Klarinet Archive - Posting 000769.txt from 1997/09

From: "Diane Karius, Ph.D." <dikarius@-----.edu>
Subj: Re: Let us settle this mouthpiece question
Date: Tue, 16 Sep 1997 03:02:29 -0400

I can work the stats on this one - but you'll need more mouthpieces
to start (and their starting measurements). 10/group wouldn't be a
bad place to start.

> I believe it would be valuable (and fun) to try to settle this question and
> propose the following single blinded study:
>
> A high level clarinet technician who has the expertise to accurately measure
> mouthpieces needs to volunteer his/her time to perform measurements.
>
> A supplier will need to lend 10 mouthpieces for the study.
>
> The mouthpieces will initially be sent to our technician, who will take an
> average of several measurements of each mouthpiece.
>
> The mouthpieces are then sent to me (no measurements attached).
>
> My daughter, Allegra, whose only reward will be name recognition in the
> clarinet world early in her clarinet career (she just started clarinet one
> month ago) will randomly choose two mouthpieces. She will pull a silk swab
> through one mouthpiece and a cotton swab through another 5000 times
> (equivalent to swabbing more than once daily for 10 years). - the
> mouthpieces, of course need an identifying mark.
>
> I will then send the mouthpieces back to the technician for remeasurement
> without informing him/her of which 2 mouthpieces were swabbed.
>
> The numbers will then be sent to a statistician (list member or perhaps
> spouse) who will come up with p values for whether differences are
> significant or not.
>
> The results will then be presented with great fanfare to the list (and
> perhaps published in The Clarinet?
>
> What does everyone think?
>
> Needed:
> Volunteer to temporarily donate 10 mouthpieces, identical brand, hard rubber.
> Volunteer clarinet technician to donate time and expertise.
> If one of the supply houses would be willing to supply 2 free cotton swabs
> and 2 silk swabs, it would be greatly appreciated.
> Volunteer statistician to crunch the numbers (and also to provide input on
> study design - eg., is 10 a good number?)
> Input (serious - no jokes) from teachers/perfomers on list concerning other
> aspects of study design, eg., is it necessary to dip the mouthpiece in water
> between each swabbing to more approximate real conditions? Is this a fair 10
> year approximation? Is 10 years a reasonable goal or do people think a
> mouthpiece should last 30 years?
>
> Thanks in advance
>
> Steve Gordon, M.D.
> Beaverton, OR
>
Diane R. Karius, Ph.D.
Department of Physiology
University of Health Sciences
2105 Independence Ave.
Kansas City, MO 64124
email: dikarius@-----.EDU