It is currently Tue Jul 17, 2018 6:45 pm




Post new topic This topic is locked, you cannot edit posts or make further replies.  [ 28 posts ]  Go to page Previous  1, 2
 Cognitive Illusion 
Author Message
Chibi-Czar
User avatar

Joined: Wed Jun 09, 2004 9:14 pm
Posts: 2769
Location: Location, Location!
Post 
<Edit> I thought I'd make a more detailed critique of Twyst's post instead of just a simple "no." </Edit>

Your methodology is flawed. You're not looking at the right thing. You include a lot of erroneous data. Allow me to critique your test.

Twyst wrote:
I tested 5 people 100 times.

Did you test five people a hundred times each for a total of five hundred tries or did you test five people for a total of one hundred tries?

Twyst wrote:
68% of the time people will pick either the center or the right side box (or left if their left handed)

Which box they take does not matter. This data set is irrelevant.

Twyst wrote:
50% of the time they picked the other box after removing one false box
25% of those times they picked the correct box

This is irrelevant. You're only testing how likely the box they do choose is correct.

Twyst wrote:
50% of the time the correct box was picked the first time

So the experiment shows that you have a 50% chance no matter which box you chose at first. A third box is irrelevant and is the true illusion, making you think you have a 1 in 3 chance. However, if you decide to switch boxes, your chances of picking the correct box is reduced to 25%.

First, your math is flawed. If half the time the subjects switch boxes and when they switch boxes 25% of the time it's the correct box, that does not mean that the other box is right only 25% of the time. In order to figure that out, you'd need to determine how often the first box was right. If the first box is right 50% of the time, when they kept the first box it would also have to be right only 25% of the time. Your math simply doesn't work out.

Second, your numbers are too neat. Real probability doesn't come out like that. Flip a coin one hundred times. You will almost never get fifty heads and fifty tails. Your total will be close to the optimum fifty, but rarely bang on. This is a good allegory as you claim each box has a fifty percent chance of being right. Even if both boxes had a fifty-fifty chance of being right, the chance it would show up as exactly 50% is 8.8817841970012523233890533447266e-16-to-1 against. In case you're wondering, that means no way in hell.

Third, the numbers you have given are very unlikely statistically, but they are very psychologically compelling. They appeal to human tastes. 50% is exactly half the total possible. 25% is exactly one quarter. That you claim to have tested five people a hundred times is less compelling in this way as this is one part of the experiment that you admittedly made up and thus would naturally be prone to the vagaries of human though patterns. None the less, when taken in context with the other numbers, they are suspect.

Finally, I am forced to concluded you made this data up. You did no real tests and simply fabricated the supposed results. Your math that you use o draw conclusions is incorrect. The probabilities you concluded did not equate with the data you posted. The numbers you listed show clear signs of fabrication. This would explain why they don't mesh with your conclusions. Simply put, you tried to fix both the data set and the conclusions but did not think to make them agree. The probability of the data you suggested showing up, even if there was a probability of fifty-fifty, is astronomically small. When viewed as a whole the only conclusion I can reach is that you are a lying liar who lies, and lies poorly at that.

Having concluded that you are a liar, nothing you further say in this matter can be admitted. Everything you say from now on will be suspect as, having been lied to by you once, there can be no trust that you are not lying now. One can only assume that if you were to post corrected data from another supposed test, that there would be a strong possibility that you merely used this critique to generate a more compelling lie the second time around. Simply put, STFU & GTFO.

_________________
Image


"Understanding the scope of the problem is the first step to true panic."
--Freefall

A file that big?
It might be very useful.
But now it is gone.
-- David J. Liszewski


Fri May 04, 2007 1:55 am
Profile
Chibi-Czar
User avatar

Joined: Wed Jun 02, 2004 5:01 pm
Posts: 287
Location: Co. Springs, CO, USA
Post 
Quote:
Did you test five people a hundred times each for a total of five hundred tries or did you test five people for a total of one hundred tries?


Each person was tested 100 times.

Quote:
Originally posted by Twyst:
68% of the time people will pick either the center or the right side box (or left if their left handed)


Which box they take does not matter. This data set is irrelevant.


I just thought it was interesting.

Quote:
First, your math is flawed. If half the time the subjects switch boxes and when they switch boxes 25% of the time it's the correct box, that does not mean that the other box is right only 25% of the time. In order to figure that out, you'd need to determine how often the first box was right. If the first box is right 50% of the time, when they kept the first box it would also have to be right only 25% of the time. Your math simply doesn't work out.


and ect... to keep from continuous quotes...

The truth is the answer is rounded out. I told the participants to chose the other box for the 2nd half of the test. As in, to stick with their first choice for 50 tries, then to switch for the other 50. As for the clean cut 50%...that's the average for all test participants. During the first half, it was nearly 50% for everyone. Doing the math to average it out made it 50.2%

During the second half, the results were a bit more screwed up... One person managed an exact 40%, while another got (checks notes) only a 10%.

Quote:
Even if both boxes had a fifty-fifty chance of being right, the chance it would show up as exactly 50% is 8.8817841970012523233890533447266e-16-to-1 against.


That only applies if you account for human error. If each box was assigned purely randomly, and the box picked randomly, after 500 tries it would be so close to 50% that just about anyone would round it to 50.

Quote:
Having concluded that you are a liar, nothing you further say in this matter can be admitted. Everything you say from now on will be suspect as, having been lied to by you once, there can be no trust that you are not lying now. One can only assume that if you were to post corrected data from another supposed test, that there would be a strong possibility that you merely used this critique to generate a more compelling lie the second time around. Simply put, STFU & GTFO.


Whatever man...

_________________
Sic gorgiamus allos subjectatos nunc - We gladly feast on those who would subdue us


Fri May 04, 2007 12:01 pm
Profile
Moderator of Pain!
User avatar

Joined: Mon Mar 15, 2004 2:52 pm
Posts: 314
Location: Somewhere where I don't know where I am
Post 
Ok, what started out as an interesting debate has ONCE AGAIN devolved into a childish arguement. Closing before it gets worse.

_________________
They say 'Give me a long enough lever and I will move the heavens and the earth'
I say 'Give me a long enough crowbar and I'll break into Heaven'


Fri May 04, 2007 1:16 pm
Profile
Display posts from previous:  Sort by  
Post new topic This topic is locked, you cannot edit posts or make further replies.  [ 28 posts ]  Go to page Previous  1, 2


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © phpBB Group.
Designed by Vjacheslav Trushkin for Free Forums/DivisionCore.