Re: Final Question for Sylvia Else (Inability to fullfill contract)

Article View: sci.skeptic

Article #5292

#5292

From: |-|ercules
Date: Sun, 19 Sep 2010 05:56

69 lines
4670 bytes

"Richard Cornford" <Richard@litotes.demon.co.uk> wrote ...
> |-|ercules wrote:
>> Richard Cornford wrote
>>>
>>> It is still far too small a test set for this protocol, but if the 
>>> effect was significant you would expect to see 5 out of 5 being a
>>> common enough result to have already shown up. If the effect is no
>>> more than "better than random" then it would probably take at least
>>> a few hundred samples to decide the issue.
>>
>>
>> I suggest a 3 question set in future.
> 
> The appropriate test is determined by what it is that you are trying to 
> test for. If it is only a better than random correlation that you are 
> trying to demonstrate then a better test would be, for each question 
> (where the specific question doesn't matter much), mix the 'channelled' 
> answer with, say, 9 random quotes form the same book (so that all the 
> options have the same prose style and set of possible characters/subject 
> matter) and see if people can pick out the 'channelled' answer. Of 
> course both the 'channelled' responses and the random quotes would have 
> to follow the same word selection rules (say, from the word at the point 
> selected to the next full stop), and it would be best if someone else 
> could create the lists of random quotes.


We tried 5 options (without substitution) and the markers were too thick.

I don't think they had the concentration span to consider 25 possible matches
and pick the best one for each question.

3 options is where the distinctions in the phrases will be obvious.





> 
>> It will detect the subtle answers more readily.
>>
>> 0/3 -> 2 possibilities -> bottom 33.3%
>> 1/3 -> 3 possibilities -> top 66.6%  (2 chances out of 3 to score 1 or
>> more right)
>> 3/3 -> 1 possibility -> top 16.6%
> 
> Your figures don't add up again.

And yet you offer no solution.  Maybe you are wrong yet again.




> 
>> I can do 3 sets of 3 questions per day.
>>
>> I have to work out a FAIL benchmark.
> 
> If it is a better than random correlation then results that were no 
> better then random would be a fail.

Yes but a non psychic could float just above 50% for 100 tests without
any significant result.

So a slight benchmark above 50% is in order.

3/4 Vs the expected 2/3 (to score 1 or more from 3 options)


Herc

Technical Details

Message-ID: <8fln0pFahtU1@mid.individual.net>

Path: arc1-ca.pugleaf.net!g2news1.google.com!news4.google.com!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail

References:

<8fdahaFa4mU1@mid.individual.net> <8fdaraFboaU1@mid.individual.net> <8fdmijF2j6U1@mid.individual.net> <8fdrn9Ftr8U1@mid.individual.net> <8fdt99F722U1@mid.individual.net> <8fe0gdFpt5U1@mid.individual.net> <8fe6meFvqpU1@mid.individual.net> <8febmjFu75U1@mid.individual.net> <8fecanF2enU1@mid.individual.net> <ooj596t86sm70tnvd36h6l16q57745f345@4ax.com> <8fg3r4F909U1@mid.individual.net> <p786965ojgbq67g14lqqkn5d370vogin60@4ax.com> <8fgtijFm15U1@mid.individual.net> <272896tjcpl4d9grofheek5c5ge4r08vh5@4ax.com> <8fil0tFp9dU1@mid.individual.net> <z_-dnWvgLcKqoQnRnZ2dnUVZ7oadnZ2d@giganews.com> <8fiu22F3ldU1@mid.individual.net> <I5Kdnb99b6C4xwnRnZ2dnUVZ7s-dnZ2d@giganews.com> <8fj4pbF428U1@mid.individual.net> <24-dnV0oQqpVYQnRnZ2dnUVZ8nydnZ2d@giganews.com> <8fl4hqFj94U1@mid.individual.net> <-86dnRUCcNL1CQjRnZ2dnUVZ7sWdnZ2d@giganews.com> <8flj1aFneuU1@mid.individual.net> <kdqdnbHigoKpPwjRnZ2dnUVZ7sGdnZ2d@giganews.com>

Show Full Headers

Complete Article Headers

X-Google-Thread: f44bc,c689aad361b93075
X-Google-NewGroupId: yes
X-Google-Thread: 102c27,c689aad361b93075
X-Google-NewGroupId: yes
X-Google-Thread: 1038e3,c689aad361b93075
X-Google-NewGroupId: yes
X-Google-Thread: 10a774,c689aad361b93075
X-Google-NewGroupId: yes
X-Google-Thread: 107787,250192c4cd10ee66
X-Google-NewGroupId: yes
X-Google-Attributes: giddc1377d6e,gid35c5869fd6,gidb0034e805d,gid3ac807a62f,gid2f40725cd2,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Path: arc1-ca.pugleaf.net!g2news1.google.com!news4.google.com!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: "|-|ercules" <radgray123@yahoo.com>
Newsgroups: sci.skeptic,aus.legal,misc.legal,aus.tv,alt.paranormal
Subject: Re: Final Question for Sylvia Else (Inability to fullfill contract)
Date: Sun, 19 Sep 2010 15:56:35 +1000
Lines: 69
Message-ID: <8fln0pFahtU1@mid.individual.net>
References: <8fdahaFa4mU1@mid.individual.net> <8fdaraFboaU1@mid.individual.net> <8fdmijF2j6U1@mid.individual.net> <8fdrn9Ftr8U1@mid.individual.net> <8fdt99F722U1@mid.individual.net> <8fe0gdFpt5U1@mid.individual.net> <8fe6meFvqpU1@mid.individual.net> <8febmjFu75U1@mid.individual.net> <8fecanF2enU1@mid.individual.net> <ooj596t86sm70tnvd36h6l16q57745f345@4ax.com> <8fg3r4F909U1@mid.individual.net> <p786965ojgbq67g14lqqkn5d370vogin60@4ax.com> <8fgtijFm15U1@mid.individual.net> <272896tjcpl4d9grofheek5c5ge4r08vh5@4ax.com> <8fil0tFp9dU1@mid.individual.net> <z_-dnWvgLcKqoQnRnZ2dnUVZ7oadnZ2d@giganews.com> <8fiu22F3ldU1@mid.individual.net> <I5Kdnb99b6C4xwnRnZ2dnUVZ7s-dnZ2d@giganews.com> <8fj4pbF428U1@mid.individual.net> <24-dnV0oQqpVYQnRnZ2dnUVZ8nydnZ2d@giganews.com> <8fl4hqFj94U1@mid.individual.net> <-86dnRUCcNL1CQjRnZ2dnUVZ7sWdnZ2d@giganews.com> <8flj1aFneuU1@mid.individual.net> <kdqdnbHigoKpPwjRnZ2dnUVZ7sGdnZ2d@giganews.com>
Mime-Version: 1.0
Content-Type: text/plain;
 format=flowed;
 charset="iso-8859-1";
 reply-type=response
Content-Transfer-Encoding: 7bit
X-Trace: individual.net kkmJOAuhjo64YwY0GHwhaAmAAseufBZ/LR3ExOk8HHL1+uWOvk
Cancel-Lock: sha1:TkVgKLVN81LfdHzq54rmd6p/6cY=
In-Reply-To: <kdqdnbHigoKpPwjRnZ2dnUVZ7sGdnZ2d@giganews.com>
X-Priority: 3
X-MSMail-Priority: Normal
Importance: Normal
X-Newsreader: Microsoft Windows Live Mail 14.0.8089.726
X-MimeOLE: Produced By Microsoft MimeOLE V14.0.8089.726

Reply to Parent Message