Re: what good is "percentage error"?

["Donald E. Simanek" <dsimanek@eagle.lhup.edu>]

On Mon, 12 May 1997, brian whatcott wrote:

> A facet of the distaste for such measures is the word 'error' apparently.
> Though this word figures in the general run of books of experimental statistics,
> such as
>
> Data Reduction & Error Analysis For The Physical Sciences,
>  Bevington, McGraw-Hill
> The Statistical Analysis of Experimental data
>    Mandel, Dover
> Experiments in Physical Chemistry    
>  Wilson, Newcombe, Denaro, Rickett    Pergamon
>
> -it apparently is a taboo word?

Your bibliography could be expanded manyfold. In the literature of
statistical analysis of data, both in science and engineering, the word
"error" has a specialized meaning distinct from the colloquial meaning of
"blunder". "Error" is indeed synonymous with the mathematican's term
"experimental uncertainty". One would, however, suspect that authors of
high school and freshman college lab manuals had never read these books,
when they confuse experimental discrepancy with experimental uncertainty.

Is this any different, or any harder to deal with, than the fact that in
physics we use "work" with a technical meaning far different from the
colloqual meaning?

	-- Donald

......................................................................
Dr. Donald E. Simanek                            Office: 717-893-2079
Prof. of Physics                    Internet: dsimanek@eagle.lhup.edu
Lock Haven University, Lock Haven, PA. 17745          CIS: 73147,2166 
Home page: http://www.lhup.edu/~dsimanek            FAX: 717-893-2047
......................................................................