Page 1 of 2
Multiple lines of evidence and increasing probability
Posted: Thu Sep 15, 2022 9:43 pm
by drumdude
“DCP’s favorite book on NDEs” wrote: By scientifically studying the more than 1,300 cases shared with NDERF [by the time the book was written, more than a decade ago], I believe that the nine lines of evidence presented in this book all converge on one central point: There is life after death.
The convergence of several lines of evidence — like the nine presented in this book — builds a much stronger case than only a single line of evidence.
For example, suppose we had only two lines of NDE evidence. We may not be 100 percent convinced that these two lines of evidence prove an afterlife, but perhaps each line of evidence by itself is 90 percent convincing. Combined, these two lines of evidence by mathematical calculation are 99 percent convincing that the afterlife exists. (4; italics in the original)
If each of two lines of evidence from near-death experiences (NDEs) is 90 percent convincing of the existence of an afterlife, then the combination of these two lines of evidence may be considered as follows: The probability that either of these lines of NDE evidence individually is not convincing of the existence of an afterlife is 10 percent, or 0.1. The probability that the combination of these two lines of NDE evidence is not convincing of the existence of an afterlife is (0.1 x 0.1), or 0.01, which is 1 percent. Thus the combination of two lines of NDE evidence, each of which is 90 percent convincing of the existence of an afterlife, gives 100 percent minus 1 percent, or 99 percent confidence that the afterlife is convincingly felt to exist. (203, note 1; italics in the original)
“DCP” wrote: I’m interested in feedback on the reasoning that is set forth above. Not so much about reader opinions regarding near-death experiences in and of themselves but on the subject of whether the methodology of assessing levels of persuasion is substantially sound. Please advise.
“DCP” wrote: I was nonetheless concerned at IANDS, and have been concerned for quite a while when I’ve seen it elsewhere, to see a mingling of what I regard as pretty solid material with other things that seem to me rather dubious at best — e.g., crystal-gazing, drum circles, mediums, past-life regressions, and various kinds of New Age spirituality that have been cherry-picked and decontextualized from world religious traditions and Third World indigenous practices. (I know from speaking with him that at least one of the officers of IANDS himself shares my concern, that association with a whole host of such fringy and undisciplined things detracts from the credibility of what we both see as credible claims of out-of body experiences [OBEs] and near-death experiences [NDEs]. And they are already dismissed by quite a few people without serious consideration.)
I think Daniel is starting to actually get at the heart of his long-standing epistemological problem. He has yet to finally realize it, but the scientifically dubious company he was surrounded with (crystal healers, etc) at the NDE conference he spoke at seems to have got him really thinking.
Do all of DCP’s deep seated beliefs rely on this faulty (many assumed 90% true lines of evidence multiplied together) thinking?
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 1:33 am
by Res Ipsa
He also skips over the requirement that the lines of evidence must be completely independent before simply multiplying probabilities. That’s not a small detail.
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 1:58 am
by High Spy
Res Ipsa wrote: ↑Fri Sep 16, 2022 1:33 am
He also skips over the requirement that the lines of evidence must be completely independent before simply multiplying probabilities. That’s not a small detail.
Once upon a time in a math forum I requested probability analysis of the Mormon Temple Quake occurring at only one minute in a 100+ year interval. Even math heads lost it, and couldn’t get past the problem proposition. See march8miracle.org for the basic premise. There are many LDS type threads on various forums too.
1 chance in 57 million works for me with slight preference taking into account the magnitude of said earthquake.
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 3:19 am
by Gadianton
But High Spy, the probability that everything that happened to me the way it did today; the exact number of steps on my walk and the exact blades of grassed pushed into the ground with my shoe, the number of barks of my dog, the exact shape of the puddle by my spilled water, and on and on, was a lot less likely than 1 in 57 million. Every hand is just as rare as a royal flush.
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 3:36 am
by Res Ipsa
High Spy wrote: ↑Fri Sep 16, 2022 1:58 am
Res Ipsa wrote: ↑Fri Sep 16, 2022 1:33 am
He also skips over the requirement that the lines of evidence must be completely independent before simply multiplying probabilities. That’s not a small detail.
Once upon a time in a math forum I requested probability analysis of the Mormon Temple Quake occurring at only one minute in a 100+ year interval. Even math heads lost it, and couldn’t get past the problem proposition. See march8miracle.org for the basic premise. There are many LDS type threads on various forums too.
1 chance in 57 million works for me with slight preference taking into account the magnitude of said earthquake.
The problem is that it is a statistical certainty that some very low probability event or events will happen on a regular basis. As an example, if I draw one ball out of a container of one million consecutively numbered balls, the odds that I will draw any specific ball are one in a million, but the odds that I will draw a ball are 1-1. Once I draw the ball, the odds of getting that ball before I drew are meaningless in terms of trying to place any significance on the fact that I drew that number.
At any point in time, we could define a near infinite list of low probability events that could occur. With so many possible events, it is a virtual certainty that one of them will occur at any point in time. Claiming that the event with the Moroni statute is so rare that it can’t be due to chance is just like claiming that drawing ball number 41,227 has some special significance after the ball was drawn. You’re focusing on only the ball without considering the bag from which it was drawn and its contents.
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 9:47 am
by toon
Res Ipsa wrote: ↑Fri Sep 16, 2022 1:33 am
He also skips over the requirement that the lines of evidence must be completely independent before simply multiplying probabilities. That’s not a small detail.
I tested positive for COVID a few weeks past. At precisely the same time, I also had a fever, congestion, was extremely tired, and easily out of breath despite being relatively fit. What were the odds of all that happening at the same time?
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 4:56 pm
by drumdude
toon wrote: ↑Fri Sep 16, 2022 9:47 am
Res Ipsa wrote: ↑Fri Sep 16, 2022 1:33 am
He also skips over the requirement that the lines of evidence must be completely independent before simply multiplying probabilities. That’s not a small detail.
I tested positive for COVID a few weeks past. At precisely the same time, I also had a fever, congestion, was extremely tired, and easily out of breath despite being relatively fit. What were the odds of all that happening at the same time?
By the rules of Mormon logic since you’ve experienced such a miraculous coincidence, you’re now allowed to assert anything.
Re: Multiple lines of evidence and increasing probability
Posted: Fri Sep 16, 2022 6:50 pm
by Marcus
drumdude wrote: ↑Fri Sep 16, 2022 4:56 pm
toon wrote: ↑Fri Sep 16, 2022 9:47 am
I tested positive for COVID a few weeks past. At precisely the same time, I also had a fever, congestion, was extremely tired, and easily out of breath despite being relatively fit. What were the odds of all that happening at the same time?
By the rules of Mormon logic since you’ve experienced such a miraculous coincidence, you’re now allowed to assert anything.
By any chance did you see a dead relative or find a water source? If you did, get your testimony ready. The Mopologists will want to write you into a movie.
Re: Multiple lines of evidence and increasing probability
Posted: Sat Sep 17, 2022 8:26 am
by Physics Guy
Lines of evidence can fail to be independent even if they have had no effect on each other whatever.
I can look at a large set of poker hands, all of which were played by different people, at different tables, having been dealt by different dealers from decks that were shuffled by different machines, with the dealers and shuffling machine suppliers all having strong legal and financial incentives to be fair. I can nonetheless conclude from my large set of quite independently generated poker hands that the odds of getting a straight or better in poker are extremely good, to the point where one is almost compelled to infer a mysterious force that overcomes chance to bring good cards together.
The set of hands I consider? All the winning hands from a tournament. They were independently generated, all right, but my selection method introduced a dependence.
NDEs that sound uplifting and fit the established tropes are much more likely to be reported and collected as evidence than discouraging ones that diverge from what people want to hear.
Re: Multiple lines of evidence and increasing probability
Posted: Sat Sep 17, 2022 9:40 am
by High Spy
Physics Guy wrote: ↑Sat Sep 17, 2022 8:26 am
Lines of evidence can fail to be independent even if they have had no effect on each other whatever.
I can look at a large set of poker hands, all of which were played by different people, at different tables, having been dealt by different dealers from decks that were shuffled by different machines, with the dealers and shuffling machine suppliers all having strong legal and financial incentives to be fair. I can nonetheless conclude from my large set of quite independently generated poker hands that the odds of getting a straight or better in poker are extremely good, to the point where one is almost compelled to infer a mysterious force that overcomes chance to bring good cards together.
The set of hands I consider? All the winning hands from a tournament. They were independently generated, all right, but my selection method introduced a dependence.
NDEs that sound uplifting and fit the established tropes are much more likely to be reported and collected as evidence than discouraging ones that diverge from what people want to hear.
Vonnegut and The Catcher in the Rye are rare exceptions to concocted stories. Kinda like the thousands of poor children that go missing that the media could care less about. But Elizabeth Smart and Jon Bonay receive endless coverage.