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

[John Denker <jsd@av8n.com>]

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.

> It seems not 
> quite correct to do any of the following: Ignore it, call it zero, use 
> the threshold detect value, 

OK.

>  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.

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...............