Journal Published Online: 15 July 2024
Volume 52, Issue 5

The Joint Consequence of Detection Limit Censoring and Detection Limit Determination Uncertainty

CODEN: JTEVAB

Abstract

Method detection limits estimated from specialized data studies are frequently used as a basis for left censoring reported data. Estimated method detection limits are frequently impacted by the failure to fully mimic all measurement system characteristics represented in process quality data collection. This can result in use of method detection limit estimates in the censoring process that are likely to be biased on the small side. A large set of product quality data analytically provides the substantial combined impact of detection limit censoring in conjunction with common precision and accuracy issues found in classical detection limit estimation processes. A methodology for estimation of practical method detection limits directly from uncensored quality data rather than through a formal detection limit study is provided. The utility and improved accuracy and stability of such practical detection limit estimates versus classically derived detection limits is demonstrated. Guidance is provided with respect to how to approach detection limit censoring whether one chooses to apply it or not to their reported data.

Author Information

Bzik, Thomas J.
StatsOnTheGo, Inc., Macungie, PA, USA
Ketkar, Suhas N.
Elemental Scientific, Omaha, NE, USA
Pages: 9
Price: $25.00
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: JTE20230754
ISSN: 0090-3973
DOI: 10.1520/JTE20230754