Do you think the 5-Sigma rule is too harsh for paranormal proof?

[Taverner]

Carl Sagan once said "Extraordinary claims demand extraordinary proof." Building on this, most scientists and researchers demand that for any paranormal claims to be statistically significant, they must have at least 5 standard deviations. This means something like 97%.

Do you think this is too harsh?

[redqueen]

1. Hmmm... is it the same for other proofs? n/t

[Taverner]

2. Each proof is different

For pharmaceuticals, I think they only need two standard deviations.

[EstimatedProphet]

4. Most significance testing (in most fields anyway) is done at the 95% level

This translates to a 1 in 20 chance of making a mistake based on the results of the testing. Being that it is convenient is part of the reason that it is used so commonly. 5 standard deviations is a very significant level, and translates to somewhere around a 1 in 10,000 chance of making a mistake from the test results.

[Taverner]

6. Hence "Extrordinary Claims Demand Extrordinary Proof"

[EstimatedProphet]

9. Yes, and yes I would agree with it

[Taverner]

11. Me too

[Xipe Totec]

14. Depends on the system

Medical devices and aerospace go for six sigma for man-rated systems.

Even then...

[EstimatedProphet]

3. Five standard deviations is approximately 99.99%

[redqueen]

5. based on this and your other response

that does seem somewhat extreme



oh and p.s. thanks for the geek speak guys... LOVE that shit! :evilgrin:

[EstimatedProphet]

7. Hey, I was trying to put it in English!

[redqueen]

8. still geeky enough to be sexy!

:bounce:

[wildhorses]

10. works for me --

bring on the paranormal :bounce:

[Beetwasher]

12. No, Not Harsh At All

And for the record, I don't believe anything has even come close to standard for that matter.

[Taverner]

13. Yup, the best paranormal claims can do is 75% in a fair experiment

And even then, it could easily be explained by a natural cause

[Orrex]

17. Then on what basis can you claim that it's paranormal?

And which paranormal claim has had a success rate of 75%?

[Taverner]

19. The prayer/healing one USED to be 75% - that is until they did a less biased study

The problem with the 75% study was that the participants who were prayed for were specifically told they were prayed for. In the current study, the participants did not know whether they were being prayed for or not.

[Fire Walk With Me]

15. Meditate.

Then explain the various things you find along the way.

http://www.accesstoinsight.org/tipitaka/mn/mn.118.than.html

Most often, it's left to the meditator to discover them on their own, which is why there are not volumes filled with this information. Discovering on your own is the best teacher, when the rules actually work.

[Taverner]

21. But wouldn't this lend the mind to trick itself?

Don't get me wrong - I'm all in favor of meditation. But some of the extraordinary claims made by some practitioners seem like they might be tricks of the mind.

[Orsino]

16. Proof is a strong word.

But yeah, if I wanted funding to test the reproducibility of your results, I'd probably better be armed with 90 or 95% C.I. And a well-documented methodology to lend credence to any extraordinary claims.

[RetroLounge]

18. Our Software works on a 3-sigma

and a 6-sigma level, depending on the calculations needed.

RL

[Taverner]

20. Ahhh but software is different

And you can always patch if worse comes to worse.