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

Variable error: variance-covariance heterogeneity, block size and type L error rates.

M E Gessaroli1, R W Schutz

  • 1University of British Columbia.

Journal of Motor Behavior
|March 1, 1983
PubMed
Summary

Blocked data analysis using variance equalization (VE) was studied for repeated measures ANOVA. Results show VE

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

Absolute error.

Journal of motor behavior·2013
Same author

Absolute error.

Journal of motor behavior·2013
Same author

Statistical power for the two-factor repeated measures ANOVA.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2000
Same author

Analysis of proximal femur DXA scans in growing children: comparisons of different protocols for cross-sectional 8-month and 7-year longitudinal data.

Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research·2000
Same author

Lifestyle determinants of bone mineral: a comparison between prepubertal Asian- and Caucasian-Canadian boys and girls.

Calcified tissue international·2000
Same author

Augmented trochanteric bone mineral density after modified physical education classes: a randomized school-based exercise intervention study in prepubescent and early pubescent children.

The Journal of pediatrics·2000

Area of Science:

  • Statistics
  • Psychometrics
  • Experimental Design

Background:

  • Repeated measures ANOVA assumes compound symmetry, which is often violated.
  • Violation of homogeneity of covariance can inflate Type I error rates.
  • Blocked data analysis, specifically variance equalization (VE), is a method to address these violations.

Purpose of the Study:

  • To investigate how blocked data (VE) affects the Type I error rate of repeated measures ANOVA when the compound symmetry assumption is violated.
  • To examine the influence of varying block sizes and covariance heterogeneity on statistical assumptions.
  • To assess the robustness of repeated measures ANOVA under different data blocking strategies.

Main Methods:

  • Monte Carlo simulations were employed to generate data from populations with violated covariance assumptions.
  • Data were blocked using variance equalization (VE) across different block sizes.
  • Simulated data were analyzed using repeated measures ANOVA, and Type I error rates were calculated.
  • The study involved 22,000 simulated experiments across various homogeneity conditions and data sets.

Main Results:

  • Type I error rates increased with greater heterogeneity in variance-covariance matrices for raw (unblocked) data.
  • The impact of within-matrix heterogeneity on Type I error rates with VE data was inconclusive.
  • Block size influenced the probability of significant interactions, but the relationship was inconsistent.
  • No inflation of Type I errors occurred when covariances within a matrix were homogeneous, irrespective of group differences.

Conclusions:

  • While raw data analysis is sensitive to covariance heterogeneity, the effect of VE on Type I error rates under such conditions requires further investigation.
  • Block size in VE analysis impacts interaction significance, but the precise nature of this relationship remains unclear.
  • Homogeneous within-matrix covariances ensure reliable ANOVA results, even with differing group variance-covariance structures.

Related Experiment Videos