Trimmed Means in Split-Plot Repeated Measurement Designs.Report as inadecuate




Trimmed Means in Split-Plot Repeated Measurement Designs. - Download this document for free, or read online. Document in PDF available to download.





The Welch-James (WJ) and Improved General Approximation (IGA) tests for the within-subjects main and interaction effects in a split-plot repeated measurement design were investigated when least squares estimates and robust estimates based on trimmed means were used. Variables manipulated in the Monte Carlo study included the degree of multivariate normality, degree of departure from the assumption of multisample sphericity, total sample size, degree of sample size imbalance, and number of levels of within-subjects factor. Consistent with previous research, the WJ and IGA procedures based on least squares estimates were not always robust to violations of the multisample sphericity assumption when the data were obtained from multivariate nonnormal distributions. Adoption of trimmed mean estimators resulted in dramatic improvements in the Type I error performance of the WJ procedure. These results suggest that it is possible to obtain tests of within-subjects effects in split-plot designs that are robust to the combined effects of assumption violations. (Contains 6 tables and 21 references.) (Author/SLD)

Descriptors: Foreign Countries, Least Squares Statistics, Monte Carlo Methods, Research Design, Sample Size











Author: Lix, Lisa M.; And Others

Source: https://eric.ed.gov/?q=a&ft=on&ff1=dtySince_1992&pg=12387&id=ED410298







Related documents