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

The statistical basis of meta-analysis

J L Fleiss1

  • 1Division of Biostatistics, Columbia School of Public Health, New York 10032-3799.

Statistical Methods in Medical Research
|January 1, 1993
PubMed
Summary

This study presents two meta-analysis models: fixed effects and random effects. It details their application to clinical trial data for comparing experimental interventions versus controls using various effect measures.

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

ON THE METHODS AND THEORY OF CLUSTERING.

Multivariate behavioral research·2016
Same author

A simple approximation for calculating sample sizes for comparing independent proportions.

Biometrics·2015
Same author

Power law behavior of RR-interval variability in healthy middle-aged persons, patients with recent acute myocardial infarction, and patients with heart transplants.

Circulation·1996
Same author

Vagal modulation of RR intervals during head-up tilt and the infusion of isoproterenol.

The American journal of cardiology·1995
Same author

RR variability in healthy, middle-aged persons compared with patients with chronic coronary heart disease or recent acute myocardial infarction.

Circulation·1995
Same author

Exploratory or analytic meta-analysis: should we distinguish between them?

Journal of clinical epidemiology·1995

Area of Science:

  • Biostatistics
  • Clinical Epidemiology
  • Medical Research

Background:

  • Meta-analysis is a statistical method to synthesize results from multiple studies.
  • Understanding study-to-study variation is crucial for accurate meta-analysis interpretation.
  • Fixed effects and random effects models offer different approaches to account for this variation.

Purpose of the Study:

  • To present and critique two models for study-to-study variation in meta-analysis: fixed effects and random effects.
  • To illustrate the application of these models using clinical trial data.
  • To compare the utility of both models across different summary measures of treatment effects.

Main Methods:

  • The paper describes the conceptual differences between fixed effects and random effects models.
  • It details the application of both models to three common summary measures: standardized difference for means, relative risk for proportions, and odds ratio for proportions.
  • Illustrative examples using clinical trial data are provided for each model and measure.

Main Results:

  • The fixed effects model assumes studies are the entire population of interest.
  • The random effects model assumes studies are a sample from a larger population, accounting for between-study heterogeneity.
  • Both models were applied to standardized mean difference, relative risk, and odds ratio, demonstrating their practical use.

Conclusions:

  • The choice between fixed and random effects models depends on the research question and the nature of the studies included.
  • Both models provide valuable insights into treatment effects in meta-analyses of clinical trials.
  • Understanding the assumptions and applications of each model is essential for robust evidence synthesis.

Related Experiment Videos