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

Tests for departure from normal in laboratory data.

R G Rossing, A Hillis, D Y Chou

    American Journal of Clinical Pathology
    |May 1, 1985
    PubMed
    Summary

    Evaluating laboratory data for Gaussian distribution is crucial. While the Kolmogorov-Smirnov test has limitations with rounding and skewness, kurtosis and skewness coefficients offer valuable insights for specific alternatives.

    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

    Unintended changes in cognition, mood, and behavior arising from cell-based interventions for neurological conditions: ethical challenges.

    The American journal of bioethics : AJOB·2009
    Same author

    Cell-based interventions for neurologic conditions: ethical challenges for early human trials.

    Neurology·2008
    Same author

    Diffusion- and perfusion-weighted brain magnetic resonance imaging in patients with neurologic complications after cardiac surgery.

    Archives of neurology·2001
    Same author

    Patient Tan revisited: a case of atypical global aphasia?

    Journal of the history of the neurosciences·2001
    Same author

    In cirrhotic patients variceal bleeding is more frequent in the evening and correlates with severity of liver disease.

    Hepato-gastroenterology·1999
    Same author

    Three years of experience with prospective randomized effectiveness studies.

    Controlled clinical trials·1998

    Area of Science:

    • Biostatistics
    • Statistical analysis of laboratory data

    Background:

    • Laboratory data often assumed to follow a Gaussian (normal) distribution for analysis.
    • Accurate statistical methods are essential for validating this assumption.
    • Previous methods have shown limitations in practical applications.

    Purpose of the Study:

    • To compare various statistical tests for assessing the Gaussian distribution assumption in laboratory data.
    • To identify reliable methods for evaluating data normality.
    • To address limitations of existing normality tests.

    Main Methods:

    • Comparison of the traditional Kolmogorov-Smirnov test.
    • Evaluation of a modified Kolmogorov-Smirnov test.
    • Assessment of kurtosis and skewness coefficients.
    • Use of simulated data, including positively skewed datasets.

    Main Results:

    • The traditional Kolmogorov-Smirnov test is sensitive to data rounding, reducing its reliability.
    • A modified Kolmogorov-Smirnov test demonstrated low sensitivity to positive skewness.
    • Kurtosis and skewness coefficients proved valuable for detecting deviations from normality under specific conditions.

    Conclusions:

    • No single "omnibus" test is currently satisfactory for all scenarios of normality testing in laboratory data.
    • The choice of normality test depends on data characteristics, such as the degree of rounding and skewness.
    • Kurtosis and skewness offer complementary information for assessing Gaussian assumptions.

    Related Experiment Videos