Re: [Phys-l] Another uncertainties question...

[Jason Alferness <alf@u.washington.edu>]

John Denker wrote:
> On 01/25/2008 10:36 AM, Jason Alferness wrote:
>
>   
> > Is there a standard accepted way to deal with non-detects?  
> >     
>
> The general rule is "say what you mean, and mean what you say".
>
> The rest is details.  The details are not particularly standardized.
>
>   

I don't quite understand what is meant by this...

>   
> >  or to use the threshold detect as the 
> > uncertainty for either.
> >     
>
> I assume the point is that the result is not "plus or minus"
> anything, because it can't extend to the "minus" side.
>   

If you mean by this my point... Not quite... see below...
> The fancy way to report an asymmetric distribution is to
> give the nominal value "plus" something "minus" something
> else, as in 10 (plus 3, minus1) ... or in this case
> 0 (plus 3, minus 0).  I've seen that in plenty of official
> documents, so it must be fairly widely understood, at
> least among experts.  However in this application, it
> may be fancier than you need.
>
> It is always acceptable to spell it out in plan English:
>   none detected (threshold 3)
>   none detected (sensitivity 3)
>   below threshold (threshold 4, sensitivity 2)
>
> or whatever...............
>   

You'll receive no argument from me that there are better ways about 
reporting than this... Nonetheless, my question is not really how it 
should be reported, for which I'll concede that they are better and 
clearer ways... For example the above asymmetric distribution is far 
more meaningful to me, as is the "plain english" option... but my point 
is that this is not the form of the data received.  So what to do?

The report simply of "Less than (value)" seems to be quite common... 
I've seen it in contamination testing, medical reports, blood tests and 
a number of other things.  So I don't think it's a quirk of one 
particularly stupid lab.  I'm not asking really how it _should_ have 
been reported, I guess, but rather if this is what one has (and it seems 
this is what one is likely to get in many places) and is forced to work 
with it, what's the most intelligent thing to do with it, especially 
with respect to crunching an "average" out of some stack of these 
numbers (and in some cases lack of numbers).  I could work much better 
with the above asymmetric and plain english descriptions, but it seems 
fairly unlikely that we'll get them.

I don't know if there's a clear answer even, but I figured others must 
have run into this before, so was hoping somebody could tell me what 
they did...

Thanks for your thoughts, all -

Best - and best wishes for the weekend,

Jason



> _______________________________________________
> Forum for Physics Educators
> Phys-l@carnot.physics.buffalo.edu
> https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l
>   


--

-------------------------------------------------------------------------------------------
A person, who is nice to you, but rude to the waiter/waitress, is not a 
nice person. (This is very important. Pay attention - It never fails.) 
Dave Barry from "FIFTEEN THINGS THAT IT TOOK ME OVER 50 YEARS TO LEARN"

Jason Alferness
University of Washington
Room B256B, Physics & Astronomy Building
Campus Box 351560
Seattle, Washington.  98195-1560


Phone: (206) 221-2974
FAX : (206) 685-0635
email: alf@phys.washington.edu