Re: [Phys-l] experimentation 101

[ludwik kowalski <kowalskil@mail.montclair.edu>]

Last summer I measured track densities on 40 CR-39 surfaces. And had  
three outliers (12,000, 200,000, and 15,000 tr/cm^2). The track  
densities on remaining 37 surfaces  were between 3 and 31 (mean=16 tr/ 
cm^2 and std.dev=8 tr/cm^2). Three days ago I showed the draft of our  
paper to a statistician. She said that she is against ignoring the  
outliers. Data, she said, should be respected. That is a tricky issue;  
on one hand this and on the other hand that. I am rejecting the  
outliers, blaming a possibility of migrating alpha radioactive  
contamination. My attempts to confirm this hypothesis were not  
successful.

Ludwik






On Jul 14, 2010, at 3:53 PM, Hugh Haskell wrote:

> At 15:42 -0400 07/14/2010, chuck britton wrote:
> > Some would say - and I think that I would agree - the value of 492.5
> > cm must have resulted from a fairly serious experimental blunder -
> > since it differs by half a meter from any other value.
> > The other values all fall into a 'significantly' smaller range than  
> > this.
> I wouldn't throw that one out without further investigation. It may
> be the "outlier" that is the nearest one to correct. What if there
> was some ambiguity in where the hall began and/or where it ended? It
> is possible, although not likely, that the outlier measured it
> correctly and the others got it wrong, so the criteria that each used
> for measurement need to be investigated.
>
> I believe that it was Asimov who once said that the most important
> thing a scientist can say is "That's interesting . . ."
>
> Hugh
> --
>
> Hugh Haskell
> mailto:hugh@ieer.org
> mailto:haskellh@verizon.net
>
> It isn't easy being green.
>
> 		--Kermit Lagrenouille
> _______________________________________________
> Forum for Physics Educators
> Phys-l@carnot.physics.buffalo.edu
> https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l

Ludwik

http://csam.montclair.edu/~kowalski/life/intro.html