A Statistical Case for global consciousness

[Copernicus]

A search didnt come up with anything, apologies if its already been discussed. I'd love a better mathematician than me to explain the statistical tests in English!

An interesting project run out of Princeton that statistically tests various hypotheses that world events will measurably perturb what would otherwise be random numbers.

The methodology takes the output of random number generators at sites around the world. The data is bursts of 200 bits of data which, if random, would have a mean sum of 100. The data is then analyzed during various periods surrounding global events and varous statistical tests used to measure correlations in deviations from the mean results around the world.

This links to the page on 9/11 and the remarkable results surrounding that day's events.

The project does not yet make any claims as to existence of a global consciousness, or explanations for their results. However the results are statistically significant and say "theres something happening here, what it is aint exactly clear".

[Borodog]

It's crap. Data abuse. If you torture the data enough, it will confess to anything.

[bunny]

I'm afraid I cant give you the intelligent comentary from a statistician you are looking for (stats was a minor part of my degree). However, I would have to say that quotes like

"Dean observes that "such large changes will eventually occur by chance, of course, but this particular change happened during an unprecedented event, suggesting that this `spike' and `rebound' were not coincidental.""

do not fill me with confidence. I would be interested to see their data in "quiet" periods and see whether the same fluctuations occur from time to time. I looked over their website a little when this was posted a couple of months ago - there is such a lot of data to wade through though.

It does seem like they are also attempting to form a hypothesis and decide on testing techniques ahead of time - this would have much more meaning to me.

[Copernicus]

[ QUOTE ]
I'm afraid I cant give you the intelligent comentary from a statistician you are looking for (stats was a minor part of my degree). However, I would have to say that quotes like

"Dean observes that "such large changes will eventually occur by chance, of course, but this particular change happened during an unprecedented event, suggesting that this `spike' and `rebound' were not coincidental.""

do not fill me with confidence. I would be interested to see their data in "quiet" periods and see whether the same fluctuations occur from time to time. I looked over their website a little when this was posted a couple of months ago - there is such a lot of data to wade through though.

It does seem like they are also attempting to form a hypothesis and decide on testing techniques ahead of time - this would have much more meaning to me.

[/ QUOTE ]

From what ive gathered so far they do test quiet periods, but im not convinced by any means.

[Sharkey]

Arnold Toynbee once remarked before a lecture audience that an original thought anywhere in the world was instantly transmitted to all four corners via some sort of mental aether.

[bunny]

[ QUOTE ]
[ QUOTE ]
I'm afraid I cant give you the intelligent comentary from a statistician you are looking for (stats was a minor part of my degree). However, I would have to say that quotes like

"Dean observes that "such large changes will eventually occur by chance, of course, but this particular change happened during an unprecedented event, suggesting that this `spike' and `rebound' were not coincidental.""

do not fill me with confidence. I would be interested to see their data in "quiet" periods and see whether the same fluctuations occur from time to time. I looked over their website a little when this was posted a couple of months ago - there is such a lot of data to wade through though.

It does seem like they are also attempting to form a hypothesis and decide on testing techniques ahead of time - this would have much more meaning to me.

[/ QUOTE ]

From what ive gathered so far they do test quiet periods, but im not convinced by any means.

[/ QUOTE ]
I think the only way I would place any weight on their research is if they successfully made predictions in advance (perhaps focussed on the world cup or events of such importance - I expect (shocking as it may be to americans) that this would have more impact on most people's lives than the 9/11 tragedy). The limited reading I did on their site seemed to show much weaker results here (and from memory they modified what tests they applied to the data after the fact - another red flag in my mind).

[ffredd]

You can always find something that seems non-random in random data. When you do, you can formulate a hypothesis based on what you've found. You can't however use the same data to check if your hypothesis is correct! That would just be ridicilous.

Suppose e.g. that I examine my hand histories and find that I got dealt 1% more pocket pairs than the statistical expectation. Now I have a hypothesis: "This site deals 1% too many pocket pairs". I can test that hypothesis by checking another large database of hand histories, but I can't do it by examining the same hand histories.

What these guys should have done is to use the data collected from a subset of the random number generators to formulate a hypothesis, and the data collected from the remaining random number generators to test the hypothesis. Their failure to do so invalidates their "evidence" completely.

It's also interesting that the largest contribution to the "unlikeliness" of the data came from the time before planes started crashing into buildnings. If they had chosen their start and end times a little bit differently, their result would have been very different.

[bearly]

and the "emanations and pneumbras" from the constitution?.........b

[Copernicus]

[ QUOTE ]


It's also interesting that the largest contribution to the "unlikeliness" of the data came from the time before planes started crashing into buildnings. If they had chosen their start and end times a little bit differently, their result would have been very different.

[/ QUOTE ]

I believe they do take different starting times on the same data and test the consistency of results using the different start times.

Bunny, as I recall World Cup is one of the events they hypothesize in advance about and then test, though I dont recall the results.

It is also interesting that (at least as far as Ive seen) there are no disclaimers that Princeton University disavows any connection with the group, and I dont see any statements from PU to that effect. Since they are using PU servers I would think they would distance themselves from any blatantly bad science.

[SWB]

A few statisticians and other researchers reviewed their work a few years back, I don't recall whether it was because they submitted findings to a scientific journal or if Skeptical Inquirer went out to take a look at them. At any rate, there were two major problems:

First, they do a lot of data-fitting, and not in a particularly consistent manner. Basically, when some event occurs, they go and see if there was some kind of spike around that time. Anywhere from a day before to a day after the event is counted as a hit. Obviously, that gives them a lot of leeway to play with.

Second, there are many spikes that are not associated with any particular event. There are spikes much larger than the 9/11 examples which don't correspond to anything in particular. The researchers hypothesize that these spikes represent events important to some smaller group of people and unknown to the masses, but then why would they have a bigger reaction in the "global conciousness" than 9/11?

So basically, it's junk science right now. Maybe there's something there, maybe they'll figure out enough to put it in publishable form and get their Nobel prizes... or maybe they won't. Right now there's nothing to get excited about.