Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Evaluation of censored contamination data

I G Vlachonikolis1, F H Marriott

  • 1Department of Mathematical Sciences, University of Technology, Loughborough, Leicestershire, UK.

Food Additives and Contaminants
|September 1, 1995
PubMed
Summary

Accurate laboratory analysis of low-level contamination requires careful statistical modeling. This study evaluates methods for fitting data distributions with values below the detection limit, improving contamination assessment.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

The inflammatory bowel disease questionnaire: a review of its national validation studies.

Inflammatory bowel diseases·2004
Same author

ZD1839, a novel, oral epidermal growth factor receptor-tyrosine kinase inhibitor, as salvage treatment in patients with advanced non-small cell lung cancer. Experience from a single center participating in a compassionate use program.

Lung cancer (Amsterdam, Netherlands)·2003
Same author

Urinary N-telopeptide levels in multiple myeloma patients, correlation with Tc-99m-sestaMIBI scintigraphy and other biochemical markers of disease activity.

Hematological oncology·2003
Same author

HLA phenotypes as promoters of cardiovascular remodelling in subjects with arterial hypertension.

Journal of human hypertension·2003
Same author

Association of the HLA antigens with early atheromatosis in subjects with type 2 diabetes mellitus.

International angiology : a journal of the International Union of Angiology·2003
Same author

Intra-articular morphine and postoperative analgesia after knee arthroscopy.

The Knee·2002

Area of Science:

  • Environmental Science
  • Analytical Chemistry
  • Statistics

Background:

  • Low-level contamination measurements often yield skewed data.
  • Values below the detection limit are reported as non-detects, complicating distribution analysis.

Purpose of the Study:

  • To discuss and evaluate methods for fitting statistical distributions to skewed laboratory data with values below a specified detection limit.
  • To compare parameter estimates derived from complete data versus those based on assumed detection limits.

Main Methods:

  • Exploration of statistical distribution fitting techniques for data with censored values (below detection limit).
  • Evaluation of methods using laboratory results from specialized equipment with very low detection limits.
  • Comparison of estimation strategies: using complete data versus incorporating assumed detection limits.

Main Results:

  • Assumptions about the lower tail of the distribution are necessary when dealing with non-detects.
  • Specialized equipment allows for checking these distributional assumptions.
  • Estimates derived from complete data can differ from those based on assumed detection limits.

Conclusions:

  • Appropriate statistical methods are crucial for accurately characterizing low-level contamination.
  • The use of advanced equipment with lower detection limits enhances the reliability of contamination data analysis.
  • Careful consideration of data below detection limits improves the precision of contamination estimates.

Related Experiment Videos