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

Imputing cross-sectional missing data: comparison of common techniques.

Graeme Hawthorne1, Peter Elliott

  • 1Department of Psychiatry, Australian Centre for Posttraumatic Mental Health, The University of Melbourne, PO Box 5444, West Heidelberg, Melbourne, Victoria 3081, Australia.

The Australian and New Zealand Journal of Psychiatry
|July 6, 2005
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

Ultrafast Saddletronics: Current and Charge Control at the M Point.

ACS nano·2025
Same author

Combining THz and Infrared Light to Control Valley Charge and Current in Gapless Graphene.

Nano letters·2025
Same author

Evaluating limited biopsy templates for men with markedly elevated PSAs.

The Canadian journal of urology·2024
Same author

THz induced giant spin and valley currents.

Science advances·2023
Same author

Massed <i>v</i>. standard prolonged exposure therapy for PTSD in military personnel and veterans: 12-month follow-up of a non-inferiority randomised controlled trial.

Psychological medicine·2023
Same author

Cognitive Behavioral Therapy Enhances Survival in Cardiac Patients Aged Under 60: 14-YEAR OUTCOMES OF THE BEATING HEART PROBLEMS PROGRAM TRIAL.

Journal of cardiopulmonary rehabilitation and prevention·2023
Same journal

Letter to the Editor regarding 'Middle childhood profiles of social-emotional competencies and difficulties differentiate risk of health service presentations with adolescent mental disorders'.

The Australian and New Zealand journal of psychiatry·2026
Same journal

Diagnostic complexity and comorbidity: Implications for identification, outcomes, health care and involuntary treatment.

The Australian and New Zealand journal of psychiatry·2026
Same journal

Defining and measuring psychosocial disability in Australia: Assessment of national surveys and administrative datasets.

The Australian and New Zealand journal of psychiatry·2026
Same journal

Academic performance in children of mothers with perinatal depressive disorder.

The Australian and New Zealand journal of psychiatry·2026
Same journal

The Multisensory Hallucinations Schedule to assess hallucinations across multiple sensory modalities in psychosis: A pilot validation study.

The Australian and New Zealand journal of psychiatry·2026
Same journal

Student wellbeing supports in Australian and New Zealand medical schools: A post-guidance snapshot from a 14-institution survey.

The Australian and New Zealand journal of psychiatry·2026
See all related articles

For missing data in health research, person mean imputation and hot deck imputation are recommended. Listwise deletion and item mean substitution perform poorly and should be avoided.

Area of Science:

  • Biostatistics
  • Health Services Research
  • Epidemiology

Background:

  • Missing data significantly impacts clinical and public health intervention analyses.
  • Numerous procedures exist for handling missing data, but guidance on selection is limited.

Purpose of the Study:

  • To compare the performance of six common missing data procedures.
  • To provide recommendations for selecting appropriate methods based on data characteristics.

Main Methods:

  • Evaluation of listwise deletion, item mean substitution, person mean substitution (two levels), regression imputation, and hot deck imputation.
  • Analysis of a complete dataset under varying sample sizes and missing data levels.
  • Comparison based on true t-values for the entire sample.

Related Experiment Videos

Main Results:

  • Person mean substitution and hot deck imputation are superior when approximately half of scale items are present.
  • Hot deck imputation is recommended for scales with over half missing items or single-item measures.
  • Listwise deletion and item mean substitution demonstrated poor performance across scenarios.

Conclusions:

  • Person mean substitution and hot deck imputation are the preferred methods for handling missing data in health research.
  • The widespread use of less effective methods (listwise deletion, item mean substitution) has broad implications for data analysis accuracy.