Re: [Phys-l] Basic statistics

[Jack Uretsky <jlu@hep.anl.gov>]

Hi all-
	I find all these generalities unhelpful,  because I don't know 
what Ludwik is talking about.  My tiny mind can only start from specific 
problems, and, on a good day, I can generalize from these.

	Yesterday I attended a seminar by a member of an experimental 
collaboration that was measuring the mass and width of the W-particle.
He compared the most recent result with several others, from different 
experiments, each with a mass value and an uncertainty.  How would you set 
about determining (1) a world average and uncertainty, and (2) the 
consistency of various experiments.
	For guidance as to current practice, see the Particle Data Group 
booklet, or their web-site.
					Regards,
						Jack



On Thu, 9 Nov 2006, ludwik kowalski wrote:

> In section 8.3, at
>
>   http://www.av8n.com/physics/uncertainty.htm#sec-samples
>
> John Denker wrote: "Nowadays experts generally avoid using the term
> ?precision", and concentrate instead on quantifying the uncertainty."
>
> 1) I think that the words "uncertainty," like words "errance," or
> "estimator", would be good replacements for what is usually called
> "standard deviation of mean values."
>
> 2) Do you people agree that two similar phrases: (a) "standard
> deviation" (referring to a set of data in a sample) and (b)  "standard
> derivation of  mean values" (referring to data on <x> from many sample)
> often confuse students?
>
> 3) John suspects that many people use the word sample as a reference to
> one number. Statisticians, however, decided to define  samples as sets
> of n numbers. That is why John suggested to replace the word "sample"
> with the word "cluster." Will this suggestion be implemented? I do not
> think so.
>
> 4) I suggest that, at least on this list, we start using the word
> "errance" instead of "standard deviation of the mean". By errance, E,
> we mean s/sqr(n), where s is a standard deviation from a representative
> sample of size  n.
>
> 4) What should one do to make this happen? I do not know. In fact, I do
> not believe that  this will happen. The proposition would have a much
> higher chance of being implemented if it came from a famous scientist,
> for example, from a winner of Nobel Prize in pedagogy, or from an
> author of a widely used laboratory manual for students.
> Ludwik
> _______________________________________________
> Forum for Physics Educators
> Phys-l@carnot.physics.buffalo.edu
> https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l

--
"Trust me.  I have a lot of experience at this."
		General Custer's unremembered message to his men,
		just before leading them into the Little Big Horn Valley