(2018-10-15, 11:33 PM)malf Wrote: [ -> ]From the GCP themselves
Quote:The procedures for event selection cannot be strictly formulaic because we are attempting something without precedent, addressing complex issues in a very big world. In brief: Event selection was necessarily flexible at the inception of the Project because we did not have precedents to tell us what kind of events would be useful.
That doesn't at all contradict what Chris wrote. The selection of events might be somewhat flexible, but - according to the same project website - events are selected
before analysing the data, so the hypotheses are, as Chris wrote, fixed in advance. From the project's more detailed
page on event selection (emphasis added by me):
Quote:About half the events in the formal series are identifiable before the fact; the accidents, disasters, and other surprises must, of course, be identified after they occur. To eliminate a frequent misconception, we do not look for spikes in the data and then try to find what caused them. Such a procedure, given the unconstrained degrees of freedom, is obviously inappropriate. There is no data mining, and there is no post hoc inclusion or exclusion of events. All events are entered into the formal experiment registry before the corresponding data are extracted from the archive.
(2018-10-16, 07:45 AM)Laird Wrote: [ -> ]That doesn't at all contradict what Chris wrote. The selection of events might be somewhat flexible, but - according to the same project website - events are selected before analysing the data, so the hypotheses are, as Chris wrote, fixed in advance. From the project's more detailed page on event selection (emphasis added by me):
But it seems generally agreed that there is data mining, hence Bancel's appeal to "researcher psi" in their selection of events and time frames.
We have a project that:
1. Has no testable hypothesis,
2. Has scant prior plausibility,
3. Has enough flexibility in the process to allow conscious or unconscious bias.*
There are enough conditions present here to remain justifiably skeptical IMO. For anyone unaware of how science can let us down if we trust it too much see here:
http://www.skeptiko-forum.com/threads/cr...iced.2959/
*I don't think we would need that much either; it's not like the data, as it is presented, is "slam dunk" in that regard, albeit mathematically significant.
(2018-10-16, 10:43 PM)malf Wrote: [ -> ]But it seems generally agreed that there is data mining
Hmm? How could you write that in response to a post in which I quoted the GCP website as stating (which I bolded) "There is no data mining"? Nor is it agreed to by Chris. So... who's in this general agreement?
(2018-10-16, 10:43 PM)malf Wrote: [ -> ]hence Bancel's appeal to "researcher psi" in their selection of events and time frames.
Even if psi guided the selection of events (and I'm not sure, because I haven't read his paper, that you're correct that this is what he's claiming), this would still not be data mining: again, because the data are not extracted and analysed until
after the events are selected.
(2018-10-16, 10:43 PM)malf Wrote: [ -> ]We have a project that:
1. Has no testable hypothesis,
Hmm? What do you call this then? (Quoting from the same link as in my previous post; emphasis in the original)
Quote:The first step was to define our experimental hypothesis, but this could not be strictly formulaic since our target system, global consciousness, can only be usefully defined in operational terms. Our approach has been to work with a general hypothesis describing a range of conditions rather than a narrow set of parameters:
Periods of collective attention or emotion in widely distributed populations will correlate with deviations from expectation in a global network of physical random number generators.
(2018-10-16, 10:43 PM)malf Wrote: [ -> ]2. Has scant prior plausibility,
That's a matter of opinion, no? Given prior work done with RNGs in, for example, the PEAR lab, and/or a worldview which does not conceive of consciousness as utter slave to matter, an opinion opposite to the one you expressed seems to have plenty of justification.
(2018-10-16, 10:43 PM)malf Wrote: [ -> ]3. Has enough flexibility in the process to allow conscious or unconscious bias.
If it does, you haven't pointed out where or how.
(2018-10-16, 10:43 PM)malf Wrote: [ -> ]*I don't think we would need that much either; it's not like the data, as it is presented, is "slam dunk" in that regard, albeit mathematically significant.
"Albeit mathematically significant" is kind of grudging respect for results
over seven standard deviations from the norm, don't you think? Seems pretty Michael Jordan-like to me...
(2018-10-17, 04:22 PM)Steve001 Wrote: [ -> ]Keep in mind this
Sure, there seem to be different ways to interpret the data; different ways in which the results might have come about. The two ways that
don't seem available (assuming honest reporting) are "blink luck" (unless you're willing to accept seven standard deviations worth of luck) and "data mining".
(2018-10-18, 06:29 AM)Laird Wrote: [ -> ]Sure, there seem to be different ways to interpret the data; different ways in which the results might have come about. The two ways that don't seem available (assuming honest reporting) are "blink luck" (unless you're willing to accept seven standard deviations worth of luck) and "data mining".
You do know whom Mr. Nelson was?
A little something on p-value. This is just one examination on the limitations of the p-value.
Quote:Three Things the P-Value Can't Tell You about Your Hypothesis Test
"A p-value is the probability of seeing something as extreme as was observed, if the model were true."
http://blog.minitab.com/blog/understanding-statistics/three-things-the-p-value-cant-tell-you-about-your-hypothesis-test
You see even Mr. Nelson doesn't know what the rng machines are respondinging to, yet some of you are quick to shout eureka.
Yes, but it has no relevance to my response: assuming honest reporting, the null hypothesis can be decisively rejected. You do accept that?
Edit: I see you edited your post. To be clear: the above was in response to your "You do know whom Mr. Nelson was?" In response to your additions: I understand all of that already. Nobody's saying the effect size is huge (it isn't). But whatever the limitations to p-values are, a value this small justifies decisively rejecting the null hypothesis - on the assumption that the procedure and results have been reported accurately. If you don't accept that, then it would be helpful if you explained why.
(2018-10-18, 12:01 PM)Laird Wrote: [ -> ]Yes, but it has no relevance to my response: assuming honest reporting, the null hypothesis can be decisively rejected. You do accept that?
Honesty doesn't necessarily equate with truth with much left to interpretation in the on going GCP. So no.
(2018-10-18, 01:48 PM)Steve001 Wrote: [ -> ]Honesty doesn't necessarily equate with truth with much left to interpretation in the on going GCP. So no.
Well, I think the null hypothesis can reasonably be rejected with respect to the GCP. The problem you are describing, I think, is that there are a plethora of alternative hypotheses left to choose from, most of which are not anomalous.
Linda
(2018-10-18, 12:01 PM)Laird Wrote: [ -> ]Yes, but it has no relevance to my response: assuming honest reporting, the null hypothesis can be decisively rejected. You do accept that?
Edit: I see you edited your post. To be clear: the above was in response to your "You do know whom Mr. Nelson was?" In response to your additions: I understand all of that already. Nobody's saying the effect size is huge (it isn't). But whatever the limitations to p-values are, a value this small justifies decisively rejecting the null hypothesis - on the assumption that the procedure and results have been reported accurately. If you don't accept that, then it would be helpful if you explained why.
The GCP is a giant widget detector attempting to demonstrate human collective will can influence classical phenomena. The problem arises because there's no evidence will can influence classical reality. Now you might balk at that and point out The Copenhagen Interpretation of QM theory. But that is one interpretation, there are other interpretations of equal validity. This giant widget detector is an extension of the PEAR lab. During the entire history not one independent lab was able to replicate PEARS's claimed positive results. Is that answer satisfying?