Related Experiment Videos
Experimental evaluation of different precision criteria applicable to microbiological counting methods
Bertrand Lombard1, Marie Cornu, Cecile Lahellec
1Agence Française de Sécurité Sanitaire des Aliments-Laboratoire d'Etudes et de Recherches sur la Qualité des Aliments et sur les Procédés Agro-Alimentaires, 23 Avenue du Général De Gaulle, 94 706 Maisons-Alfort, France. b.lombard@afssa.fr
Journal of AOAC International
|July 9, 2005
Summary
This study compares methods for measuring variability in microbiological testing. Robust median-based estimators for reproducibility standard deviations showed consistent results with classical mean-based methods when enumerating Listeria monocytogenes in food.
Area of Science:
- Microbiology
- Analytical Chemistry
- Food Safety
Background:
- Microbiological counting measurements exhibit dispersion due to repeatability and reproducibility.
- Interlaboratory studies are crucial for characterizing measurement variability.
- Accurate enumeration of pathogens like Listeria monocytogenes in food is vital for public health.
Purpose of the Study:
- To evaluate robust median-based estimators for reproducibility standard deviations.
- To compare these robust estimators with classical mean-based estimators.
- To assess the impact of log10 transformation on enumeration data analysis.
Main Methods:
- Conducted an interlaboratory study using a standardized reference method for Listeria monocytogenes enumeration.
- Applied two types of robust estimators based on the median.
- Compared results with classical estimators based on the mean.
- Investigated the effect of log10 transformation on enumeration data.
Main Results:
- Robust median-based estimators and classical mean-based estimators yielded consistent results for reproducibility standard deviations in most cases.
- The study provided experimental validation for the chosen statistical approaches.
- The necessity and impact of the log10 transformation were questioned.
Conclusions:
- Median-based robust estimators are a reliable alternative to mean-based estimators for assessing reproducibility in microbiological testing.
- The findings support the consistent application of statistical methods in interlaboratory studies for food safety analysis.
- Further investigation into data transformation methods may refine microbiological measurement analysis.