod Erat Dorkandstrandum

Man, I'm so glad the mods removed the ability to edit posts!
 
BillHoyt said:
Man, I'm so glad the mods removed the ability to edit posts!

Not just you.

Every real skeptic is glad that the mods removed the ability to edit posts after one hour.

You only hear the woowoos complain.

You may guess why.
 
T'ai Chi said:
But the main point still stands, obviously. I repeat, would you like me to construct an example where alpha = .05 (or another number of your choosing)? Let me know.[/B]
None for me thanks. Your first example that you tried to sneak by us people who don't have your claimed degrees was an elementary "error" that puported to show a large (2X) difference between a p derived from Fisher's and a Chi squared.
You failed the test because either you don't know the difference between 2 tails and 1 (F in Stat 101, for that one). Or you didn't check the tables very closely ( C or D in Stat 101). Sometimes people who get a B- in Stat 101 cry out, "But it's just a rounding error!"
So you are either a poseur or a student ( and teaching aide) from the Bad Statistics Department, Arkham University.
You also offer Bad Astronomy and Fraudian Psychology at the university, I understand.
 
Jeff Corey said:

You failed the test because either you don't know the difference between 2 tails and 1 (F in Stat 101, for that one). Or you didn't check the tables very closely ( C or D in Stat 101).


Well since I admitted the second scenario, duh. C or D, yeah... keep saying that to yourself. :) PM me a number if you'd like transcripts for any class and/or proof that I taught that class (and higher classes).


So you are either a poseur or a student ( and teaching aide) from the Bad Statistics Department, Arkham University.


Yawn. Ditto here. You must have cognitive dissonance.

Jeff, you failed to address the actual issue of Fisher's vs. chi-square after I submitted the revised version. Feel free to do. Or, you can simply latch on to an honest mistake which I've already admitted. Choice is yours oh open-minded one.
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom