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

Multiple comparisons distortions of parameter estimates.

Neal O Jeffries1

  • 1National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD 20892, USA. neal.jeffries@nih.gov

Biostatistics (Oxford, England)
|September 15, 2006
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

Mitral Valve Repair for Degenerative MR With Moderate or Less Tricuspid Regurgitation: 2-Year Outcomes From a Multicenter Echocardiographic Core Laboratory-Adjudicated Cohort.

Journal of the American College of Cardiology·2026
Same author

Prospective Study of Risk Factors for Postoperative Atrial Fibrillation After Cardiac Surgery.

The Journal of surgical research·2023
Same author

Pacemaker implantation associated with tricuspid repair in the setting of mitral valve surgery: Insights from a Cardiothoracic Surgical Trials Network randomized trial.

The Journal of thoracic and cardiovascular surgery·2023
Same author

Evaluating treatment effects in group sequential multivariate longitudinal studies with covariate adjustment.

Biometrics·2022
Same author

Concomitant Tricuspid Repair in Patients with Degenerative Mitral Regurgitation.

The New England journal of medicine·2021
Same author

Initial Invasive or Conservative Strategy for Stable Coronary Disease.

The New England journal of medicine·2020

Multiple testing can bias observed effect sizes, impacting study validation and confidence intervals. A bootstrap method adjusts for this bias in significant findings, ensuring more accurate results.

Area of Science:

  • Statistics
  • Biostatistics
  • Experimental Design

Background:

  • Multiple comparison procedures are standard in experiments with numerous variables to identify significant differences.
  • The potential bias in effect size magnitude due to multiple testing is often overlooked.
  • Biased effect sizes can compromise the validation of results and the power calculations for subsequent studies.

Purpose of the Study:

  • To investigate and quantify the bias in observed effect sizes resulting from multiple testing procedures.
  • To assess the impact of this bias on confidence intervals.
  • To introduce and validate a bootstrap approach for bias correction.

Main Methods:

  • Utilized a bootstrap approach to estimate and adjust for bias in effect sizes.

Related Experiment Videos

  • Analyzed factors influencing the degree of bias.
  • Provided a proof for the convergence of the bootstrap distribution.

    • Main Results:

    • Demonstrated that multiple testing can lead to significantly biased effect size estimates.
    • Showed that standard confidence intervals can be unreliable due to this bias.
    • The bootstrap method effectively corrects for bias in the largest observed effect sizes.

    Conclusions:

    • Researchers must consider multiple testing bias when interpreting effect sizes, especially those driving study validation.
    • The proposed bootstrap method offers a reliable way to adjust biased effect sizes and improve confidence interval accuracy.
    • Understanding factors contributing to bias can help mitigate its impact in experimental research.