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

Incorrect least-squares regression coefficients in method-comparison analysis.

P J Cornbleet, N Gochman

    Clinical Chemistry
    |March 1, 1979
    PubMed
    Summary
    This summary is machine-generated.

    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

    Natural killer cell precursor acute lymphoma/leukemia presenting in an infant.

    Archives of pathology & laboratory medicine·2001
    Same author

    High and higher sensitivity CRP.

    Clinical chemistry·2000
    Same author

    Blastic/blastoid transformation of follicular lymphoma: immunohistologic and molecular analyses of five cases.

    The American journal of surgical pathology·2000
    Same author

    Proficiency testing of therapeutic drug monitoring techniques.

    Clinical chemistry·2000
    Same author

    Aggressive natural killer-like T-cell malignancy with leukemic presentation following solid organ transplantation.

    American journal of clinical pathology·1999
    Same author

    The new millennium in laboratory automation.

    Annals of clinical and laboratory science·1999
    Same journal

    Comparison of Information-Dependent Acquisition and Sequential Window Acquisition of All Theoretical Mass Spectra for Untargeted Drug Testing on a Linear Ion Trap-Pulsing Quadrupole-Time of Flight Mass Spectrometer.

    Clinical chemistry·2026
    Same journal

    Patterns of One-Year Change in HbA1c and Continuous Glucose Monitoring (CGM) Metrics in Older Adults with Type 2 Diabetes.

    Clinical chemistry·2026
    Same journal

    TSH Pediatric Reference Intervals: Lack of CALIPER Applicability to US-Based Populations.

    Clinical chemistry·2026
    Same journal

    Rapid Detection of Hemoglobinopathy Variants Using One-Step Library Preparation and Nanopore Sequencing.

    Clinical chemistry·2026
    Same journal

    Editor's Note: Circulating Proteolytic Products of Carboxypeptidase N for Early Detection of Breast Cancer.

    Clinical chemistry·2026
    Same journal

    In Reply to Reflexing NT-proBNP for sFlt-1/PlGF Ratios That Fall into the Measurement Uncertainty for Preeclampsia Risk Classification.

    Clinical chemistry·2026
    See all related articles

    Least-squares regression can be inaccurate due to measurement errors in the x-variable or outliers. The Deming method is recommended for improved slope estimation when the independent variable has imprecision.

    Area of Science:

    • Biostatistics
    • Data Analysis
    • Regression Analysis

    Background:

    • Least-squares regression is a common method for determining the best-fit line in data analysis.
    • However, its accuracy can be compromised by violations of underlying assumptions, particularly measurement error in the independent variable and the presence of outliers.
    • These issues can lead to incorrect regression coefficients (slope and intercept).

    Purpose of the Study:

    • To compare the performance of Deming, Mandel, and Bartlett methods for estimating regression slopes when the independent variable is subject to measurement imprecision.
    • To identify the threshold at which imprecision in the independent variable significantly impacts least-squares slope estimation.
    • To provide a method for mitigating errors caused by outliers in regression analysis.

    Main Methods:

    Related Experiment Videos

    • Comparative analysis of Deming, Mandel, and Bartlett regression methods.
    • Simulation or experimental design to introduce controlled imprecision in the independent variable.
    • Statistical analysis to quantify errors in slope estimation under different conditions.
    • Outlier detection using the vertical distance from the regression line and standard error of the estimate.

    Main Results:

    • The Deming method demonstrated superior performance in estimating the known slope compared to Mandel and Bartlett methods when the independent variable had imprecision.
    • Significant errors in least-squares slope estimation were observed when the ratio of the standard deviation of a single x-value measurement to the standard deviation of the x-data set exceeded 0.2.
    • Eliminating data points with a vertical distance greater than four times the standard error of the estimate effectively reduced errors attributable to outliers.

    Conclusions:

    • The Deming method is recommended for regression analysis when the independent variable is measured with imprecision.
    • Researchers should be cautious of least-squares regression when the measurement error ratio exceeds 0.2.
    • Implementing outlier removal based on a defined threshold can improve the reliability of regression coefficients.