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

Attrition in longitudinal studies. How to deal with missing data.

Jos Twisk1, Wieke de Vente

  • 1Institute for Research in Extramural Medicine, Vrije Universiteit, Vd Boechorststraat 7, 1081 BT, Amsterdam, The Netherlands. jwr.twisk.emgo@med.vu.nl

Journal of Clinical Epidemiology
|April 3, 2002
PubMed
Summary

Missing data significantly impacts longitudinal analyses like MANOVA, necessitating imputation. Longitudinal imputation methods are often superior to cross-sectional ones, with multiple imputation offering more adequate standard errors.

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

CRB1-Associated Inherited Retinal Dystrophies: Prospective Natural History Study With 4 Years of Follow-Up.

Clinical & experimental ophthalmology·2026
Same author

Malignant transformation of sacrococcygeal teratoma versus presacral teratoma in Currarino syndrome: Results of 'The SCT-study'.

Journal of pediatric surgery·2026
Same author

Amateur Soccer Heading and Acute Elevations in Blood-Based p-Tau217 and S100B.

JAMA neurology·2026
Same author

Clinical validation of video-based vital sign monitoring in the intensive care unit: a prospective cohort study : Clinical validation of video-based vital sign monitoring.

Journal of clinical monitoring and computing·2026
Same author

Effect of a protein intervention during resistance training with varying training intensities on muscle outcomes in frail community-dwelling older adults: a randomized controlled trial.

The journal of nutrition, health & aging·2026
Same author

The time-varying prognostic value of stenosis and plaque burden in coronary artery disease.

European heart journal. Cardiovascular Imaging·2026

Area of Science:

  • Statistics
  • Biostatistics
  • Longitudinal Data Analysis

Background:

  • Missing data is a common challenge in longitudinal studies.
  • Statistical analyses can be biased by missing data.
  • Various imputation methods exist to handle missing data.

Purpose of the Study:

  • To assess the impact of missing data on longitudinal statistical analyses.
  • To compare the effectiveness of different data imputation methods.
  • To evaluate imputation strategies for MANOVA and GEE.

Main Methods:

  • Simulated missing data (10% and 25%) in an observational longitudinal dataset.
  • Compared cross-sectional (mean of series, hot deck, regression) and longitudinal (last value carried forward, interpolation, regression) imputation methods.

Related Experiment Videos

  • Applied multiple imputation and analyzed results using MANOVA for repeated measurements and Generalised Estimating Equations (GEE).

    • Main Results:

    • MANOVA (using listwise deletion) requires imputation for accurate results.
    • GEE analyses were robust to missing data, not requiring imputation.
    • Longitudinal imputation methods yielded estimates closer to complete data than cross-sectional methods.
    • Multiple imputation provided more adequate standard errors, reflecting uncertainty from missing values.

    Conclusions:

    • Imputation is crucial for MANOVA but not GEE. Longitudinal imputation methods are preferred over cross-sectional ones.
    • Multiple imputation, while theoretically sound, did not significantly alter point estimates but improved standard error accuracy.
    • The choice of imputation method depends on the statistical analysis and data characteristics.