Home |
Search |
Today's Posts |
#82
|
|||
|
|||
science vs. pseudo-science
(Audio Guy) wrote in message news:MpFcb.576680$YN5.411073@sccrnsc01...
In article , (ludovic mirabel) writes: (Audio Guy) wrote in message news:zalcb.565817$Ho3.102946@sccrnsc03... (ludovic mirabel) writes: If you are happy with a "test" that gives as many different results as there are people doing it, who am I to stop you? Use it. You'll get yours. Audio Guy: Where is your evidence that audio DBTs "gives as many different results as there are people doing it"? So far it is only your mistaken interpretation of the test statistics. How about some real evidence? Below find the results of of Greenhill's ABX cable test (The Stereophile ,1983) A "hit" is 12 correct answers out of 15. Note different performers, performing differently. (Surprise, Surprise!). Note Nr. 6; 1.75db level difference but music is the signal. Compare with test 1 and test 4. I will not rediscuss the "statistics". This was thrashed out ad nauseam here. If it tells you something different from what it tells me, well and good. SUBJECTS: A B C D E F G H I J K Test1: Monster vs. 24 g. wire,Pink noise 1.75db level difference 15 14 15 15 15 15 15 15 15 15 15 2. Same but levels matched 9 13 7 10 na. 8 9 6 14 12 12 3. Monster vs. 16 gauge zipcord, Pink noise 13 7 10 7 11 12 9 9 11 12 7 4.. 16 ga vs. 24 ga., Pink noise 15 15 na. 14 15 na 15 14 15 15 15 5. Monster vs. 16ga., choral music 4 6 11 8 9 5 5 7 6 10 10 6. Monster vs. 24ga, choral music 1.75db. level difference 14 7 15 10 8 10 6 10 11 12 10 ______________________________________________ % of "hits" in the total of 6 tests, 90 tries. 67. 50 40 33 40 40 33 33 50 83 50 L.M.: It tells me that people can easy tell level differences with pink noise, not so easily with music. Where does it "gives as many different results as there are people doing it"? You're right. I got sort of dozed off looking at it all and got carried away. It is sometimes 10/11 the same results sometimes 3/11 and a few in between. It all adds up beautifully. I wish you and Mr Nunes who "has just done that" many happy hours with the ABX and pink noise. Ludovic Mirabel I will not stop you and I will not continue this pointless scholastic argument. How is it a "scholastic argument"? Are you using this label so you can side-step the issue? Ironically many would consider all of your arguments purely "scholastic arguments". Yes you're correct: scholastic arguments, including mine, are about something unproven. When you or someone like Mr. JJnunes comes up with experimental evidence that ABX is the right tool for COMPARING COMPONENTS and that for instance it does not interfere with perception of their musical characteristics we'll be talking about realities. I believe Mr. Junes has just done that. excessive quoting snipped |
#83
|
|||
|
|||
science vs. pseudo-science
In truth I have little to substantiate to you. Every few
weeks you post a personal attack ( quotes below) with no other content or you truncate and distort my words . Every time when shown up you go silent for a month or two. I find it distasteful to go over all this stuff again but since you force me you'll find requotes below. There's more if you want it. Just ask.. . You seem to have some kind of immunity for this kind of thing here. I won't claim the same. Now you thought up a new wrinkle: insinuation that I'm lying about my professional record.. This of course has nothing to do with an argument about the proper way to compare components but anything will do. One thing I know : However inadequate my DBT exposure cxould be it exceeds yours by miles. I apologise to the readers for what follows and invite you to skip it. It is not MY choice of the way to discuss opinions. ----------- ------- After 5 years of postgraduate training in internal medicine I became a full time, and the only resident researcher in the schistosomiasis (bilharzia) unit, a division of the Tropical Disease Research of the Med. Research Ccil of U.K. in Hertford in 1951/2. Head: Dr Newsome. My lab technician: Mr England (yes!). We had also an Egyptian on a fellowship from his Govt. I forgot his name but remember him for memorising a 900 page textbook of Neurology in one week. I already had some drug research DBT experience.. I was a Senior House Officer in Brook Hospital, London. A Dr. G. Loxton, rheumatologist, was trying out a "wonder drug" for rheumatoid arthritis (dexoxycorticosterone with vitamin C). Initial enthusiasm cooled after a DBT. M.R.C. was the organisation and that was the time when and where the principles of randomised DB drug testing were being developed principally by the statistician Bradford Hill. My unit was researching the proposed bilharzia drugs effects on infected animals (baboons and "desert rats") and planning human trials but none of the drugs we tested warranted it as yet.. I resigned after one year after passing my specialty exam in int. medicine (M.R.C.P. Ed). I decided that I preferred clinical medicine to research and as there were no openings for me in U.K I emigrated to Canada. Hundreds of others newly qualified specialists in U;K; had to (or U.S. or Australia) because under the system in U.K. a specialist- consultant doesn't just hang out his shingle. You have to wait for someone to retire or die and then be selected by a hospital in preference to others, most at least equally bright. Afterwards, as the consultant cardiologist ( solo for a few years till others joined me) in a large suburban hospital I HAD TO keep up with DBT's. Randomised DBT drug trial has been staple in medicine for decades. No new treatment without DBT. It was the air we breathed. At that there are constant arguments about the adequate selection of controls, significance of results etc. . Proper DBT at that with objective body changes to assess at the end, symmetrical placebo control group etc.- not a question and answer ad hoc "listening test" or a home ABX switch kit.------ End of personal stuff The only reason I first brought up my research experience here was because people like you with qualifications in eg. electronic engineering kept questioning my right as an audio consumer to express my views on the DBT tunes sung in RAHE. Perhaps you felt I was trespassing on your territory as the all- round audio oracle. I happen to react to the local authorities laying down the law about things they know no more about than anyone else. Someone like me was long overdue in RAHE. If only to infuriate the pompous importances. That you'd imagine that disagreeing with you is important enough for anyone to falsify his credentials tells more about you than I care to know. As for your "references"- whom are you kidding Pierce? You know perfectly well that the argument is about ABX as THE test for ordinary audio consumers for COMPARING COMPONENTS- NOT ABOUT PSYCHOACOUSTIC RESEARCH. Of course I do not spell it in full every time- I'm assuming the minimum of decent discussion manners. After all I said it at least twenty times already. Of course you would quote a dozen references of which only one (Toole) may have some bearing on the subject. Who is actually comparing *what* components in your "references"?. With what results? You hope no one will read this stuff carefully , right? Why don't you quote the index of JAES for one year? You don't want to compete with Jjnunes 18 titles.? For your information this is what a proper reference looks like.: S. J. Wilson & al "Comparing the quality of oral anticoagulant management by....A randomised controlled clinical trial" C.M.A.J., vol.169, No4, 293, '03. This IS a reference. You want to know about management of anticoags, this is what you look up. You want to know about the usefulness of ABX to untrained, unselected audio fans for comparing components you don't look in these irrelevant collections In the meantime I'll repeat what I said to Jjnunes: " "Just to spur you on I will now state emphatically that you talk about "trajectories" for lack of anything better. NOTHING in Fletcher, NOTHING in Yost, NOTHING in good old Moore. And you know what else? NOTHING ANYWHERE ELSE. The reputable, published basic research for the consumer use of DBTs in comparing audio components does NOT EXIST. "Starting points" and "trajectories" will not replace it. You were asked for nothing complicated. Just a very simple thing called: quotable evidence. Remember "evidence"?. Remember quote?" And I'll add this: you Mr. Pierce do not have the foggiest about how to transfer Component Comparison DBT from the lab to the street. If you did you would not be quoting your pseudo "references"..Read my soon to appear ( I hope) posting about S. Toole's loudspeaker comparison in his laboratory to begin learning. I can not stop wondering what accounts for the hostility in the samples quoted below. Gourmets argue about food, wine drinkers about wines, piano players about pianos. No Gallo drinkers claim that they have a "test"that will show up those damn Burgundy and Bordeaux lovers. Audio seems to breed a particularly embittered and combative swarm of discussants.. Ludovic Mirabel Samples of Mr. Pierce debating methods. June 25 "Why Dbts in audio do not deliver?" "Well, the answer is VERY simple: DBT does not deliver what people like Ludovic want. It does not support THEIR agenda, it does not validate THEIR preferences, indeed, it does not elevate their preferences to the level of universal fact. In that sense, indeed, ANY testing of ANY kind will NEVER deliver what they want, except that testing that gives the results they expect. It basically reduces to the fact that if you don't get the answer you expect, blame the question, but NEVER entertain the posibility that not so much the expectation itself is wrong, but the very fact that you HAVE an expectation is the issue. Science certainly works hard to give you answers, it just doesn't give a sh*t whether you like the answer or not. THAT'S why DBT doesn't work: because it does." No quoted argument of mine in the whole posting, Just this. July 8 same thread: "I'd posit, instead, that Ludovic simply engages in a continuous 2. stream of misrepresentation. Why? 1. It's inadvertant. He doesn't no better. Poor Ludovic. Poor us for having to slog through his irrelevant misrepresentations. 2. It's deliberate. He has no sound foundation for whatever the hell it is he's arguing about and simply to keep his side of the conversation going, he just makes stuff up because he has absolutely nothing to contrinute of any relevance or substance. The evidence, especially in the form of the quoted text above, would seem to have one lean in the direction of deliberate and malicious misrepresentation." It continues in the same vein. And there are plenty more like this. It goes back two years. Just ask, Mr. Pierce, and I'll oblige. T. Poulsen, "Application of psychoacoustic methods," H. Staffeldt, "Evaluation and scaling of timbre in listening tests on loudspeakers," F. Toole, "Planning of listening tests - technical and environmental variables," A. Gabrielsson, "Planning of listening tests - listener and experimenatl variables," S. Bach, "Planning of listening tests - choice of rating scale and test procedure," N. Kousgaard, "The applicatin of binary paired comparisons to listening tests," M. Williams, "Choice of programme material for critical listening to loudspeakers," S. Pramanik, "Inadvertant bias in listening tests," F. Toole, "Correlation between the results of objective and subjective tests," All present at and found in the Proceedings of the Symposium on Perception of Reproduced Sound, Gammel Avernaes, Denmark, 1987. __________________________________________________ __________ "Dick Pierce" wrote in message ... wrote in message ... On Wed, 24 Sep 2003 17:08:01 GMT, (ludovic mirabel) wrote: This is the 4th request to Mr. Jjnunes for references I don't take you seriously, especially since these have posted before and you argued against them with your usual absurd rhetorical games. I'm not interested in continuing further. You apparently can't even come to terms with audio components being reproducers of sound and play rhetorical games about them being 'producers of music' to thus provide yourself with an avenue to argue from the same pretext as to how musical instruments are compared. That was why I made the mistake of pointing you at the books I mentioned --- to provide a foundation to look further into the subject. You aren't interested in rational debate about this, but rather rhetoric and rhetoric only. I should have learned this lesson sooner. Indeed, but hope springs eternal even, it seems, for our persistant Mr. Ludovic. In addition to Mr. Nunes' excellent references, every one of which I would wager Mr. Ludovic has never read and will ignore, I would merely AGAIN, as I did some months ago, point out the following references: T. Poulsen, "Application of psychoacoustic methods," H. Staffeldt, "Evaluation and scaling of timbre in listening tests on loudspeakers," F. Toole, "Planning of listening tests - technical and environmental variables," A. Gabrielsson, "Planning of listening tests - listener and experimenatl variables," S. Bach, "Planning of listening tests - choice of rating scale and test procedure," N. Kousgaard, "The applicatin of binary paired comparisons to listening tests," M. Williams, "Choice of programme material for critical listening to loudspeakers," S. Pramanik, "Inadvertant bias in listening tests," F. Toole, "Correlation between the results of objective and subjective tests," All present at and found in the Proceedings of the Symposium on Perception of Reproduced Sound, Gammel Avernaes, Denmark, 1987. Through all, this, our dear Mr. Ludovic has pounded his fist and the occasional show on the table DEMANDING references, when provided with same, he has simply pounded louder. And he has ALSO made claims about the unsuitability of blind testing, claiming his experience in the medical field and testing. He has done so, it would seem with NO substantiation of those claims. I think its time we called his bluff: Mr. Ludovic, you have provided NO substantiation that you have ANY experience in the field of blind testing or medical research. Your claims, indeed, could well be interpreted as belonging to someone who has, at best, very limited, casual and peripheral experience in the realm. We, thus, kindly ask YOU to substantiate YOUR claims of experience in those fields which you claim some experience. Where are YOUR published papers? With whom were YOU affiliated? What research projects have YOU been a principal or support investigator on? Please, we have provided DOZENS of references for YOU, how about providing us the same. After all, aren't YOU subject to the very same criteria that you subject others to? |
#84
|
|||
|
|||
science vs. pseudo-science
(Stewart Pinkerton) wrote in message ...
On Wed, 24 Sep 2003 23:18:09 GMT, (ludovic mirabel) wrote: (Stewart Pinkerton) wrote in message ... On Tue, 23 Sep 2003 17:31:45 GMT, (ludovic mirabel) wrote: "All Ears" wrote in message news:97Fbb.406782$cF.126279@rwcrnsc53... Mirabel wrote: In '89 a rather elaborate listening test for audibility of distortion was performed .(Masters and Clark, St. Review, Jan. '89). Various types of distortion with different signals were tested.. There were 15 TRAINED listeners -? Gender?. At 2 db. distortion level (2db), playing "natural music" the "average" level of correct hits was 61% (barely above the minimum statistically significant level of 60%). The individual scores varied from perfect 5/5 to 1/5 Similar discrepancies were observed in phase shift recognition.: Authors' conclusion: "Distortion has to be very gross and the signal very simple for it to be noticed" ... by the "average" Will it do for the time being? All Ears commented: I think this is good "food for thoughts" because it gives an idea of how large a margin there is to really detect a difference. I answered: Note that the performance varies from one listener to other- inspite of training and retraining-. A few have 5 out of 5 correct responses, a few 1 out of 5 and most fall in the average middle. As you would expect. Pinkerton: Indeed, as you would expect for an effect which is on the threshold of audibility. Note that the "objectivist", objectively unbiased, proctors showed no interest in the few who heard the DIFFERENCES accurately. They just lumped them together with the most who DID NOT and got an average for an average, fictitious Mr. Average Listener who hears no differences-ever. This of course was in acordance with the "Stereo Review" guiding principle- "the high end does not exist, our big account advertisers sound just as good." You are once again making the classic mistake (or is it yet another deliberate distortion?) of ignoring the basis of statistics. *You* invariably 'cherry pick' the results that suit your preconceptions, the researchers above very properly included *all* the responses. They most certainly did *not* 'ignore' the 5/5 response, they *included* it in the results. Incidentally, you once again alter the facts to suit yourself. There is no indication in the above report that there were 'a few' listeners who scored 1/5 or 5/5, there may have been only one of each - as a standard distribution curve would suggest. Now, if the researchers had *repeated* the experiment, do you presume that the same listeners would score the same results, i.e. that the 5/5 scorer(s) really do have 'Golden Ears'? That would suit *your* preconceptions, but the results of the recently posted TAG McLaren tests did not show this. Once again, you attempt to ignore the very basis of statistical probability, in an attempt to shore up your prejudices. I love "probability". But when someone is correct 5 times out of 5 and someone else 1 out of 5 ,or 74 times out of 90 (like Greenhill's "golden ear") and not 30 out of 90 like his other testees I would be curious how they would do on a repeat. In other words I'd experiment. Of course things are diferent when one is a pure, scientific statistician/mathematician like Mr. Pinkerton. Cherry-picking is taboo, experiments-hell, we have probabilities and OUR probabilities are certainties. And paper is patient. Now for the "golden ear". For the last time (What a hope!) Greenhill said: "The final significant conclusion is that at least one genuine "golden ear" exists". The only time I used the term was when relating his results. Greenhill was the cable test proctor and the writer ("The Stereo Review" Aug. 1983, p. 51), He is also former collaborator of Mr. Krueger, who I believe invented ABX. He is also alive and well and writing for "The Stereophile". WHY don't you tell him what you think of his statistics and his "claim". You have a TAG test (whatever that is) on your side. One more little thing: I said all this stuff to you before. You still twist the facts to suit your polemic. Go home and write the Pinkerton comment you would write if I did such a thing. For economy I'll refer to my today's answer to Mr. Strong. Which is completely refuted by the TAG test, which found that the better performers in one test, were average or worse performers in the other test. I don't know the test but I guess that you're saying that it contains the ultimate truth that makes any further experiment unnecessary, I'll add only that when Mr. Pinkerton posts his results of his tests in this group that is not "cherry picking". Indeed it's not, since I posted both positive *and* negative results. You no doubt would have claimed that the negative results were in some mysterious way flawed, and/or that you were simply 'bad at DBTs', and that some unnamed other person would of course have obtained no negative results. 1) Why would anyone in his senses say that ALL the negative results are flawed? I would not. Is this what some discussants here wittily call a "strawman?" 2) And Greenhill posted only the positive ones?? My poor nonmathematical head is spinning. I see- He had himself and one or two of his friends in his amplifier "test". If he added 10 "audiophiles" he would add up all their results- and let the dice fall as they may- even if Krell turned out not distinguishable from Panasonic integrated- right? Right. I wonder if he ever sat an exam.? Far too many! :-) I wonder if he'd like the collective results averaged or would he want himself to be cherry-picked. Different situation, as I have a personal interest in my own results. Audiophile friends with an interest in their own results would no doubt conduct further tests for themselves. The analogous situation is where I perform lots of tests on myself, to verify my own abilities, but only limited tests on others, to verify that I am just one of many with similar perceptual abilities. You have shown absolutely *no* evidence of the existence of 'Golden Ears', indeed all the available evidence suggests nothing more than standard statistical distributions according to random chance. Sorry. Your mathematics are beyond me. I have no idea what you're saying. This does not seem to prevent you from *claiming* that such people somehow must exist - somewhat like Bigfoot. This is witty. I go to the Okanagan (in B.C.) and I saw the Bigfoot talking to the "Golden Ear". In person Absurdity in the service of winning a debate on paper could not go any further I entirely agree............. This is witty too. The insinuating: " Or is it yet ANOTHER DELIBERATE DISTORTION" is par for the gentleman. It tells more about the way he thinks than he'd like to be known and is another one of his contributions to the gentler , kinder RAHE debating manners. It tells people *exactly* what I'd like to be known, which is that you simply *refuse* to engage in honest debate, instead using every possible trick to avoid the inevitable conclusion, that you simply do not have a case to argue. Name one participant in RAHE discussions who disagrees with you on DBT matters and who does "engage in honest debate" with you. If you recall it is not Mkuller or Harry Lavo. Name one "honest" man who disagrees with you ,honestly by your lights, about testing cables. It is probably ( statistically of course) true that you really believe that people normally lie to argue a point. That is a kind of insight. Ludovic Mirabel |
#85
|
|||
|
|||
science vs. pseudo-science
(Audio Guy) wrote in message news:XWldb.603104$o%2.282900@sccrnsc02...
In article O36db.436378$Oz4.243723@rwcrnsc54, (ludovic mirabel) writes: (Audio Guy) wrote in message news:MpFcb.576680$YN5.411073@sccrnsc01... In article , (ludovic mirabel) writes: Below find the results of of Greenhill's ABX cable test (The Stereophile ,1983) A "hit" is 12 correct answers out of 15. Note different performers, performing differently. (Surprise, Surprise!). Note Nr. 6; 1.75db level difference but music is the signal. Compare with test 1 and test 4. I will not rediscuss the "statistics". This was thrashed out ad nauseam here. If it tells you something different from what it tells me, well and good. SUBJECTS: A B C D E F G H I J K Test1: Monster vs. 24 g. wire,Pink noise 1.75db level difference 15 14 15 15 15 15 15 15 15 15 15 2. Same but levels matched 9 13 7 10 na. 8 9 6 14 12 12 3. Monster vs. 16 gauge zipcord, Pink noise 13 7 10 7 11 12 9 9 11 12 7 4.. 16 ga vs. 24 ga., Pink noise 15 15 na. 14 15 na 15 14 15 15 15 5. Monster vs. 16ga., choral music 4 6 11 8 9 5 5 7 6 10 10 6. Monster vs. 24ga, choral music 1.75db. level difference 14 7 15 10 8 10 6 10 11 12 10 ______________________________________________ % of "hits" in the total of 6 tests, 90 tries. 67. 50 40 33 40 40 33 33 50 83 50 L.M.: It tells me that people can easy tell level differences with pink noise, not so easily with music. Where does it "gives as many different results as there are people doing it"? You're right. I got sort of dozed off looking at it all and got carried away. It is sometimes 10/11 the same results sometimes 3/11 and a few in between. It all adds up beautifully. I wish you and Mr Nunes who "has just done that" many happy hours with the ABX and pink noise. This just shows again that you have an incomplete understanding of statistics. You state that a hit is 12 out of 15. If it is not a hit, then it is considered to be within the realm of random chance and so is not counted any differently whether it is 1 out of 15 or 11 out of 15. That does not by any stretch of the imagination "gives as many different results as there are people doing it". I said I will not discuss satististics but I'm always eager to learn. You are right. It is all the same. No differences. Or maybe there are SOME differences but they are not statistical. They are just an illusion, a puff in the wind, a knock and we are through the mirror in the Wonderland. Ludovic Mirabel |
#86
|
|||
|
|||
science vs. pseudo-science
In article a0Fdb.611941$Ho3.119231@sccrnsc03,
(ludovic mirabel) writes: (Audio Guy) wrote in message news:XWldb.603104$o%2.282900@sccrnsc02... This just shows again that you have an incomplete understanding of statistics. You state that a hit is 12 out of 15. If it is not a hit, then it is considered to be within the realm of random chance and so is not counted any differently whether it is 1 out of 15 or 11 out of 15. That does not by any stretch of the imagination "gives as many different results as there are people doing it". I said I will not discuss satististics but I'm always eager to learn. You are right. It is all the same. No differences. Or maybe there are SOME differences but they are not statistical. They are just an illusion, a puff in the wind, a knock and we are through the mirror in the Wonderland. Did you not employ statistical analysis in any of the medical DBTs you were involved in? Because it's key in analyzing the results of the test or tests. Without the statistical analysis you cannot tell which results are just random chance and which are really significant. I will agree with your point that when results get very close to shoving significance the individual in question should be retested. But without the retesting you cannot state that they did hear something, only that they may have heard something. But I have yet to see you make that qualification. |
#87
|
|||
|
|||
science vs. pseudo-science
|
#88
|
|||
|
|||
science vs. pseudo-science
|
#89
|
|||
|
|||
science vs. pseudo-science
Stewart -
Since my name has been dragged in here, and you are apparently calling me "dishonest" and a "liar," would you please cite where and when I have been "dishonest" or "lied" in a discussion of dbt testing. Of cables no less, which I have studiously avoided. If you can't find any (which will be the case) I would appreciate an apology. Harry "Stewart Pinkerton" wrote in message ... On Sun, 28 Sep 2003 17:37:55 GMT, (ludovic mirabel) wrote: (Stewart Pinkerton) wrote in message ... It tells people *exactly* what I'd like to be known, which is that you simply *refuse* to engage in honest debate, instead using every possible trick to avoid the inevitable conclusion, that you simply do not have a case to argue. Name one participant in RAHE discussions who disagrees with you on DBT matters and who does "engage in honest debate" with you. If you recall it is not Mkuller or Harry Lavo. Name one "honest" man who disagrees with you ,honestly by your lights, about testing cables. That is of course entirely my point. It is *very* obvious that those in this forum who support DBTs rely on logic and on the results of actual tests, whereas those who disagree rely on polemic, distortion and cherry-picking. It is probably ( statistically of course) true that you really believe that people normally lie to argue a point. Not perhaps in general, but it certainly seems to form a pattern in *this* instance. -- Stewart Pinkerton | Music is Art - Audio is Engineering |
#90
|
|||
|
|||
science vs. pseudo-science
(Audio Guy) wrote in message news:4dJdb.455493$Oz4.260164@rwcrnsc54...
In article a0Fdb.611941$Ho3.119231@sccrnsc03, (ludovic mirabel) writes: (Audio Guy) wrote in message news:XWldb.603104$o%2.282900@sccrnsc02... This just shows again that you have an incomplete understanding of statistics. You state that a hit is 12 out of 15. If it is not a hit, then it is considered to be within the realm of random chance and so is not counted any differently whether it is 1 out of 15 or 11 out of 15. That does not by any stretch of the imagination "gives as many different results as there are people doing it". I said I will not discuss satististics but I'm always eager to learn. You are right. It is all the same. No differences. Or maybe there are SOME differences but they are not statistical. They are just an illusion, a puff in the wind, a knock and we are through the mirror in the Wonderland. Did you not employ statistical analysis in any of the medical DBTs you were involved in? Because it's key in analyzing the results of the test or tests. Without the statistical analysis you cannot tell which results are just random chance and which are really significant. I will agree with your point that when results get very close to shoving significance the individual in question should be retested. But without the retesting you cannot state that they did hear something, only that they may have heard something. But I have yet to see you make that qualification. Dear man , I qualified thusly at least ten times in the last two years, asking why didn't the proctord pay attention to the only interesting results, namely those of the exceptional performers and rechecked them SOS. But I can't expect you to memorise my collected writings. So for your convenience: Yes they should have done it. Even though Greenhill ran not just one but six tests and only two of his subjects scored consistenly well in five of them. You don't say how many times they should have repeated it to satisfy you and Pinkerton. Twice? 3 times? Ten times like Norman Strong once suggested? So let's collaborate on an ideal design Your statistical prowess encourages me. Let us get an ABX project, you and I together, based on Sean Olive's results. (of course he did not use ABX but we respect his results, right?). To do justice to the differences in performance between trained (the best) , semitrained (in the middle - 3 times worse) and the great unwashed (us audio consumers- just like the audio students- 27 times worse) we'll get three groups going. First the random collection of audiophiles. Get them ABXed on anything reasonably comparable other than the grossly unlike loudspeakers. The result almost guaranteed: "No difference, no preference". All is for the best in this best of all possible worlds.. Just what you all would have wanted. The result is accepted without a murmur just like your predecessors in the "Stereo Review" days accepted Greenhill, Clark, Masters and so on. As long as their ABX manipulated results on cable, preamp, amp, cdplayer, dac were "They are all the same" . Electronics is wonderful Now group 3, the trained. Some get 80% correct. Panic in the ranks. This couldn't be! Repeat please. And keep repeating till they say "uncle" ie. till they are half-deaf and ready to confess that there is "No difference"- and can I go home, please? The intermediate group , the salesmen, doesn't count. They convince easy. Isn't statistics wonderful too? Ludovic Mirabel. |
#91
|
|||
|
|||
science vs. pseudo-science
(Audio Guy) wrote in message news:4dJdb.455493$Oz4.260164@rwcrnsc54...
In article a0Fdb.611941$Ho3.119231@sccrnsc03, (ludovic mirabel) writes: (Audio Guy) wrote in message news:XWldb.603104$o%2.282900@sccrnsc02... This just shows again that you have an incomplete understanding of statistics. You state that a hit is 12 out of 15. If it is not a hit, then it is considered to be within the realm of random chance and so is not counted any differently whether it is 1 out of 15 or 11 out of 15. That does not by any stretch of the imagination "gives as many different results as there are people doing it". I said I will not discuss satististics but I'm always eager to learn. You are right. It is all the same. No differences. Or maybe there are SOME differences but they are not statistical. They are just an illusion, a puff in the wind, a knock and we are through the mirror in the Wonderland. Did you not employ statistical analysis in any of the medical DBTs you were involved in? Because it's key in analyzing the results of the test or tests. Without the statistical analysis you cannot tell which results are just random chance and which are really significant. I will agree with your point that when results get very close to shoving significance the individual in question should be retested. But without the retesting you cannot state that they did hear something, only that they may have heard something. But I have yet to see you make that qualification. Dear man , I qualified thusly at least ten times in the last two years, asking why didn't the proctord pay attention to the only interesting results, namely those of the exceptional performers and rechecked them SOS. But I can't expect you to memorise my collected writings. So for your convenience: Yes they should have done it. Even though Greenhill ran not just one but six tests and only two of his subjects scored consistenly well in five of them. You don't say how many times they should have repeated it to satisfy you and Pinkerton. Twice? 3 times? Ten times like Norman Strong once suggested? So let's collaborate on an ideal design Your statistical prowess encourages me. Let us get an ABX project, you and I together, based on Sean Olive's results. (of course he did not use ABX but we respect his results, right?). To do justice to the differences in performance between trained (the best) , semitrained (in the middle - 3 times worse) and the great unwashed (us audio consumers- just like the audio students- 27 times worse) we'll get three groups going. First the random collection of audiophiles. Get them ABXed on anything reasonably comparable other than the grossly unlike loudspeakers. The result almost guaranteed: "No difference, no preference". All is for the best in this best of all possible worlds.. Just what you all would have wanted. The result is accepted without a murmur just like your predecessors in the "Stereo Review" days accepted Greenhill, Clark, Masters and so on. As long as their ABX manipulated results on cable, preamp, amp, cdplayer, dac were "They are all the same" . Electronics is wonderful Now group 3, the trained. Some get 80% correct. Panic in the ranks. This couldn't be! Repeat please. And keep repeating till they say "uncle" ie. till they are half-deaf and ready to confess that there is "No difference"- and can I go home, please? The intermediate group , the salesmen, doesn't count. They convince easy. Isn't statistics wonderful too? Ludovic Mirabel. |
#92
|
|||
|
|||
science vs. pseudo-science
|
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Simple science question | Audio Opinions | |||
rec.audio.opinion, isn't exactly rocket science | Audio Opinions |