Premium
Detection limits of organic contaminants in drinking water
Author(s) -
Draper William M.,
Dhoot Jagdev S.,
Dhaliwal Joginder S.,
Remoy John W.,
Perera S. Kusum,
Baumann Frank J.
Publication year - 1998
Publication title -
journal ‐ american water works association
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.466
H-Index - 74
eISSN - 1551-8833
pISSN - 0003-150X
DOI - 10.1002/j.1551-8833.1998.tb08456.x
Subject(s) - pooling , minimum description length , statistics , detection limit , computer science , mathematics , artificial intelligence
Better control of MDL determination will substantially reduce interlaboratory variation. This article examines some of the experimental variables that can contribute to the observed variability in laboratory performance. The examples provided suggest that method detection limits (MDLs) would be more uniform among laboratories if (1) uniform spike concentrations were used in MDL determination; (2) analytical methods were more uniform as to procedures, reagents, and materials; and (3) tighter guidelines were established for conducting MDL experiments and handling MDL data. The pooling of data from multiple spike levels (or any other means to increase sample size) minimizes random error in MDL determination. Improved control in MDL determination would lead to better information on laboratory capabilities, and this in turn would improve the technical basis for reporting limits, trigger levels, and water quality standards.