Home Critic's Corner

Discuss a review. Provide constructive feedback. Talk to the industry.

[ Follow Ups ] Thread: [ Display  All  Email ] [ Critic's Corner ]
[ Alert Moderator ]

You have to take the "sucker factor" into account

From Harry Brown's book "Why Most Investment Strategies Fail", a guy receives a letter from a stock forecaster he doesn't know and it says a certain stock will go up tomorrow. And sure enough it does. He gets another letter the next day and it says the sock will go down and sure enough it does. He gets a total of five correct predictions in a row and each one is right. At the end of the week, the broker says OK, I've given you five correct predictions for free, the next one will cost you $10,000. Should he pay? The scam is easy to understand. On the first day, he sends out 16 letters half of which say the stock will go up and half of which say it will go down. To the 8 he got right, the next day he sends half letters saying the following day it will go up and the other half get letters saying it will go down. You know the rest, one out of 16 will get 5 correct predictions making it look like the guy is a genius. What does this have to do with DBT audio tests? Just this, there is a certain percentage of tests JA will be right on a significant number of trials, not because he actually could hear a difference but because of the statistical probability that eventually he will. Taking 100 such tests, unless the number of trials in each one is enormous, there is a likelihood that for some of them he will get a significant number right simply by random chance. And what about the other participants? How many participants, how many trials. JA doesn't give you that data. It's easy to be tricked, you have to know ALL of the details of the test before you can say whether or not his conclusion is fair and based on actually hearing a difference and not on random chance. It's not that statistics lie, it's just that you have to know the entire context to be able to say what they mean. In his case, they may mean nothing. We just don't know.

This post is made possible by the generous support of people like you and our sponsors:
  Schiit Audio  

Follow Ups Full Thread
Follow Ups
  • You have to take the "sucker factor" into account - Soundmind 14:06:20 04/28/06 (0)
You can not post to an archived thread.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]