Jasper JP (2001): Quantitative estimates of precision for molecular isotopic measurements. RAPID
COMMUNICATIONS IN MASS SPECTROMETRY 15 (17): 1554-1557.

Bob Dias wrote:

> Dear Pier (and All),
>
> I second your concern.  After having been a part of a similar set of
> discussions at a Canadian Users Meeting a few years ago, John Jasper was
> inspired to publish a paper on experimental error and reporting in Rapid
> Communications in Mass Spectrometry (2001).  That I don't have the full
> reference at hand , speaks volumes to the current state of affairs with
> regard to reporting error in isotope measurements.  I would suggest that we
> all read that paper, take it to heart and hold to ourselves to a standard
> that explicitly states in our work what our error (stdev, whatever)
> represents...even if it takes a paragraph or two in our experimentals.
> Then we will be able to unambiguously distinguish between the reporting of
> actual error in the isotope values and the reporting of standard deviation
> between replicate analyses.  With new instrumentation arriving shortly in
> my lab, and having to teach new grad students about isotopic errors for the
> first time, we will be erring on the side of diligence.
>
> Cheers,
> Bob Dias
>
> Dr. Robert F. Dias
> Assistant Professor
> Department of Chemistry and Biochemistry
> Old Dominion University
> Norfolk VA 23529
> (757) 683-4093
>
>                       Pier De Groot
>                       <pier.de-groot@i         To:      [log in to unmask]
>                       rmm.jrc.be>              cc:
>                       Sent by: Stable          Subject: [ISOGEOCHEM] About uncertainty budgets...
>                       Isotope
>                       Geochemistry
>                       <ISOGEOCHEM@LIST
>                       .UVM.EDU>
>
>                       02/21/02 03:40
>                       AM
>                       Please respond
>                       to pier.de-groot
>
> Dear all,
>
> I like to ventilate my growing unhappy feelings about the way some stable
> isotope analysts are reporting uncertainty budgets for their results.
> Increasingly, reports contain 'very precise' results, claimed to be even
> external precision or repeatability.
> Sometimes even the external precision/repeatability is reported to be
> better
> than realistic internal MS precisions, and sometimes even better than
> precisions given by manufacturers, which values can be considered to be
> generated under very ideal conditions, and which generally are not
> considered obtainable during routine work.
> What might be the reason of this generation of 'very precise' data is
> erroneous use of statistics: 'the more measurements the better the mean
> becomes'!!!. In worst cases this also might be caused by exclusion of
> uncertainty related to preparation of samples. Or by neglecting or deleting
> some measurement values which did not fit in with the supposed precise
> results..... basically meaning cheating with the results!
>
> This reporting of unrealistic uncertainty budgets, generally in form of
> precisions or repeatabilities, is alarming, since those who do their job
> seriously and give correct uncertainty budgets, become increasingly 'the
> ones not being able to work very precisely...' while they do a proper job.
>
> It is not the intention to point towards any person or organization. But it
> is rather the intention to raise awareness of this alarming fact. We are
> moving into a wrong direction with this way of reporting is the opinion
> brought forward with this meassage.
>
> I very much like to hear your opinions, comments, proposals, discussions,
> ... on this message.
>
> Pier de Groot.
> ****************************************************************************
>
> ***************
> Dr. Pier A. de Groot
> European Commission
> Joint Research Centre
> Institute for Reference Materials and Measurements (IRMM)
> Retieseweg
> 2440 Geel
> Belgium
> Tel. +32 (0)14 571 628
> Fax +32 (0)14 571 685
> e-mail: [log in to unmask]
>
> Visit my WEB-site about my "Handbook of Stable Isotope Analytical
> Techniques" at:
> http://www.geocities.com/padegroot/index.html
> last update: 4 February, 2002.
> ****************************************************************************
>
> *********************

--
P.S.: If you like to send me e-mail attachments please
      read information below address. Thank you

********************************************************************************

Dr. Roland A. Werner
Max-Planck-Institut fuer Biogeochemie
Adresse / Street address:
Carl-Zeiss-Promenade 10, 07745 Jena, Germany
Postanschrift / Postal address:
Postfach 10 01 64, 07701 Jena, Germany
Tel.: ++49-3641-643719, -643825
Fax:  ++49-3641-643710
e-mail: [log in to unmask]

********************************************************************************

The following file extensions are not accepted by our mail server:

 *.dll|*.pif|*.vb|*.vbe|*.vbs|*.exe|*.com|*.bat|*.lnk
  (instead of Winword *.doc files please use *.rtf format !!)

  Our mailserver does not digest messages larger than about 10 MB !!
  For attachments > 10 MB a ftp server is available. Please contact
  me in advance if you need to send me a larger attachment!