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

Towards a reduction in publication bias.

R G Newcombe1

  • 1Department of Medical Computing and Statistics, University of Wales College of Medicine, Cardiff.

British Medical Journal (Clinical Research Ed.)
|September 12, 1987
PubMed
Summary

Many medical studies are published with inadequate sample sizes and low statistical power, leading to false positives. Implementing a priori power analysis through peer review of study protocols can improve research quality and reduce bias.

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

Corrigendum to "Diagnostic accuracy of on-scan assessments compared to clinical assessments using a periodontal probe for detecting gingival recession: A cross-sectional study" [Journal of Dentistry Volume 166, March 2026, 106504].

Journal of dentistry·2026
Same author

Diagnostic accuracy of on-scan assessments compared to clinical assessments using a periodontal probe for detecting gingival recession: A cross-sectional study.

Journal of dentistry·2026
Same author

Prevalence of gingival recession and study of associated related factors in young UK adults.

Journal of dentistry·2018
Same author

Axillary tumour burden in women with one abnormal node on ultrasound compared to women with multiple abnormal nodes.

Clinical radiology·2018
Same author

Axillary tumour burden in women with a fine-needle aspiration/core biopsy-proven positive node on ultrasonography compared to women with a positive sentinel node.

The British journal of surgery·2017
Same author

A randomised clinical trial to determine the abrasive effect of the tongue on human enamel loss with and without a prior erosive challenge.

Journal of dentistry·2017

Area of Science:

  • Medical research methodology
  • Biostatistics
  • Scientific publishing

Background:

  • Current publication practices often favor statistically significant results, even from studies with insufficient sample size and statistical power.
  • This leads to the publication of numerous underpowered studies, increasing the risk of false positive findings and exaggerated effect sizes.

Purpose of the Study:

  • To highlight the issues arising from inadequate sample size and statistical power in medical research publication.
  • To advocate for the adoption of a priori power analysis as a criterion for study design and publication suitability.

Main Methods:

  • Discussion of the limitations of current peer review practices regarding statistical power.
  • Critique of a posteriori power analysis and endorsement of a priori power analysis.

Main Results:

  • Published reports from small, underpowered studies frequently contain false positive results.
  • These studies tend to overestimate the true effects due to publication bias towards significant findings.

Conclusions:

  • A priori power analysis, assessed during the protocol planning stage, is crucial for ensuring research validity.
  • Peer review of study protocols by funding bodies and journals is recommended to enforce adequate statistical power and improve the reliability of published medical research.

Related Experiment Videos