This is, of course, John's thread and he has the final word, but... I'd like to appeal to the forum moderators - can you, please, move this DBT/ABX chatter and attacks on J.C. views re. DBT to the separate thread? This one has become barely readable.
Thank you!
Last edited:
No, my argument is that one designs an experiment appropriate to what you're trying to determine rather than using mantras, catchphrases, and inappropriate "controls" to match the mantra.
Homework question: If you take a room of (say) 100 people and give them 5 presentations, each with a 50/50 chance of being correct, and we further assume totally random guessing, what is the probability that someone in the room will score 0/5? 5/5? (Remember, there's no a priori designation of who will score 0/5 or 5/5)
Bonus question: if there are a spread of abilities and 10% of the participants can nonrandomly make a choice at each presentation, with the remaining 90% guessing randomly, what percentage of guesses overall will be correct?
Bonus question: if there are a spread of abilities and 10% of the participants can nonrandomly make a choice at each presentation, with the remaining 90% guessing randomly, what percentage of guesses overall will be correct?
Homework answer: 13.48% chance that just one person will get 0/5. Same for 5/5.
95.82% chance that someone (i.e. one or more people) will get 0/5. Same for 5/5.
That is about the limit of my knowledge of statistics, so I won't attempt the bonus question!
I have a question: what is the chance that some (or most) of those scoring 5/5 will believe it is due to their skill rather than luck? I guess that is psychology rather than statistics.
95.82% chance that someone (i.e. one or more people) will get 0/5. Same for 5/5.
That is about the limit of my knowledge of statistics, so I won't attempt the bonus question!
I have a question: what is the chance that some (or most) of those scoring 5/5 will believe it is due to their skill rather than luck? I guess that is psychology rather than statistics.
I find that very fine differentiations, AT MY AGE, are more practical with long term listening. For example, the difference in sound quality of the JC-80 and the CTC Blowtorch line stages, with their associated switching and volume controls.
However, in 1978, when I did the listening test in Japan, I was 36 years old, about 1/2 my present age, and I could hear differences quicker and easier, especially 'flawed' designs not of my own making.
However, in 1978, when I did the listening test in Japan, I was 36 years old, about 1/2 my present age, and I could hear differences quicker and easier, especially 'flawed' designs not of my own making.
Yes. Also a correct answer PM'ed to me (thanks, Jay!).
There used to be a stock scam some years ago based on this notion.
100%. Especially if there's no-one to collect and verify signed scoresheets before the results are announced.
Fund managers work on the same basis. It has been shown that on average almost all of them do worse than random choice, if you look over a long enough period, but for shorter periods many seem to be either clever or 'unlucky'. I think I saw one test where, over a year, a cat did better at picking stocks than a team of fund managers.
There may be a few fund managers with genuine skill. The snag is that you would need a time machine to take advantage of this as over any finite period you can never be sure whether they are just being lucky, but if you had a time machine you would not need a fund manager!
There may be a few fund managers with genuine skill. The snag is that you would need a time machine to take advantage of this as over any finite period you can never be sure whether they are just being lucky, but if you had a time machine you would not need a fund manager!
I don't get it.
How can you have a 13.48% that just one person gets a certain score, but 95.82 that one OR MORE get that score?
jan
- Status
- Not open for further replies.