Back to search

North Dakota State University

Power Analysis to Determine the Importance of Covariance Structure Choice in Mixed Model Repeated Measures Anova

Abstract

dc:description.abstract

Repeated measures experiments involve multiple subjects with measurements taken on each subject over time. We used SAS to conduct a simulation study to see how different methods of analysis perform under various simulation parameters (e.g. sample size, autocorrelation, repeated measures). Our goals were to: compare the multivariate analysis of variance method using PROC GLM to the mixed model method using PROC MIXED in terms of power, determine how choosing the incorrect covariance structure for mixed model analysis affects power, and identify sample sizes needed to produce adequate power of 90 percent under different scenarios. The findings support using the mixed model method over the multivariate method because power is generally higher when using the mixed model method. Simpler covariance structures may be preferred when testing the within-subjects effect to obtain high power. Additionally, these results can be used as a guide for determining the sample size needed for adequate power.

Degree

thesis:*
Grantor dc:publisher
North Dakota State University
Year dc:date.issued
2017

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • King, Taylor J.

Rights

dc:rights
Statement dc:rights
  • NDSU Policy 190.6.2

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/10365/28656

Chain of custody

source
Harvested from
North Dakota State University
Base URL
library.ndsu.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

King, Taylor J.. Power Analysis to Determine the Importance of Covariance Structure Choice in Mixed Model Repeated Measures Anova. North Dakota State University, 2017. https://hdl.handle.net/10365/28656