Robust estimation of between and within laboratory standard deviation with measurement results below the detection limit

  • Steffen UhligEmail author
Announcements and Reports


In method validation studies, the question arises how to deal with <LOD results in order to ensure reliable precision estimates. It is proposed to elaborate strategies for dealing with <LOD results on the basis of the Q/Hampel method, a particularly robust statistical method. Two different strategies are presented and assessed: censoring <LOD results and setting them equal to LOD/2. The two strategies are then examined in the light of the reliability of the precision estimates obtained on the basis of a comprehensive simulation study. The first method exhibits better precision, while the second method displays better trueness. However, the two methods are only satisfactory as long as the percentage of <LOD results is less than 25 %. For higher percentages of <LOD results, more sophisticated methods should be applied.


LOD <LOD results Method validation Censoring Robust statistical methods Q method Q/Hampel Reproducibility Repeatability 


Conflict of interest

The author declares that there are no conflicts of interest.


  1. Colson B (2015) The Untold History of Algorithm A.
  2. Müller CH, Uhlig S (2001) Estimation of variance components with high breakdown point and high efficiency. Biometrika 88:353–366CrossRefGoogle Scholar
  3. Rousseeuw PJ, Croux C (1993) Alternatives to the median absolute deviation. J Am Statist Assoc 88:1273–1283CrossRefGoogle Scholar
  4. Uhlig S (1997) Robuste Schätzung in Kovarianzmodellen mit Anwendung im Qualitätsmanagement und in Finanzmärkten. Habilitationsschrift, Freie Universität Berlin, Berlin. Available at Scholar
  5. Uhlig S, Colson B (2015) Q/Hampel*: A new robust method for interlaboratory studies with high percentages of “less than” values. (in preparation).

Copyright information

© Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL) 2015

Authors and Affiliations

  1. 1.QuoData GmbHDresdenGermany

Personalised recommendations