Klarinet Archive - Posting 000147.txt from 2005/03

From: Tom Flavel <tom@-----.net>
Subj: Re: [kl] German sound
Date: Thu, 3 Mar 2005 17:10:00 -0500

On 03/03/2005 16:56:23, David Blumberg wrote:
> Ok, so what would be considered to be an acceptable % of correct vs
> incorrect answers? If there are no differences than it would be just
> "guessing" so what is considered "guessing" % wise?

If I remember correctly, the t-test will tell you if the difference
between two groups of data is significant or not. I'd guess two
appropiate groups for this would be 1. peole who know nothing of
clarinets, as a control to give your "guessing" data - I propose using
humans rather than random numbers, as you'd be using humans for the
second group: 2. the people in the contest.

I have absolutley no idea as to how useful this would be, but I hope
that helps from my limited knowledge of statistics...

Regards,

--
Tom

-------------------------------------------------------------------
Klarinet is a service of Woodwind.Org, Inc. http://www.woodwind.org

   
     Copyright © Woodwind.Org, Inc. All Rights Reserved    Privacy Policy    Contact charette@woodwind.org