Klarinet Archive - Posting 000804.txt from 2003/08

From: ormondtoby@-----.net (Ormondtoby Montoya)
Subj: [kl] 'Measuring' music (was: klarinet Digest 28)
Date: Thu, 28 Aug 2003 12:12:04 -0400

> For example, if I ask 100 people to each play
> ten different clarinets and then I ask them,
> "Which clarinet has the least fuzzy, least
> buzzy sound?", can I tabulate their answers
> and declare that I have achieved an
> "objective" standard for fuzzy & buzzy?

David Blumberg replied:

> That depends on the levels of the players.
> Is it 100 Professional Players, Professors, or
> are they Marching Band Students....... ;)

I promised to leave this topic alone, but my intent was to point out
that while a large sample size reduces statistical uncertainty, it does
not solve the problem of: "What are you measuring? Are you measuring
the audience's subjective response, or are you measuring an objective,
quantifiable aspect of the sound itself?"

There's nothing wrong with statistical analysis of subjective topics.
The Gallup Poll and Nielsen surveys do it every day, and people pay big
money for the results. But large sample sizes do not transform a
person's personal opinions about (say) clothing styles or TV preferences
into "objective" statements.

....I really will leave this topic alone now.

---------------------------------------------------------------------
Klarinet is supported by Woodwind.Org, http://www.woodwind.org/