Homeopathy - I am ready for open challenge

[JWideman]

Ah, I understand now. You can only take 2 measurements each day, at 8am. It seems to me that a simple alternative is to just give you 10 charts, and you identify which ones indicate medicine. Or even just the raw data, which you can chart yourself.

 

[PingOfPong]

Sure I will discuss my results in that science forum.

Actually I thought better of it and realized that I shouldn't distract you from your negotiations. I'm still interested in how you produced your spectral graphs but it's not that important. Good luck with your protocol.

 

[steenkh]

As far as the challenge is concerned, the principles are irrelevant: see Rule 3, and the comment immediately following Rule 16.

But muntadev2in is constructing a protocol intended to prove his paranormal event, and while the JREF is not interested, it may well be that the underlying principles are important for the demonstration to be successful.

 

[Mojo]

But muntadev2in is constructing a protocol intended to prove his paranormal event, and while the JREF is not interested, it may well be that the underlying principles are important for the demonstration to be successful.

I disagree.

The underlying principles are irrelevant to the Challenge, and the rules are quite explicit on this.

Muntadev2in claims that he can distinguish a homoeopathic remedy from a placebo by measuring the skin temperature of volunteers who have taken remedy or placebo. Fine. All we need is a protocol designed to allow him to do this. A discussion of the "principles" involved will only delay discussion of the actual protocol.

We do need to know some practical details, such as how long the test will take, whether it has to be carried out at certain times etc. It is up to muntadev2in to make his requirements clear.

The test should be reasonably simple to set up. The only potential difficulty I can see is the fact that, as currently proposed, it will need to be carried out over several days. We need to prevent the possibility of collusion between muntadev2in and anyone who knows which is remedy and which is placebo (I'm not suggesting that muntadev2in would get up to anything underhand, just that we need to avoid any possibility of this happening). Since the individual tests will apparently only take a short amount of time each day, I would suggest randomising the samples to be used on each day immediately before use.

 

[fls]

The pattern of wave the remedy produces changes from time to time.

I know the wave patern only at a particular time of day. if u we obtain wave patern of a medicine that it can produce all along 24 hours, we can conduct exp at any time, but is very difficult.

If I understand correctly, if you know the wave pattern for a particular time of day for a particular medicine, you can look at two graphs and say "the wave pattern is present" or "the wave pattern is absent". Is this correct?

The sensor variation and subject variation cause changes in the amplitude of the wave, but not in the frequency of the wave. Is this correct?

How difficult would it be for you to repeat your experiments at different times of the day (maybe at two hour intervals) so that you know the wave pattern for 5 different times? That way we could do the test in 2 days, which may be acceptable for the Challenge.

Linda

 

[Cainkane1]

Dear critics of Homeopathy,

I am Dr.Devendra Kumar munta viswakarma MD(Homeo).Long back I had posted a thread "Homeopathy not a placebo". It is known to the JREF regular users.

I am back again with my positive experimental results with Homeoapthic medicines.

Now I am ready for open challenge to prove that Homeopathic dilutions are different from alcohol or water.

Prior to comment, you can have a look at my experimental results and my proposed natural laws upon which Homeopathic dilutions effecting living beings on the earth.

Please have a look at http://homeoresearch.blogspot.com

You can reach me@: muntadev2in@yahoo.co.in

Have a brainfeed.

Ready for challenge.

Dr.Devendra Kumar munta
MD(Homeo)

Herbal medicine is not worthless. This is how modern medicine began in fact. To this day the herbs and potions of healers in primitive societies are found to be useful in the treatment of disease.

 

[steenkh]

I disagree.

The underlying principles are irrelevant to the Challenge, and the rules are quite explicit on this.

Muntadev2in claims that he can distinguish a homoeopathic remedy from a placebo by measuring the skin temperature of volunteers who have taken remedy or placebo. Fine. All we need is a protocol designed to allow him to do this. A discussion of the "principles" involved will only delay discussion of the actual protocol.

But Muntadev2in needs to take the underlying principles into account when he makes the claim and states a protocol. If he thinks the underlying principles make it more advantageous to make the test during full moon (or whatever), then he needs to make a claim and protocol that reflects this. The JREF does not have to know anything about the underlying principles, but here on the forum, if we want to help Muntadev2in construct a claim and a protocol to test for the claim, we will be in a better position to know something about these underlying principles also. Otherwise we may risk proposing protocols that do not take essential principles of the claim into account.

We do need to know some practical details, such as how long the test will take, whether it has to be carried out at certain times etc. It is up to muntadev2in to make his requirements clear.

At this stage, Muntadev2in has not even made a workable claim, and "we" are not going to test him anyway. The JREF will test him, and "we" can choose to help him getting his act together, or not. We do not need to know the underlying principles (which we assume are BS anyway), but if Muntadev2in is unable to formulate a claim (as they often are), we may be able to help if we an idea of what he is thinking.

The test should be reasonably simple to set up.

I agree. But I predict that a workable claim will not be forthcoming nonetheless.

Since the individual tests will apparently only take a short amount of time each day, I would suggest randomising the samples to be used on each day immediately before use.

I agree completely. I think that Muntadev2in has in fact not understood the need for blinding at all, and we may have to explain what it is and why it is necessary.

 

[Mojo]

But Muntadev2in needs to take the underlying principles into account when he makes the claim and states a protocol. If he thinks the underlying principles make it more advantageous to make the test during full moon (or whatever), then he needs to make a claim and protocol that reflects this. The JREF does not have to know anything about the underlying principles, but here on the forum, if we want to help Muntadev2in construct a claim and a protocol to test for the claim, we will be in a better position to know something about these underlying principles also. Otherwise we may risk proposing protocols that do not take essential principles of the claim into account.

He says he can tell the difference using certain equipment. In principle, all we need to do is set him up in a room with his equipment, and give him a couple of volunteers and remedy and placebo to give them. He then does his thing and says which is which. We don't need to know what his "principles" are; just what he claims to be able to do. If he insists that we demonstrate an understanding of his "principles", the whole thing will just get bogged down in irrelevancies.

At this stage, Muntadev2in has not even made a workable claim, and "we" are not going to test him anyway. The JREF will test him, and "we" can choose to help him getting his act together, or not. We do not need to know the underlying principles (which we assume are BS anyway), but if Muntadev2in is unable to formulate a claim (as they often are), we may be able to help if we an idea of what he is thinking.

If he's unable to formulate a claim, I suspect he will also fail to adequately explain his "underlying principles".

I think that Muntadev2in has in fact not understood the need for blinding at all, and we may have to explain what it is and why it is necessary.

He certainly seems to have managed to avoid answering any of the posts that have explicitly mentioned blinding.

 

[Ocelot]

I think I can reduce the time required for a definitive test.

If I'm correct muntadev2in can distinguish not just between a single homeopathic preparations and a placebo but between muliple homeopathic preparations.

He also has two sensors.

With one homeopathic preparation (A) and one placebo (X) there are four combinations that may be given to two volunteers.

AA
AX
XA
XX

Pick one combination at randon and apply it to the volunteers. Correctly determining the true combination has a 1 in four chance of success, by chance alone.

To raise this to the 1 in 1000 chance usually used for a preliminary test we'd need 5 repetitions.

100% accuracy after 5 repetitions (10 trials) would have a 1 in 1024 chance of sucess by chance alone.

However if muntadev2in claim to be able to perform the following test thing start to look up.

Two homeopathic preparations A or B and a placebo X are prepared. Each volunteer may be given one of these at random - there are 9 combinations.

AA
AB
AX
BA
BB
BX
XA
XB
XX

A 100% accuracy over just 4 repetitions (8 trials) is now an adequate test having odds of sucess by chance alone of just 1 in 6561.

Introduce a third homeopathic preparation and we're good with just 3 repetitions (6 trials) a 100% accuraccy would require a 1 in 4096 longshot.

With four homepathic preparations (and still one placebo) we still need 3 repetitions (6 trials). Getting 100% accuracy is a one in 15,625 longshot.

With 5 homeopathic preparations and 1 placebo, correctly guessing who got what, with 100% accuracy over just two repetiaions (4 trials) is a 1 in 1296 longshot.

At this stage we start to think why not do three repetitions (6 trials) and set 5 out of 6 sucessess as the pass mark (1 in 1505)

If muntadev2in can indeed distinguish between multiple different preparations in a single test then I shall prepare some stats to further illustrate what pass mark is required for number of trial vs number of preparations in play. It'd reduce the workload to know the maximum number of different options (including a placebo) that muntadev2in can distinguish between.

 

[GzuzKryzt]

But muntadev2in is constructing a protocol intended to prove his paranormal event, and while the JREF is not interested, it may well be that the underlying principles are important for the demonstration to be successful.

I agree with Mojo.

1. Claim + application.
2. Protocol negotiations.
3. Test.

If muntadev2in delivers two successful tests, he could write (and sell) books about the underlying principles.

Plain and simple: I want to see a proper controlled test showing positive results. Nothing more, nothing less.

 

[steenkh]

I agree with Mojo.

1. Claim + application.
2. Protocol negotiations.
3. Test.

If muntadev2in delivers two successful tests, he could write (and sell) books about the underlying principles.

Plain and simple: I want to see a proper controlled test showing positive results. Nothing more, nothing less.

Of course.

But what is our role in this sub-forum? Some people here offer to help the applicants formulating their claims and protocols, because of the curious inability exhibited by most applicants of forming coherent thoughts by themselves ...

 

[Mojo]

But what is our role in this sub-forum?

Trying to get a definitive claim, for a start. And explaining the concept of blinding.

Then maybe we can move on to a protocol.

 

[JWideman]

Of course.

But what is our role in this sub-forum? Some people here offer to help the applicants formulating their claims and protocols, because of the curious inability exhibited by most applicants of forming coherent thoughts by themselves ...

But that doesn't mean we have to understand the principles, nor is discussion of such helpful. He says that there are identifiable patterns and has images which he believes display these patterns. Thus, all he needs for a successful demonstration is to reliably identify the homeopathic preparations by their patterns. Unless I'm missing something important, he can do two readings per day for five days and identify all ten on the fifth.

 

[muntadev2in]

For the question why have u taken readings at a perticular time I asked to go through my principles.

You no need to consider my principles. Just agree to the time I specified.

I must be able to differentiate which one is medicine and which one placebo. thats it.

Before that I want to conduct a blind study. But even I am confidential. I can perform the test now.

At this stage I can only differentiate a Homeopathic preparation from placebo. but not two preparations.

So protocol should be like this,

8 samples - some are homeopathic preparations, some are placebo (sugar pills without alcohol mix I prefer, but even alcohol mix if you want)

Cover the bottles, keep it in a box and lock, keep it with third party, open the box every day before 1/2 hour the trail starts.

4 days trail period,

8 subjects - 2 per a day - 10 mits record readings per day, at time X (8am - 8.10 am),

4th day - analysis of data for 1 hour - reveal which one medicine and placebo,

uncover the bottles and check results.

thats it,

Is this protocol considerable?

 

[Pixel42]

You no need to consider my principles. Just agree to the time I specified.

Fine. That's part of the protocol then.

I must be able to differentiate which one is medicine and which one placebo. thats it.

So you need volunteers to take a placebo and a medicine each day? Or are you happy for them to select a bottle at random, so some days they might both take placebos and other days both take the medicine? This is an important point because it significantly affects the probability of you getting the right answers by chance, and hence the number of trials you need to do.

Before that I want to conduct a blind study. But even I am confidential. I can perform the test now.

I urge you again to do a blind study before even applying.

At this stage I can only differentiate a Homeopathic preparation from placebo. but not two preparations.

That's fine, but again it increases the number of trials you need to do to produce a result which is significantly better than chance.

So protocol should be like this,

8 samples - some are homeopathic preparations, some are placebo

For a statistically significant trial (see below) you need 10 homeopathic preparations and 10 placebos, and the volunteers need to be able to choose randomly from all 20 each day.

(sugar pills without alcohol mix I prefer, but even alcohol mix if you want)

If the homeopathic preparation uses an alcohol solvent then the placebo must use the same alcohol solvent. If the homeopathic medicine is water based then the placebo must be water. The only difference between the medicine and the placebo must be that the medicine is solvent that has been homeopathically treated. You do understand that such a medicine contains not a single molecule of the original tincture, and is hence indistinguishable from the solvent by any scientific means? That is the paranormal power you are claiming: that you can tell the difference between two solvents which appear identical to science by giving them (neat, or on sugar pills) to volunteers and monitoring their skin temperature.

Cover the bottles, keep it in a box and lock, keep it with third party, open the box every day before 1/2 hour the trail starts.

4 days trail period,

8 subjects - 2 per a day - 10 mits record readings per day, at time X (8am - 8.10 am),

4th day - analysis of data for 1 hour - reveal which one medicine and placebo,

uncover the bottles and check results.

thats it,

Is this protocol considerable?

If you need volunteers to take a placebo and a medicine each day, no. There's a 50% chance you could get the right answer by chance each day, so the chances of getting four right answers by chance is 0.5 X 0.5 X 0.5 X 0.5 i.e. 1 in 16, which is nowhere near the 1 in 1000 usually required for the preliminary test. I calculate 10 days of trials are required to produce the required odds against chance success with this protocol. If you're happy with the possibility of both volunteers taking either two lots of placebo or two lots of medicine each day, Ocelot's calculations suggest 5 trials are required.

 

[JWideman]

For the needed 1 in 1000 odds, you'll need 10 samples. Each sample should randomly be either placebo or homeopathic medicine. Also, the bottles should be labeled by a third party with a code, with the code key held until you've completed your analysis.
Each day you take out any two untested bottles, record their labels and take your readings. For the code itself, I think the numbers from 10 to 19 (to avoid confusing 6 with 9 or 2 with 5) will be sufficient.
On the 5th day, you make your analyses and pass it on to the third party with the code key.
With the coded labels, and the code known only to the third party, there is no need for you to keep the bottles with another 3rd party.
That's basically how you do a double-blind test.

 

[Ocelot]

For the question why have u taken readings at a perticular time I asked to go through my principles.

You no need to consider my principles. Just agree to the time I specified.

I must be able to differentiate which one is medicine and which one placebo. thats it.

Before that I want to conduct a blind study. But even I am confidential. I can perform the test now.

At this stage I can only differentiate a Homeopathic preparation from placebo. but not two preparations.

So protocol should be like this,

8 samples - some are homeopathic preparations, some are placebo (sugar pills without alcohol mix I prefer, but even alcohol mix if you want)

Cover the bottles, keep it in a box and lock, keep it with third party, open the box every day before 1/2 hour the trail starts.

4 days trail period,

8 subjects - 2 per a day - 10 mits record readings per day, at time X (8am - 8.10 am),

4th day - analysis of data for 1 hour - reveal which one medicine and placebo,

uncover the bottles and check results.

thats it,

Is this protocol considerable?

Under this protocol where you already know that in each day one person gets a placebo and the other gets a homeopathic preparation the odds are not good enough for the JREF to stake $1 million.

The chance of getting your first test right by chance alone is 1 in 2 or 0.5 or 50%.

The chance of getting all four tests correct is 0.5 x 0.5 x 0.5 x 0.5 = 0.0625 or 6 1/4 % or 1 in 16

That would be a nowhere near significant result. Even given that this is a preliminary test and that the same test must be repeated before the $1m is awarded there's a 1 in 256 chance that somebody just guessing could win the million.

If those odds were acceptable to the JREF every gambler in vegas who could come up with a protocol that cost less than $3,900 to test would jump at those odds. The JREF would be inundated with people claiming to be able to correctly guess the toss of a coin four times in a row.

As I've said above, there is a way to do this in just 5 repetitions, (10 trials)

Under that protocol, instead of knowing that one of your pair has been given the homeopathic preparation and one has been given the placebo, you deall with each person individually. All you know is that on the toss of a coin there's a 50% chance that they'll have been given the placebo and a 50% chance that they've been given the homeopathic preparation.

As such, for every repetition there are four rather than two possible outcomes.

Both have been given the placebo,
The first has been given the placebo whilst the second has been given the homeopathic preparation.
The second has been given the homeopathic preparation whilt the first has been given a placebo.
Both have been given the homeopathic preparation.

Over the course of 5 repetitions (10 trials) there are 1024 possible combinations which I won't list here except for to mention that they include nobody being given the placebo or everybody being given the placebo.

It's equivalent to being able to guess the toss of a coin 10 times in a row. Enough to scare away the gamblers.

I believe such a test should fullfill the requirements of the JREF.

 

[arthwollipot]

Herbal medicine is not worthless. This is how modern medicine began in fact. To this day the herbs and potions of healers in primitive societies are found to be useful in the treatment of disease.

[OT]That's because herbal medicine differs from homeopathy in actually having an active ingredient.[/OT]

 

[muntadev2in]

For the needed 1 in 1000 odds, you'll need 10 samples. Each sample should randomly be either placebo or homeopathic medicine. Also, the bottles should be labeled by a third party with a code, with the code key held until you've completed your analysis.
Each day you take out any two untested bottles, record their labels and take your readings. For the code itself, I think the numbers from 10 to 19 (to avoid confusing 6 with 9 or 2 with 5) will be sufficient.
On the 5th day, you make your analyses and pass it on to the third party with the code key.
With the coded labels, and the code known only to the third party, there is no need for you to keep the bottles with another 3rd party.
That's basically how you do a double-blind test.

this is acceptable to me,

No where I have mentioned, one for medicine and one for placebo,

subject can pick up samples randomly.

out of 10 samples there is no limit for medicne and placebo samples. is then ok?

 

[GzuzKryzt]

this is acceptable to me,

No where I have mentioned, one for medicine and one for placebo,

subject can pick up samples randomly.

out of 10 samples there is no limit for medicne and placebo samples. is then ok?

Perhaps you meant to say this: The subject is given a randomly selected sample.

Neither you nor the subject nor any person present in the room must know what the subject has been given. That would be proper double blinding.

What do you think of Ocelot's protocol proposal?