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

Analysis of data from group-randomized trials with repeat observations on the same groups

D M Murray1, P J Hannan, R D Wolfinger

  • 1Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis 55454, USA.

Statistics in Medicine
|August 12, 1998
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

Irish SARS-CoV-2 Convalescent Serological Status of Children Following Acute Pneumonia During Ireland's First Wave.

Irish medical journal·2022
Same author

Maternal Mid-Gestation Cytokine Dysregulation in Mothers of Children with Autism Spectrum Disorder.

Journal of autism and developmental disorders·2021
Same author

Neonatal encephalopathy: Etiologies other than hypoxic-ischemic encephalopathy.

Seminars in fetal & neonatal medicine·2021
Same author

Adherence to the infant vitamin D supplementation policy in Ireland.

European journal of nutrition·2020
Same author

Listeria Meningitis in an Immunocompetent Child: Case Report and Literature Review

Irish medical journal·2019
Same author

Impact of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers on renal and mortality outcomes in people with Type 2 diabetes and proteinuria.

Diabetic medicine : a journal of the British Diabetic Association·2019

The random coefficients model accurately analyzes group-randomized trials with multiple time points. Mixed-model ANOVA can produce biased results and inflated errors with heterogeneous group trends.

Area of Science:

  • Biostatistics
  • Clinical Trials Methodology
  • Statistical Modeling

Background:

  • Group-randomized trials (GRTs) are essential for evaluating interventions in clustered populations.
  • Analyzing GRTs with more than two data collection time points presents unique statistical challenges.
  • Existing models may not adequately account for complex temporal trends within groups.

Purpose of the Study:

  • To compare the performance of mixed-model ANOVA and random coefficients models for analyzing GRTs with multiple time intervals.
  • To evaluate model accuracy in estimating intervention effects and controlling Type I error rates under varying conditions.
  • To identify the most robust statistical approach for longitudinal GRT data.

Main Methods:

  • Monte Carlo simulations were employed to assess model performance.

Related Experiment Videos

  • Two models were compared: mixed-model ANOVA and the random coefficients model.
  • Simulations varied group-specific time trends (homogeneous vs. heterogeneous) to test model sensitivity.

    • Main Results:

    • Mixed-model ANOVA provided unbiased estimates and nominal Type I error rates only when group-specific time trends were homogeneous.
    • When group-specific time trends were heterogeneous, mixed-model ANOVA yielded biased sampling variance estimates and inflated Type I error rates.
    • The random coefficients model consistently provided unbiased sampling variance estimates and nominal Type I error rates, irrespective of trend heterogeneity.

    Conclusions:

    • The random coefficients model is a more reliable method for analyzing group-randomized trials with more than two time intervals, especially when group trends vary.
    • Mixed-model ANOVA may lead to inaccurate conclusions in longitudinal GRTs with heterogeneous group-specific time trends.
    • Statistical model selection is critical for valid inference in complex longitudinal study designs.