I don't know if this has been posted, I read several of the responses but my eyes glazed over.
In regards to the clock specifically, I think this lends itself to a nice blind (not double blind as has been so oft spoken of here) test, not necessarily random, because we won’t have a real control group, but well read on if you have the interest.
Here is the question, does the CLC make an identifiable difference in a listeners perception of music.
Note it does not ask what kind of difference, only if one exists, and if it can be correctly identified.
Second, more specifically, can those active in Audiogon forums, who have already purchased a CLC, and believe they have made a difference in there perception of music, accurately determine when a clock is or is not present. (In the building or listening room, whatever the consensus among those who believe in it feel is needed to free oneself from the clocks effect)
Note, for simplicity’s sake, and because everyone here is so damn committed to the truth, justice, and the audiophile way we rely on the honor system for these tests, both in carrying out the trial (IE no communication with acquaintances, no peaking shaking, trying to "guess" which is which etcetera) and the reporting of findings. Furthermore it is expected that subjects will go out of there way to maintain the trials validity, blinding, and reporting (self censure and the like)
I would hope for about 10 participants who fit the description above.
The test subject asks an acquaintance to place the CLC and another object of equal size and weight (TBD by people in the know about the clock) into two identical brown lunch bags lined with paper towels, folding the tops and single stapling while the subject is not in the room. the acquaintance then leaves the room and the subject enters, places each bag into yet another set of larger identical bag, folds the top, staples it and then marks one with an T (for tails) and the other with H (for heads). These bags are stored off site together, in a safe place where they will not be damaged, opened by friends/family that kind of thing (car maybe?)
The subject then for conducts 13 listening sessions of 2 hours in length. Before each session a coin is flipped and the corresponding bag is brought in from the storage area to the listening room. the listener listens for one hour to whatever music they prefer, on there home system (yes these will be the three major uncontrolled variables) they then bring the bag back to storage and bring the other bag into the listening room for an hour long listen of the same musical selection. It should be encouraged for this study that a specific piece (or set of pieces) from a specific genre that are agreed upon in the forums as adequate are listened to at the end of the first hour and beginning of the second hour for each and every listening session. The listener than makes judgment as to which session contained the CLC. Judgment should be made at the earliest after the first 15 minutes of listening hour 2, and certainly by the conclusion of the session, they will then post there decision for that day on a dedicated Audiogon thread. (a decision is coded as Order H/T or T/H and the Guess as to which bag contains CLC H or T) at the end of reporting for 9 sessions the listener may decide if they wish that they now know which end is up, so to speak and make a distinction, this provides for the notion that the effects may be subtle and cumulative, thus really knowing takes time and uncertainty in the first sessions does not accurately represent the matter being tested. Doing so however means that the remainder of the 4 listening tests will not take place, and they will be marked as the subject’s decision.
At the end of 13 tests (or 9 if the subject has made there absolute decision) the coding (IE which bag had the CLC) is revealed, posted on line, and then a score is computed, for subjects who went the distance it is an average of correct to not correct, we will call correct 1 and incorrect 0. For those with the 9 trials we compute two statistics, one is for the final decision 1 or 0, and another for overall decision (we impute the final 4 trials as all correct or incorrect based on the subjects past decision on trial 9).
The null hypothesis is that subjects will not score significantly different from .5 (IE 50/50 chance). Furthermore we are interested here (since it is difficult for a subject to remove themselves of the bias imparted from a previous listening tests H/T determination) in weather they can predict which bag is which, A score closer to 1 indicates accurate prediction, a score closer to 0 indicates inaccurate prediction. We expect that when averaged as a group the direction of biases due to previous listening would cancel each other, and the group mean would accurately reflect both the ability to hear a difference, and to make the correct judgment based on that distinction. IE if the group scores 0.5 than we conclude that no difference exists, and the CLC is a load of BS. If the group scores a >0.5 (p<0.05, or whatever threshold you like, but this is the most widely used one) we conclude that a difference exists, and that it can be correctly attributed to the actual product. If the opposite occurs than a difference exists, but that the effects thought to be associated with the CLC are actually associated with its absence.
Additionally there is a better way to do this, which is for the subject to leave the room after double bagging and shuffling the bags order, and have the friend re-enter to mark the bottom of the bags with an H or T. in the previous description it would be the friend who flipped the coin once at hour one, and again at hour two. The bag would be brought in and placed in a covered oblique container, and the subject would not be allowed to be near the fiend while all of this was happening. At the beginning of hour two the subject removes themselves to a secluded area, while the friend takes the bag back to the storage area, flips a coin and then takes whichever bag back with them to be placed in the room again. This complicates reporting as a friend would have to keep track of the actual H or T, and then would report it separately to the forum after all sessions were complete and the listener had made each decision. Of course, then the reveal and the analysis, the coding for which would be modified appropriately.
Cheers,
Windzilla
In regards to the clock specifically, I think this lends itself to a nice blind (not double blind as has been so oft spoken of here) test, not necessarily random, because we won’t have a real control group, but well read on if you have the interest.
Here is the question, does the CLC make an identifiable difference in a listeners perception of music.
Note it does not ask what kind of difference, only if one exists, and if it can be correctly identified.
Second, more specifically, can those active in Audiogon forums, who have already purchased a CLC, and believe they have made a difference in there perception of music, accurately determine when a clock is or is not present. (In the building or listening room, whatever the consensus among those who believe in it feel is needed to free oneself from the clocks effect)
Note, for simplicity’s sake, and because everyone here is so damn committed to the truth, justice, and the audiophile way we rely on the honor system for these tests, both in carrying out the trial (IE no communication with acquaintances, no peaking shaking, trying to "guess" which is which etcetera) and the reporting of findings. Furthermore it is expected that subjects will go out of there way to maintain the trials validity, blinding, and reporting (self censure and the like)
I would hope for about 10 participants who fit the description above.
The test subject asks an acquaintance to place the CLC and another object of equal size and weight (TBD by people in the know about the clock) into two identical brown lunch bags lined with paper towels, folding the tops and single stapling while the subject is not in the room. the acquaintance then leaves the room and the subject enters, places each bag into yet another set of larger identical bag, folds the top, staples it and then marks one with an T (for tails) and the other with H (for heads). These bags are stored off site together, in a safe place where they will not be damaged, opened by friends/family that kind of thing (car maybe?)
The subject then for conducts 13 listening sessions of 2 hours in length. Before each session a coin is flipped and the corresponding bag is brought in from the storage area to the listening room. the listener listens for one hour to whatever music they prefer, on there home system (yes these will be the three major uncontrolled variables) they then bring the bag back to storage and bring the other bag into the listening room for an hour long listen of the same musical selection. It should be encouraged for this study that a specific piece (or set of pieces) from a specific genre that are agreed upon in the forums as adequate are listened to at the end of the first hour and beginning of the second hour for each and every listening session. The listener than makes judgment as to which session contained the CLC. Judgment should be made at the earliest after the first 15 minutes of listening hour 2, and certainly by the conclusion of the session, they will then post there decision for that day on a dedicated Audiogon thread. (a decision is coded as Order H/T or T/H and the Guess as to which bag contains CLC H or T) at the end of reporting for 9 sessions the listener may decide if they wish that they now know which end is up, so to speak and make a distinction, this provides for the notion that the effects may be subtle and cumulative, thus really knowing takes time and uncertainty in the first sessions does not accurately represent the matter being tested. Doing so however means that the remainder of the 4 listening tests will not take place, and they will be marked as the subject’s decision.
At the end of 13 tests (or 9 if the subject has made there absolute decision) the coding (IE which bag had the CLC) is revealed, posted on line, and then a score is computed, for subjects who went the distance it is an average of correct to not correct, we will call correct 1 and incorrect 0. For those with the 9 trials we compute two statistics, one is for the final decision 1 or 0, and another for overall decision (we impute the final 4 trials as all correct or incorrect based on the subjects past decision on trial 9).
The null hypothesis is that subjects will not score significantly different from .5 (IE 50/50 chance). Furthermore we are interested here (since it is difficult for a subject to remove themselves of the bias imparted from a previous listening tests H/T determination) in weather they can predict which bag is which, A score closer to 1 indicates accurate prediction, a score closer to 0 indicates inaccurate prediction. We expect that when averaged as a group the direction of biases due to previous listening would cancel each other, and the group mean would accurately reflect both the ability to hear a difference, and to make the correct judgment based on that distinction. IE if the group scores 0.5 than we conclude that no difference exists, and the CLC is a load of BS. If the group scores a >0.5 (p<0.05, or whatever threshold you like, but this is the most widely used one) we conclude that a difference exists, and that it can be correctly attributed to the actual product. If the opposite occurs than a difference exists, but that the effects thought to be associated with the CLC are actually associated with its absence.
Additionally there is a better way to do this, which is for the subject to leave the room after double bagging and shuffling the bags order, and have the friend re-enter to mark the bottom of the bags with an H or T. in the previous description it would be the friend who flipped the coin once at hour one, and again at hour two. The bag would be brought in and placed in a covered oblique container, and the subject would not be allowed to be near the fiend while all of this was happening. At the beginning of hour two the subject removes themselves to a secluded area, while the friend takes the bag back to the storage area, flips a coin and then takes whichever bag back with them to be placed in the room again. This complicates reporting as a friend would have to keep track of the actual H or T, and then would report it separately to the forum after all sessions were complete and the listener had made each decision. Of course, then the reveal and the analysis, the coding for which would be modified appropriately.
Cheers,
Windzilla