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

Determining the number of clusters by sampling with replacement.

Scott Tonidandel1, John E Overall

  • 1University of Texas Health Sciences Center at Houston, USA. sctonidandel@davidson.edu

Psychological Methods
|May 13, 2004
PubMed
Summary

This study validates a bootstrap procedure for hierarchical cluster analysis. The method accurately identifies latent populations, improving with larger resampled data sets.

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

Examining Linguistic Differences in Electronic Health Records for Diverse Patients With Diabetes: Natural Language Processing Analysis.

JMIR medical informatics·2024
Same author

Authorship agreements benefit researchers and research culture.

Nature human behaviour·2023
Same author

How Can Big Data Science Transform the Psychological Sciences?

The Spanish journal of psychology·2020
Same author

From alpha to omega and beyond! A look at the past, present, and (possible) future of psychometric soundness in the Journal of Applied Psychology.

The Journal of applied psychology·2020
Same author

From Outcome to Process Focus: Fostering a More Robust Psychological Science Through Registered Reports and Results-Blind Reviewing.

Perspectives on psychological science : a journal of the Association for Psychological Science·2018
Same author

The effects of social contact on cocaine intake in female rats.

Drug and alcohol dependence·2017

Area of Science:

  • Statistics
  • Data Science
  • Computational Statistics

Background:

  • Hierarchical cluster analysis is a common method for grouping data.
  • Determining the optimal number of clusters (latent populations) is a challenge.
  • Existing stopping rules may lack robustness across diverse datasets.

Purpose of the Study:

  • To evaluate the effectiveness of a split-sample replication criterion as a bootstrap stopping rule for hierarchical cluster analysis.
  • To assess the bootstrap procedure's accuracy in identifying the true number of latent populations.
  • To investigate the influence of dataset characteristics (overlap, sample size) on the procedure's performance.

Main Methods:

  • Applying a split-sample replication criterion (Overall & Magee, 1992) as a bootstrap stopping rule.

Related Experiment Videos

  • Generating multiple datasets via sampling with replacement from a primary simulated dataset.
  • Testing the procedure across varying numbers of true latent populations, degrees of overlap, and sample sizes.

    • Main Results:

    • The bootstrap procedure significantly enhanced the accuracy of identifying the true number of latent populations.
    • Accuracy improved consistently across different combinations of population numbers, overlap, and sample sizes.
    • Increasing the relative size of resampled datasets further boosted identification accuracy.

    Conclusions:

    • The bootstrap stopping rule is a valid and effective method for determining the number of latent populations in hierarchical cluster analysis.
    • The procedure demonstrates robustness across various data conditions and is recommended for practical application.
    • A computer program is available to facilitate the implementation of this bootstrap stopping rule.