Carina Landin in test at this moment

The reason 16 is considered a success is because it was agreed upon.

The reason for picking the limit 16 was based in statistics. To label at least 16 diarys correctly by guesswork corresponds to approximately ½% chanse.

By the way, I find it extremely unlikely that Carina Landin would be able to sense wether a diary was written by a male or female, but not wether she gets a feeling for it or not (if it is too old). It does not matter what I think though. She should be tested according to protocol. Not that it will matter though. My guess is that she will continue to claim powers after she fails next test.
 
Last edited:
Paul2 said:
Why is 16 out of 20 successful and 15 out of 20 isn't? Is it just because that's what was agreed to, more or less arbitrarily, or is there a statistical, mathematical reason?
Click to expand...

The preliminary usually has around a 1/1000 chance of passing by random chance alone. We can calculate the chance that someone gets n correct guesses or more using this formula:
(sum of (i choose 20) from n to 20) divided by 2^20

If you plug 15 into this, you get a ~1/50 chance. 16 gives a ~6/1000 chance. 17 gives a ~1/1000 chance. I don't know why they didn't use 17. So they were actually giving slightly better odds than they say they do!

For the interested: the chance of getting 12 or better is ~1/4.
 
Randi is doing re-runs and giving nice odds. What is happening? What's next? Randi and Geller best friends?
 
I suspect he wants to show that he is not a vindictive old fart, like the woos like to paint him as, but a fair-minded investigator. What would be the objection to a retrial if the applicant thought the protocol was not being adhered to?
 
Zep said:
I suspect he wants to show that he is not a vindictive old fart, like the woos like to paint him as, but a fair-minded investigator. What would be the objection to a retrial if the applicant thought the protocol was not being adhered to?
Click to expand...


I don't think there was anything wrong with the test. It followed the protocol. But maybe I am reading the wrong protocol...

Now I like to see her tested again anyway so it is just fine by me.
 
Assuming a 50/50 probability by chance alone.

The probabiliby of getting 16 or more out of 20 correct is 0.0013
The probability of getting 15 or more out of 20 correct is 0.0059

So it takes 16 or more out of 20 to meet the JREF requirements for the preliminary test.

The probability of getting 12 or more out of 20 correct is 0.2517

If we drop the 5 older diaries that she missed,
the probability of getting 12 or more out of 15 is 0.0176
 
case sensitive said:
I don't think there was anything wrong with the test. It followed the protocol. But maybe I am reading the wrong protocol...

Now I like to see her tested again anyway so it is just fine by me.
Click to expand...

The test did not follow the protocol exactly. In the original protocol, Carina specified an upper limit on the age of the diaries to be used. The testers included several diaries that violated this limit. Therefore, the test results are invalid. Randi is not "being nice." He is doing what is correct and necessary in this situation.
 
Alkatran said:
The preliminary usually has around a 1/1000 chance of passing by random chance alone. We can calculate the chance that someone gets n correct guesses or more using this formula:
(sum of (i choose 20) from n to 20) divided by 2^20

If you plug 15 into this, you get a ~1/50 chance. 16 gives a ~6/1000 chance. 17 gives a ~1/1000 chance. I don't know why they didn't use 17. So they were actually giving slightly better odds than they say they do!

For the interested: the chance of getting 12 or better is ~1/4.
Click to expand...

Strictly speaking this is not correct because certain cases were excluded from the 2^20 possible choices...namely the cases where there were 20 males, the cases where there were 19, 18, 17, 16, 4, 3, 2, 1 or 0 males.

However it's pretty hard to talk about "chance alone" in this case. If by chance you mean Carina was simply flipping coins to make her guesses then yes, your formula is correct because we can think of 20 diaries randomly labeled not M or F but rather "right" or "wrong." There are (your formula) ways of getting 16 or more right and (your formula) ways of getting 12 right, etc.

However if we take into account the fact that Carina knows that there are actually fewer than 2^20 possible outcomes her chances go up.

But the fact is that Carina was not flipping coins, not even mentally. If you give a human being 12 diaries and ask her to label them M and F she will almost always show a great deal of non randomness in her choice.

For example, she will be unlikely to pick 15 male, let alone all 20 male. She will avoid sequences like MMMMMF even though these sorts of sequences are very likely to appear by chance.

Or let's say we gave Carina a bag of 30 tokens, 15 labeled M and 15 labeled F. She picks by grabbing a token from the bag. Afterwards tokens are not returned to the bag. But that is not how the diaries were selected. Every M you pick in that case increases the chance of your next token being an F. I would expect that this was roughly Carina's strategy.

So it's kind of pointless to talk about "chance alone" unless you know Carina's guessing strategy.
 
Gr8wight said:
The test did not follow the protocol exactly. In the original protocol, Carina specified an upper limit on the age of the diaries to be used. The testers included several diaries that violated this limit. Therefore, the test results are invalid. Randi is not "being nice." He is doing what is correct and necessary in this situation.
Click to expand...

Is the original protocol different than the one Randi signed (if that is how he does it)? How many protocols are there and isn't the last protocol the one and only?
 
case sensitive said:
Is the original protocol different than the one Randi signed (if that is how he does it)? How many protocols are there and isn't the last protocol the one and only?
Click to expand...

My apologies. The "agreed upon protocol" I should have said.
 
Gr8wight said:
The test did not follow the protocol exactly. In the original protocol, Carina specified an upper limit on the age of the diaries to be used. The testers included several diaries that violated this limit. Therefore, the test results are invalid. Randi is not "being nice." He is doing what is correct and necessary in this situation.
Click to expand...

From what I can see, that's not exactly right. It looks like the protocol was sloppily-worded, in that it stated the diaries should not be older than late 19th century if possible. And so, it was followed.

That means the testers might have thought if it was not possible to get newer diaries, it was OK to use older ones, and their interpretation of 'late' 19th century might have differed from Carina's.

Is 'late' 19th century anything after 1850? Or anything after 1875? If that's not defined up front, then it's going to lead to this sort of issue. Unfortunate, but I would say the fault was in the protocol, not the test.
 
tkingdoll said:
From what I can see, that's not exactly right. It looks like the protocol was sloppily-worded, in that it stated the diaries should not be older than late 19th century if possible. And so, it was followed.

That means the testers might have thought if it was not possible to get newer diaries, it was OK to use older ones, and their interpretation of 'late' 19th century might have differed from Carina's.

Is 'late' 19th century anything after 1850? Or anything after 1875? If that's not defined up front, then it's going to lead to this sort of issue. Unfortunate, but I would say the fault was in the protocol.
Click to expand...

I agree. And that is also the fault of the sceptical society who negotiated that protocol. They are charged with eliminating escape holes. They did not do so in this case. Telling is the fact that they included one diary from the eighteenth century. It shows that they did not pay as much attention to this condition as they should have. If they were unable to locate enough diaries that were not older than about 125 years, they should have approached Carina and discussed the problem. They absolutely should not have gone ahead with a test with a potential escape hole that large.
 
Agreed. By the sounds of it, this was hardly a scientific test of her abilities and she has every right to complain. If she's for real, they screwed up her chances of showing it. If she's full of it, they gave her a loophole and of course she's taking it.
 
Wow. I'm surprised this protocol was aggreed upon by Randi.

This protocol seems to suggest that there is no non-supernatural way to distingish between male and female entries. Handwriting alone would give a good clue, or so my gut tells me.
 
Rob Lister said:
Wow. I'm surprised this protocol was aggreed upon by Randi.

This protocol seems to suggest that there is no non-supernatural way to distingish between male and female entries. Handwriting alone would give a good clue, or so my gut tells me.
Click to expand...

I believe the protocol specified that Carina would not be allowed to examine the handwriting. As well, the diaries were selected so as to hopefully have no telltale signs of gender identity on the outside.
 
Rob Lister said:
Wow. I'm surprised this protocol was aggreed upon by Randi.

This protocol seems to suggest that there is no non-supernatural way to distingish between male and female entries. Handwriting alone would give a good clue, or so my gut tells me.
Click to expand...

She didn't look at the writings.
 
Gr8wight said:
I believe the protocol specified that Carina would not be allowed to examine the handwriting. As well, the diaries were selected so as to hopefully have no telltale signs of gender identity on the outside.
Click to expand...

You know, I've just been thinking about this for a couple of seconds, and thought that if I were to take this test, given that women's diaries (IMO) would be more likely than men's diaries to have telltale signs of gender on the outside (for example, is a brown leather diary male or female? Is a diary with pink butterflies on the outside male or female?) it seems to me that women's diaries would be more likely to be removed in the sorting process than men's. I would simply guess male for every diary and feel confident of getting a result significantly over 50%. Maybe not good enough to pass, but I like my odds.
 
Gr8wight said:
You know, I've just been thinking about this for a couple of seconds, and thought that if I were to take this test, given that women's diaries (IMO) would be more likely than men's diaries to have telltale signs of gender on the outside (for example, is a brown leather diary male or female? Is a diary with pink butterflies on the outside male or female?) it seems to me that women's diaries would be more likely to be removed in the sorting process than men's. I would simply guess male for every diary and feel confident of getting a result significantly over 50%. Maybe not good enough to pass, but I like my odds.
Click to expand...

But you could only have 15/20. Not good odds at all.
 
Gr8wight said:
You know, I've just been thinking about this for a couple of seconds, and thought that if I were to take this test, given that women's diaries (IMO) would be more likely than men's diaries to have telltale signs of gender on the outside (for example, is a brown leather diary male or female? Is a diary with pink butterflies on the outside male or female?) it seems to me that women's diaries would be more likely to be removed in the sorting process than men's. I would simply guess male for every diary and feel confident of getting a result significantly over 50%. Maybe not good enough to pass, but I like my odds.
Click to expand...

If they followed the protocol that was posted they started with 15 generic looking male and 15 generic looking female diaries, and then flipped a coin to decide which pile to select from. I'm not sure whether 19th century diaries ever had pink butterflies on them, and I wouldn't be surprised if the archives had signifigantly more male diaries to pick from in any case. So long as the collection is large enough to get 15 of each it doesn't matter how many diaries are left on the shelves.

It is possible that they didn't have 15 generic female diaries (which weren't very fragile, etc.) from "the late 19th century" or later to work with. That seems unlikely because they choose diaries in the first place because of the archives.
 
You must log in or register to reply here.

Back
Top Bottom