View Active Topics          Latest 100 Topics          View Your Posts          Switch to Mobile

Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Discussions about Psychics and Psychic Phenomena, Extra Sensory Perception, Telepathy, Psi, Clairvoyancy, 6th Sense, Psychokinesis, etc.

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby ProfWag » 07 Sep 2011, 23:07

Craig Browning wrote::, using the type of stats and 'scientific' data that keeps getting insisted upon, that really can prove things "wrong" when it comes to the traditional skeptic's position.

Craig B, I agree with you in that I am not a statistician and many of these numbers are difficult to understand which is why I was asking for a reference. If someone is going to suggest that psi is a reality, that is news to me and I would like to know the source of that information and to what extent the evidence appeared to confirm. I don't need numbers, per se', as I'll leave that to those in the field who do that sort of thing, I simply would like to know the source, whether it is someone's opinion from on online forum like the JREF, if it was written in a book, or if it was written in a peer reviewed journal. I don't think asking for that information is too much to ask. In fact, without knowing the source of data and/or information, I don't believe one should form an opinion based just on what someone else tells them, yet it happens all the time from both skeptics and "believers" alike.
So, feel free to criticize me about demanding a reference, but shame on you for not wanting to see it yourself if this subject interests you.
User avatar
ProfWag
 
Posts: 3843
Joined: 05 Aug 2009, 03:54






Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby craig weiler » 07 Sep 2011, 23:28

Hi Craig,
Actually, Ice Age is by far better with the statistics and deserves the shout out. I have a good grasp of the theory and I know what a good p value looks like, but I can't do the calculations.

As you can see, Arouet and Profwag have done an admirable job of informing themselves on the subject, and I mean that sincerely. About 95% percent of the skeptics I encounter are complete idiots spewing nothing but talking points while having not bothered to inform themselves at all. These two might be overly conservative and stubborn with their viewpoints, but they are confronting the evidence head on and not running away from it. I have respect for them.

They do have something in common with nearly every other conservative person I have ever encountered; they don't do a good job of stepping back and seeing things as a whole. They are far more comfortable examining "things", sometimes in too much detail. I think that at some point this inhibits good decision making because it ignores the concept of "enough."

A skeptic would criticize me for how I look at the Ganzfeld and how easily I come to the conclusion that it is legitimate. To their mind, you have to examine every last detail and if anything is not in place, it throws the whole thing into question. It is a very left brain approach, and while it has its strengths, it also has weaknesses.

They fail to see it from a right brain approach; that is, to see it as a whole. Things such as:
1. It has been picked over by skeptics for 30 years and modified accordingly
2. A lot of researchers have been involved at different times and different places.
3. The results stay basically the same regardless of these differences.
4. It agrees with other research in this field.
5. It agrees with field observations.
6. It behaves like entanglement.

It is not any one of these things that I have mentioned, but all of them together along with the actual ganzfeld studies that is convincing. To grasp that, you need to be comfortable with a right brain approach.
A ship in harbor is safe, but that's not what ships are for.
User avatar
craig weiler
 
Posts: 386
Joined: 03 Sep 2011, 12:08
Location: San Francisco Peninsula

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby Arouet » 08 Sep 2011, 02:21

_Ice_Ages_14_Aces_ wrote:
Arouet wrote:You've forgotten the biases in calcualting the statistics.


Such as????


I didn't have as much time as I'd hoped over the weekend and haven't gone back through the ganzfeld threads yet. I'm not a stats guy either though so I'm not the best to describe it (and stats that I've picked up have come from reading forums). but I understand there to be some real dispute.

I also understand that some feel that the dispute has been resolved. From what I can see, however, there is no consensus, which makes it difficult for me as a lay person to feel confident in the suggested interpretation.

Actually, if you look at the history of the Ganzfeld, you will clearly see its methodology has profoundly improved in a chronological order and Monica Harris, along with Robert Rosenthal (A world-renowned expert in methodology and meta-analysis) critiqued the 85 MA and they both concluded that something interesting was going on.


I agree the methodologies have improved. I'm not the best person to suggest further improvements though.

In 1975, the Parapsychological Association officers adopted a policy opposing the file-drawer problem, so both positive and negative findings have been reported at the Parapsychological Association's anuual mettings and in its affiliated publications for over nearly 4 decades. Hey, nice thinking!


That's better than nothing. What I'm suggesting is that every study is announced BEFORE it is completed and registered with the PA. All results must be reported as well. Just trusting parapsychologists to tell the PA after the fact is not good enough. Again: given how simple such a system would be to put in place (and I know I'm not the only one to suggest this): what's the hold-up?

Also: IIRC Ersby found some funky stuff in his graph that suggested that there was a file drawer problem. I can't recall if that was ever explained away, but I know it was a big issue that was discussd at the time.

Yes, they can. However (as I said before) experimenter bias, as well as bias in statistical analyses, is not a plausible explanation. The Ganzfeld studies since 1983 were done under double-blind, sensory-isolated, and automated (computer-controlled) conditions, so the possibility of experimenter bias, the lack of negative studies, and flaws have been ruled out. In my opinion, the proper way to use meta-analyses is to combined the results of independent, homogeneous studies.

Are the studies homogeneous and done independently? Check.


I know the ersby version was homogeneous, but were the others? I thought the main reason he did his review was specifically because they weren't!

Are the ganzfeld meta-analyses combined with the standard criteria (Images and videos) instead of nonstandard (ex. musical targets, etc.)? Check.


I just read a hyman article recently where he wrote IIRC that with the new studies the positive results were with the video, not the still. But my time line was off. Actually, I had gotten about 3/4 through the paper doing a summary but then closed down without saving and lost it. I might go back and try it again. So don't quote me on this. But I think Hyman had some issues there that seemed pertinent.

1974-2004 meta-analysis: 3,145 independent ganzfeld sessions, 1008 hits (32%) z=9.11113, p<0.00000000000001 Evidence: Amazing!

Andrew Endersby: (6,700) 1922.9 hits (28.7%) z=6.9547, p=0.00000000001 Evidence: Amazing!

Ersby (JREF): (2854) 782 hits (27.4%) z=2.93955, P=0.00328 Evidence: Good

All combined: (12,699) 3,712 hits (29%) z=10.9998, p<0.0000000000001) Evidence: Amazing!

So, is a total of 12,699 independent ganzfeld sessions not good enough for you??????


Again, I'm not a stats guy. But Ersby's is the tightest, right? and his hit rate is down to 27.4% just 2.4% over chance. Doesn't that bother you at all? Now, perhaps Esrby's methodology can also be challenged, but that's my point: there is still a ton of debate on the methodologies used which result in pretty different resutls (32% vs 27%!)!

Also: how are you adding all those together? Isn't there some overlap in the different meta-analyses? Ersby's wasn't new data, he was paring down the old data, getting rid of outliers and heterogeneous studies.


One thing is clear to me: none of this is clear! I think we're not doing any favours by trying to pass it off as a done deal.
User avatar
Arouet
 
Posts: 2544
Joined: 07 Aug 2010, 03:07

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby craig weiler » 08 Sep 2011, 03:46

Ersby isn't the tightest. Removing outliers just drags down the numbers without giving better data. Think of it this way: The bottom is closed because the study will not go much below chance levels, while the top is open; the numbers can go up to 100%. If you take away a high outlier and then take away a purely chance level, you will ALWAYS skew the results downwards.

In reality, this is just a clever way of statistical cherry picking.
A ship in harbor is safe, but that's not what ships are for.
User avatar
craig weiler
 
Posts: 386
Joined: 03 Sep 2011, 12:08
Location: San Francisco Peninsula

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby Arouet » 08 Sep 2011, 04:48

craig weiler wrote:Ersby isn't the tightest. Removing outliers just drags down the numbers without giving better data. Think of it this way: The bottom is closed because the study will not go much below chance levels, while the top is open; the numbers can go up to 100%. If you take away a high outlier and then take away a purely chance level, you will ALWAYS skew the results downwards.

In reality, this is just a clever way of statistical cherry picking.


Here's where I'm out of my league. I mean, I could cut and paste the detailed explanations that were in that thread, along with the explanations for why it appeared to indicate that there was a file drawer problem, but there were some apparently good reasons to do so.

This is part of the problem: these things are so dependant on statistics and statistical manipulation that its hard to have confidence in the results. And this is where bias can easily sneak in. Again: its a problem of small effect sizes. If the effect size was bigger, it would come out in the wash - the result would be staggering. It also goes to my point that NO ONE should be saying case closed on Ganzfeld right now. It is simply too muddy. It might be pithy to go down a list and snappily say "check!" but from what I've seen its not nearly so cut and dried.

I've said it above: they really need to start the replications from scratch, eliminate the file drawer and reach some consensus involving stat experts on the stat methods to be used. Meta-studies are problematic to begin with as I understand it. While their at it: forget about doing it with just any subject: focus on the artists, musicians, twins, etc. people who they seem to think have a better gift than others and may have more of a chance of a bigger hit size. They need some consistent, highly confident results. They need a clear win.

The Ganzfeld, which is known as the best evidence for psi, is extremely muddy. Which makes it not the best evidence for psi. If parapsychologists want to have any chance of breaking this into the mainstream they need to clean it up. Which means starting from scratch with the best protocols, (perhaps even improving on them). And start counting from 1.
User avatar
Arouet
 
Posts: 2544
Joined: 07 Aug 2010, 03:07

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby _Ice_Ages_14_Aces_ » 08 Sep 2011, 09:52

ProfWag wrote:You consider more than half of the experiments a failure a successful experiment? If 100 people flip 100 coins and 45 of them are heads 55% of the time and the success is reported as "45 people flip heads 55% of the time--statistical significance!", then this is actually incorrect since 55 people could have reported heads 50% or less. If the second half is not reported, the the bias lies in incomplete data as overall, the significance is actually a coincidence or not a statistically significant event.


You're absolutely correct! If 100 people independently flipped the coins let's say 1000 times, 45% got a 55% hit rate (550 heads in 1000 flips) and only the 45% of the people (55% hit rate) were reported that would be a total of 100,000 flips and 24,750 heads (z=-159.6918, p<0.00000.....1). While a result like this is highly statistically significant, it clearly suggests that there is an extremely strong evidence that many tails than heads had occurred in a way that would hardly hardly happen by chance alone (a.k.a. psi-missing). However, if you combine the total heads from 45 (24,750) and 55 (27,500 heads 50% hit rate) that would be a total of 52,550 heads (z=16.12445, p<0.000...1) which suggests that many heads than tails had occurred in a way that would hardly hardly happen by chance alone (a.k.a. psi-hitting).

A P-Value tells you the probability of an observed outcome would take a value as extreme as or more extreme in which the null-hypothesis true and it also measures how strong evidence you have against the null-hypothesis. While the P-Value tells you if you have evidence or not, there is a draw-back :? . The P-Value doesn't tell you what other factors besides chance have caused the observed outcome, so other factors besides chance must be taken under consideration before concluding that you have genuine evidence against the null-hypothesis.

If the coins were truly fair and there were no factors that influenced the overall observed outcome, then you have genuine evidence of psychokinesis.
User avatar
_Ice_Ages_14_Aces_
 
Posts: 69
Joined: 04 Sep 2011, 06:38

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby _Ice_Ages_14_Aces_ » 08 Sep 2011, 10:27

craig weiler wrote:Hi Craig,
Actually, Ice Age is by far better with the statistics and deserves the shout out. I have a good grasp of the theory and I know what a good p value looks like, but I can't do the calculations.


Thanks for the compliment, Craig :mrgreen:!

P.S. while I do sound like a knowledgeable person in statistics here, I am actually not an expert in statistics. I know some of the statistical concepts such as experimentation, data, scatterplots, significance tests, etc. but I'm no expert in all fields of statistics........
User avatar
_Ice_Ages_14_Aces_
 
Posts: 69
Joined: 04 Sep 2011, 06:38

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby _Ice_Ages_14_Aces_ » 08 Sep 2011, 10:48

Aroulet,

Thanks for informing me more about Ersby's meta-studies. I knew his meta-studies were flawed since he used one-tailed and trimmed out the outliers...

Even though if I get rid of Ersby's meta-studies and combine only Andrew's and the 1974-2004 MA, it would still be highly statistically significant with an overall 29% hit rate. (I guess JREF members who conducted their own meta-studies aren't realiable either.)

As for the 32% vs 27% hit rate you had mentioned why would that bother me???? I don't how that could possibly be a problem. In fact, there are a huge number of cases that show that all forms of human performance vary widely from one moment to the next, even among experts. Can an expert in basketball constantly get a hit rate let's say 80% of the time??? No, but that doesn't mean that expert can never reach an 80% hit rate. Even if the expert did got an 80% hit rate, it's very possible that 80% hit rate would vary in the other long run. Besides, aren't P-Values less than 0.000001 not good enough for you???

TTYL (Talk to you later)......
User avatar
_Ice_Ages_14_Aces_
 
Posts: 69
Joined: 04 Sep 2011, 06:38

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby Arouet » 08 Sep 2011, 12:30

The issue is that the tighter studies had lower hit rates.

Anyhow, I'm slowly making my way through the Ersby thread on JREF. They've been talking about the outliers - someone said that even Radin had said outliers should be removed. Jessica Utts said they shouldn't be removed just because they were outliers, but Ersby said he removed them to make the studies homogeneous.

Still making my way through the thread, will see the conclusions...
User avatar
Arouet
 
Posts: 2544
Joined: 07 Aug 2010, 03:07

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby Arouet » 08 Sep 2011, 12:36

I'll post some interesting posts from that thread. Here's one from fls on the outlier issue:


The problem is that your criteria for removing studies may introduce an additional bias. It is okay to remove outliers if they are clearly measuring something different than what the rest of the studies are measuring (Dalton is a good example). But it is not okay to remove them if they simply fall on the extremes. While outliers should be evenly distributed (making it just as likely for you to remove an usually low result as an unusually high result), since the ganzfeld distribution is uneven, I suspect that the removal of outliers leads to the preferential removal of high results. Is that right?

Instead, think of it as trying to remove sources of heterogeneity. One possibility is to add to your exclusion criteria. Remove studies which did not include a sender, consider only studies where participants were drawn from a more general population rather than subgroups (like artists or siblings), stick to a narrow methodology, etc. If you can identify a homogeneous group of studies, then it would be reasonable to perform a meta-analysis on that group.

Despite the technical ability to do so, this data simply should not be combined as is. At best, it can be used to generate hypotheses for testing. At worst, it confirms that the use of ganzfeld to measure psi should be dropped as fruitless.

Linda
User avatar
Arouet
 
Posts: 2544
Joined: 07 Aug 2010, 03:07

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby craig weiler » 08 Sep 2011, 22:33

This is moving into denier territory, rather than skeptical. As countless people have pointed out, skeptics have to be skeptical of skeptics. This business of treating the words of skeptics as gospel and giving them no scrutiny whatsoever is ahem, unfortunate. Automatically claiming that Ersby's meta analysis is somehow better is ridiculous. You don't know that.

What is it about the positive outliers that would cause them to be removed?

In fact, at least one is a study that used artistically gifted students and the study quality is ranked rather high. (But as we've seen, study quality is not a factor in the results, so this is a non issue.)

It was a standard replication study, so why remove it? Arouet, you never ask this question, you follow the skeptic narrative like it was the Word of God.

You're being very selective on what you'll question and what you won't based solely on your views.
A ship in harbor is safe, but that's not what ships are for.
User avatar
craig weiler
 
Posts: 386
Joined: 03 Sep 2011, 12:08
Location: San Francisco Peninsula

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby Arouet » 08 Sep 2011, 22:52

With all due respect, it may be you who is not looking at this objectively. I'm not sure you really carefully read the post I quoted.

Linda wasn't saying anything like you are suggesting. In fact, she was saying not to remove them just because they were outliers. But rather because those studies were not homogeneous. She said that if you are going to remove just because they are outliers that would introduce bias. She also made some suggestions for how to make the meta-study even more homogeneous which would allow one to treat all the trials as one big experiment. When the studies are heterogeneous you just can't do that.

If you've read the ganzfeld thread you will see that it is a thread filled with debate, with differing views. it is a myth that skeptics never disagree with each other. Proponents may like to think so, but I don't think it holds up to scrutiny.

Read the post again, and maybe you'll see why I thought it was interesting to post.
User avatar
Arouet
 
Posts: 2544
Joined: 07 Aug 2010, 03:07

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby craig weiler » 08 Sep 2011, 23:15

Yes, I did read it. It is skeptic nonsense that demonstrates a complete lack of understanding about psi or human potential. It is an attempt to skew results towards null to fit the skeptic narrative. Duh.

In particular, the removal of artistic individuals or siblings is completely uncalled for.

If we are going to examine whether people can shoot a basketball from a foul line and get the ball through the hoop by anything but chance, you don't remove the people who are most likely to succeed. What kind of blank up logic is that?

The objective is to determine whether the task can be done at all, not whether it can be done by a homogenous group of people. Is this really so hard to understand?

As I said, this is veering into denier territory.

(I'm off to work and won't reply for a few hours.)
A ship in harbor is safe, but that's not what ships are for.
User avatar
craig weiler
 
Posts: 386
Joined: 03 Sep 2011, 12:08
Location: San Francisco Peninsula

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby Arouet » 09 Sep 2011, 00:27

I still don't think you're reading it properly.

She wasn't saying don't look at artists she was saying that to combine studies in a meta-study the criteria has to be homogeneous for the results to be considered in any way reliable. Otherwise its simply inappropriate to do a meta-study. The have to be identical methodologies. Otherwise you should't be trying to combine results.

I'm no stats expert but that certainly sounds reasonable to me!
User avatar
Arouet
 
Posts: 2544
Joined: 07 Aug 2010, 03:07

Re: Straight Talk: The Ganzfeld (Advocates & Skeptics only)

Postby craig weiler » 09 Sep 2011, 04:05

ARRRGGGHHHH!!!!!!

No, it's complete nonsense. If we are testing for the EXISTENCE of home runs, then ALL properly tested data about hitting home runs is valid unless they are tinkering with the study methods, such as blindfolding people. If one study happens to use major league baseball players there is no good reason to exclude them just because they hit more home runs because the study itself remained the same. Is that any more clear to you?

Homogenizing the results does not produce more reliable data, just more homogenous data that is not necessarily more reliable.
A ship in harbor is safe, but that's not what ships are for.
User avatar
craig weiler
 
Posts: 386
Joined: 03 Sep 2011, 12:08
Location: San Francisco Peninsula

PreviousNext

Return to Psychic Phenomena / ESP / Telepathy

Who is online

Users browsing this forum: No registered users and 1 guest

cron