developing a double blind experiment

24 posts / 0 new
Last post
TemporalHominid TemporalHominid's picture
developing a double blind experiment

 

TemporalHominid TemporalHominid's picture

The objective is to develop a double blind experiment for the claim: 'Magnets help cheap wine taste better'.

First we must determine if the claim 'Magnets help cheap wine taste better' is a falsifiable claim. This means that there must be an experiment or possible discovery that could prove the claim untrue. The experiment must be repeatable. Any suggestions?

Note, a person does not have to be a scientist to develop a double blind experiment. Children are participating in science everyday in school, and are creating double blind experiments for science fairs.

Once we have determined some experiments that could prove the claims untrue we can set up a double blind procedure.

If I am skipping steps or I am in error at any point, please inform me and we will start all over. Science is a process and sometimes there are false starts and errors made along the way, but that does not invalidate an inquiry.

Please don't derail this thread with ad hominem attacks, or with anecdotal (testimonial) evidence.

[ 25 March 2007: Message edited by: TemporalHominid ]

Nanuq

Start by defining your terms. What kind of magnet are you going to use, how will you measure the strength of the magnet. What constitutes a "cheap" wine? How is it different from an "expensive" wine? And what do you mean by "tasting better". How is that determined?

TemporalHominid TemporalHominid's picture

how about

'exposing the wine to a magnetic field will change the shape of the tannins, which has the benefit of making the wine smoother and less harsh.'

the cheap wine would be any wine under 6 dollars a bottle.

marzo

Should the magnets be immersed in the wine?
Should the magnets be placed outside the bottle?
Should the magnets be placed in the cup or glass before drinking?
Should the wine drinker quaff the magnetized plonk with the magnets soaking in the liquid?
If this experiment does not produce better tasting wine, the cheap wine can be mixed with something that tastes better.

Unionist

I'm attracted to cheap wine. Are you looking for volunteer magnets?

HeywoodFloyd

Mythbusters did a similar bit with vodka and carbon water filters and it improved the vodka.

Lard Tunderin Jeezus Lard Tunderin Jeezus's picture

quote:


Should the wine drinker quaff the magnetized plonk with the magnets soaking in the liquid?

And if so, is the taste still improved to a greater or lesser degree by a non-magnetized piece of iron?

Stephen Gordon

Hmm. You'd have to set up several experiments to exclude alternative explanations of the results. For example, you might have

- people who can't tell the difference between cheap and expensive wine, or
- systematic preferences for the first (or second) glass.

For example, this might test the hypothesis, while at the same time, excluding these alternative explanations:

A) Try a $6 wine, and then the same wine exposed to a magnetic field.
B) The same as A), but in reverse order.
C) Try a $6 wine, and then an untreated $20 wine.
D) The same as C, but in reverse order.
E) Try a $6 wine, and then a different $6 wine.
F) Same as E, but with $20 wines.

If the hypothesis were true, we should expect that in

A) They prefer the second glass
B) They prefer the first glass
C) They prefer the second glass
D) They prefer the first glass
E) They find them the same
F) They find them the same

If you can think of other alternative explanations (I'm sure I missed many), then you have to set up more sets of experiments.

Legless-Marine

quote:


Originally posted by TemporalHominid:
[b]The objective is to develop a double blind experiment for the claim: 'Magnets help cheap wine taste better'.

First we must determine if the claim 'Magnets help cheap wine taste better' is a falsifiable claim. This means that there must be an experiment or possible discovery that could prove the claim untrue. The experiment must be repeatable. Any suggestions?

Note, a person does not have to be a scientist to develop a double blind experiment. Children are participating in science everyday in school, and are creating double blind experiments for science fairs.

Once we have determined some experiments that could prove the claims untrue we can set up a double blind procedure.

If I am skipping steps or I am in error at any point, please inform me and we will start all over. Science is a process and sometimes there are false starts and errors made along the way, but that does not invalidate an inquiry.

Please don't derail this thread with ad hominem attacks, or with anecdotal (testimonial) evidence.

[ 25 March 2007: Message edited by: TemporalHominid ][/b]


Why does this experiment have to be Double-blind? It seems to me that single-blind will be sufficiently accurate for the sake of this exercise.

The benefits of doing double-blind trials, in this context, will be negligible, but will cause a significant increase in complexity.

M. Spector M. Spector's picture

This is actually the simplest of all kinds of double blind experiments to organize.

All you need are two identical bottles marked A and B, one containing the magnetized wine, and the other containing the exact same wine, only unmagnetized. You arrange for several volunteers who self-select as wine fanciers to taste a sample from each bottle and record whether they prefer A or B.

Two provisos: All the wine would have to come from the same vintage and production batch to ensure uniform quality (maybe buy a large box or jug of it). And the "double" in "double blind" means that the person serving the wine to the volunteers (and any other persons interacting with them) must be kept in the dark as to which bottle contains the magnetized wine.

If the volunteers drink too much, give them a glass of distilled water and tell them it's a homeopathic remedy for drunkenness. It will sober them right up in no time.

Agent 204 Agent 204's picture

quote:


Originally posted by Legless-Marine:
[b]

Why does this experiment have to be Double-blind? It seems to me that single-blind will be sufficiently accurate for the sake of this exercise.
[/b]


If the person serving the wine knows which is which, s/he might unconsciously give cues to the tasters. A double blind experiment will eliminate this possibility.

Bubbles

quote:


If the person serving the wine knows which is which, s/he might unconsciously give cues to the tasters. A double blind experiment will eliminate this possibility.

Maybe not.

quote:

Increasingly, modern science proposes that there is no universal external reality that can reasonably be expected to draw all observers to the same "facts." It has become ever clearer that the hidden assumptions with which we approach a given entity largely determine the results we will get. While the generation that followed Descartes was confident that the outside world had its own independent structure, we are forced to concede that we create much of that structure ourselves.

This external independent reality was a basic presumption of the mechanists. If, however, we acknowledge our part in forming that reality, we must then admit that our perceptions shape both causative links and rules for measurement. In one fell swoop, causality and objectivity become constructs of a still-evolving human consciousness. The greatest single element of change has been a recognition that one sees what one believes, rather than the reverse. We find ourselves attempting to adjust a changing phenomenon - human consciousness - in relation to a moving target - the expectations of society.

In attempting to understand this new awareness, one must again follow the path of scientific discovery. Unfortunately, this more recent history (much of it set in the sometimes arcane world of post-quantum-theory physics) has been less accessible to nonspecialized observers than the opposing system of ideas, which has been assimilated for four centuries. Surely the time has come to recognize the changed scientific environment and its implications.


[url=http://www.humanehealthcare.com/Article.asp?art_id=248]Towards quatum reality?[/url]

[ 25 March 2007: Message edited by: Bubbles ]

Legless-Marine

quote:


Originally posted by Agent 204:
[b]
If the person serving the wine knows which is which, s/he might unconsciously give cues to the tasters. A double blind experiment will eliminate this possibility.[/b]

I understand the concept - It's the requirement for this degree of accuracy I am unclear on, particularly from someone who has limited experience designing science experiments.

[ 25 March 2007: Message edited by: Legless-Marine ]

the grey

quote:


Originally posted by Stephen Gordon:
[b]
If the hypothesis were true, we should expect that in

A) They prefer the second glass
B) They prefer the first glass
C) They prefer the second glass
D) They prefer the first glass
E) They find them the same
F) They find them the same
[/b]


Not true. You are assuming that all cheap wines are created equal -- they aren't. On top of that, some people prefer different characteristics to their wine than others. It's entirely reasonable to imagine that an individual will prefer a specific $6 bottle to a specific $20 bottle, even if that individual generally prefers $20 bottles to $6 bottles.

But none of that is relevant to the test. The question being asked is just about the impact on $6 bottles. There's a need to ensure the $6 bottles are identical (other than magnetized / unmagnetized), and that the order of serving them is controlled and rotated. But there's no need to bring $20 bottles into it.

Preference for the taste of food is subjective. When conducting an experiment testing for it, there is a need to have a large sample group to account for subjective differences.

For instance - if the hypothesis is that magnetization diminishes the impact of tannins in the wine, the results would be different if the tasters as a group prefered stronger tannins, as opposed to a group that disliked stronger tannins.

Nanuq

quote:


Increasingly, modern science proposes that there is no universal external reality that can reasonably be expected to draw all observers to the same "facts." It has become ever clearer that the hidden assumptions with which we approach a given entity largely determine the results we will get. While the generation that followed Descartes was confident that the outside world had its own independent structure, we are forced to concede that we create much of that structure ourselves.

The old "mechanistic" physics work just fine for anything larger than an electron. Attempts at applying quantum physics outside of the subatomic realm have tended to fail miserably. Trying to use it to explain psychological phenomena definitely doesn't work. When you're standing in the middle of a street and a truck is bearing down on you, it really doesn't matter how you construct your reality. If you don't get out of the way, you're going to get hit.

Bubbles

quote:


The old "mechanistic" physics work just fine for anything larger than an electron. Attempts at applying quantum physics outside of the subatomic realm have tended to fail miserably. Trying to use it to explain psychological phenomena definitely doesn't work. When you're standing in the middle of a street and a truck is bearing down on you, it really doesn't matter how you construct your reality. If you don't get out of the way, you're going to get hit.

[img]smile.gif" border="0[/img] Nanuq, I certainly would get out of the way, because that is what my believe tells me to.

I was basically going back to the health issue. If it should be purely science based? My view is that as long as science cannot account for all deseases/ sicknesses we should not purely base cures on science, and be open to alternatives. Double blind testing has its value as long as one keeps in mind that there are still many variables that we are not aware of.

M. Spector M. Spector's picture

Variations on a quantum scale are irrelevant to medical clinical trials using double-blind methods.

Name one single instance where a double-blind clinical trial has produced a statistically incorrect result as a result of quantum mechanics.

Bubbles

quote:


Name one single instance where a double-blind clinical trial has produced a statistically incorrect result as a result of quantum mechanics

Ordering! I return in kind.

Type in Google,quantum mechanics affecting drug uniformity, and go from there. Only 750,000 entries.

the grey

quote:


Originally posted by Bubbles:
[b]

Ordering! I return in kind.

Type in Google,quantum mechanics affecting drug uniformity, and go from there. Only 750,000 entries.[/b]


Gee, I only got 47,000 -- and judging from the first few, none had anything to do with what we're talking about.

Bubbles

Sorry the grey. You are right. Was in a bit of a hurry this afternoon. I cannot remember the exact wording I used in Google. But tonight I tried, double blind testing quantum effects (700 000 hits) Did a bit of searching and found [url=http://www.niehs.nih.gov/emfrapid/html/EMF_DIR_RPT/Dir_Comments/CD_Files.... Did not have the time to read the whole thing, but on a quick scan came to the conclusion that it might fit the discussion. It deals with electro magnetic effects which might apply to the cheap wine discussion.

quote:

Firstly, it must be realised that there is a duality between frequency and chemical structure
without which chemical analysis by spectroscopy would be impossible. Secondly, the sole
concern of the EMF RAPID Report (and of most research) is directed towards “classical”
“classical” physics aspects of electric and magnetic fields but, there is no chemistry in
physics. To seek for possible biochemical effects of electromagnetic fields, one must
consider “quantum” physics.

Blondin

All the discussion about quantum mechanics has nothing to do with the original question about double blind testing. The first step should be to find out if there is a detectable difference. If there is then you can argue about how & why.

As M. Spector pointed out all you need is several identical glasses of wine (some of which were "magnetized" and some which weren't), some tasters and a couple of helpers.

One person would pour the wine into the glasses which would be marked in some non-prejudicial way (numbered 1 to 20 or something). This person would use dice or some other random method for deciding which glasses were treated with the magnets and only this person would know which are which.

A second helper would serve the wine to the tasters. Neither the tasters or the server should have any clue which are which. The whole point is to see if the tasters can detect a difference by taste alone. They should only be required to mark each glass in column A or column B. Never mind if one glass tastes older or has more tannin or anything else. They don't even need to say which they think is the treated wine. They only have to detect enough of a difference to divide the glasses into 2 groups.

A significant result would be one where more than 50% of the glasses were in the correct group. If they can't detect any difference or if the treated glasses are not mostly in one group or the other then it's probably safe to say the magnets had no effect on the wine. At least if they did it wasn't detectable so you might as well save that money you were going to spend on sooper-dooper wine magnets.

M. Spector M. Spector's picture

quote:


Originally posted by Blondin:
[b]They don't even need to say which they think is the treated wine.[/b]

In fact, there is no reason to tell them the purpose of the experiment, to mention "treating" the wine, or to mention magnets. Just ask them, "which of these two tastes better to you?"

And give them an option of saying they both taste the same (because in actual fact, they do - magnetic treatment of wine is a crock!).

The magnetic "treatment" is normally done while the wine is in a bottle, rather than treating each glass of wine separately. For the purpose of the experiment it is preferable to treat the wine in the bottle, for these reasons:

1. It simulates the actual treatment methods used by the advocates of magnetic treatment, so that they would not be able to reject the results on the basis of a difference in methodology.

2. It allows for uniformity of "treatment" of the wine, eliminating the possibility of minor treatment variations form glass to glass.

3. I'm not sure what the recommended magnetic exposure time is, but while the wine is being "treated" with the magnets, it's safer from contamination, oxidization, or other degradation if it stays in the bottle, rather than being in a drinking glass.

Blondin

Thanks, M. Spector. I was going to add that myself. In fact the experiment is more objective if the subjects do not know what hypothesis is being tested.

I've seen several variations on the wine magnet fad. A couple that are contraptions that fit around the neck of the bottle so the wine is poured through the magnetic field and at least one that is just a sort of magnetic coaster to sit the bottle on. Another is a sort of holder you slip the bottle into so that magnets are positioned along the sides.

I've come across several forums where this subject has been discussed and it always strikes me that the proponents for these devices are more interested in diverting the conversation to the credentials of the tasters, theories about how or why the magnets might affect the wine, testamonials from satisfied customers or bullshit arguments about the inadequacies of double blind tests rather than about objectively testing to see if anybody really can taste a difference.

Rarely do you hear of actual objective testing being done. An awful lot like that other pseudo-scientific, pretentious community of twits - audiophiles who spend thousands of dollars on magical wires, laquer, bits of plastic, etc. But don't get me started...

Please note that I am not disparaging all wine and audio afficionados, only the ones who refuse to consider the merits of such claptrap objectively.