JREF discusses 182. Andrew Paquette Brings Statistical Rigor to Psi Experiences

[Andrew Paquette]

Even though you call it "averse" and claim that it is overly high, you don't back this up with anything but probability calculations which are invalid. If instead we look at how many amazing and ho-hum readings are produced by non-anomalous information under other circumstances (mediumship readings), we find that the number you produced was almost exactly the same. So there doesn't seem to be any room for anomalous information to have produced any extra amazing dreams.

Linda

The section of your post that I snipped was very unclear, so please pardon the deletion. The next bit doesn't make sense to me unless you have arbitrarily discounted the compounding factor I used when calculating probability and instead substituted your own completely different method, ala' Wiseman.

The key to this paper is that I am checking related items, not items that are mutually independent. The question isn't, "how many individual line items are veridical?" But how many per night are veridical AND linked to the same anchor AND day?" There is a big difference. You treat them all as if independent and you lose the compounding power of this. It is the difference between predicted one die roll at a time, and predicting twenty in a row. In the first example, each roll has a probability of 1:6, but in the second, it is 1/1*6*6*6*6...

AP

 

[fls]

The people over at JREF stopped discussing the paper seriously, though there were a number of fair questions at first. Now it's just a bunch of wandering trolls out there. It was too good to last, but for just a moment I thought they might have a couple of real skeptics. To badly paraphrase Jesus, "Why waste your time?"

AP

That's pretty typical for the JREF forum and is a big part of why I asked myself the same question (though I had no idea I was paraphrasing Jesus :)). There are some good people over there, but I never did figure out how to keep discussions from being overwhelmed by trolls (not really the right description, but it will do unless I come up with something more a propos).

Linda

 

[fls]

The section of your post that I snipped was very unclear, so please pardon the deletion. The next bit doesn't make sense to me unless you have arbitrarily discounted the compounding factor I used when calculating probability and instead substituted your own completely different method, ala' Wiseman.

What does that mean?

The key to this paper is that I am checking related items, not items that are mutually independent. The question isn't, "how many individual line items are veridical?" But how many per night are veridical AND linked to the same anchor AND day?" There is a big difference. You treat them all as if independent and you lose the compounding power of this. It is the difference between predicted one die roll at a time, and predicting twenty in a row. In the first example, each roll has a probability of 1:6, but in the second, it is 1/1*6*6*6*6...

AP

It is incorrect to compound the results in this way. It's easy to tell that this is incorrect. If you apply the same method to the mediumship studies, your probability will also tell you that it is extremely unlikely for someone to find a reading veridical due to "chance". Yet if you ask people if a reading, when psi is absent (i.e. they aren't the recipient of the reading), is veridical, they say yes far more often then your probability says they would. If there is a big discrepancy between what your model estimates, and the actual measured outcome, then your model is wrong.

Linda