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

Type I Error Rates For A One Factor Within-Subjects Design With Missing Values.

Miguel A Padilla1, James Algina

  • 1Educational Psychology, University of Florida.

Journal of Modern Applied Statistical Methods : JMASM
|July 18, 2006
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

The Perceived Internalized Sexual Stigma Measure (PrISSM): A New Sexual Stigma Measure for Lesbian, Gay, and Bisexual Individuals.

Behavioral sciences (Basel, Switzerland)·2025
Same author

The association between the multidimensional evaluation of attitudes toward security recommendations and the intention of following those recommendations.

Acta psychologica·2025
Same author

Obtaining a Bayesian Estimate of Coefficient Alpha Using a Posterior Normal Distribution.

Educational and psychological measurement·2025
Same author

Efficacy of the social-emotional learning foundations curriculum for kindergarten and first grade students at risk for emotional and behavioral disorders.

Journal of school psychology·2021
Same author

Comparison of model- and design-based approaches to detect the treatment effect and covariate by treatment interactions in three-level models for multisite cluster-randomized trials.

Behavior research methods·2018
Same author

A Comparison of Composite Reliability Estimators: Coefficient Omega Confidence Intervals in the Current Literature.

Educational and psychological measurement·2018

For educational research with missing data, the Kenward-Roger statistic in SAS PROC MIXED is superior for hypothesis testing. It effectively controls Type I errors in within-subject ANOVA when data are incomplete.

Area of Science:

  • Educational Research
  • Statistical Methods
  • Psychometrics

Background:

  • Missing data frequently occur in educational research, complicating statistical analyses.
  • Maximum likelihood estimation in SAS PROC MIXED is a viable technique for handling incomplete datasets.
  • The optimal test statistic for hypothesis testing with missing data in PROC MIXED remains unclear.

Purpose of the Study:

  • To compare the performance of Hotelling-Lawley-McKeon and Kenward-Roger omnibus test statistics.
  • To evaluate these statistics for means in single-factor within-subject ANOVA with missing data.
  • To identify which test statistic better controls Type I error rates under missing data conditions.

Main Methods:

  • Utilized SAS PROC MIXED for statistical analysis, a widely accessible software.

Related Experiment Videos

  • Employed maximum likelihood estimation to derive model parameters from data with missing values.
  • Compared the Hotelling-Lawley-McKeon and Kenward-Roger test statistics in a within-subject ANOVA context.

    • Main Results:

    • The Kenward-Roger statistic demonstrated superior performance compared to the Hotelling-Lawley-McKeon statistic.
    • The Kenward-Roger statistic maintained Type I error rates closer to the nominal alpha level.
    • This suggests better control over false positives when analyzing within-subject designs with missing data.

    Conclusions:

    • The Kenward-Roger statistic is recommended for hypothesis testing in single-factor within-subject ANOVA when educational data contain missing values.
    • Its ability to preserve accurate Type I error rates makes it a reliable choice for researchers.
    • This finding enhances the utility of SAS PROC MIXED for robust analysis of incomplete educational datasets.