Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
MULTIVARIATE COVARIANCE WITH REPEATED MEASURES--A MONTE CARLO STUDY.
Abstract
A frequently used design in educational and psychological research is the measurement of a subject on thesame variable more than once. In this design, the researcher is examining the possibility of a change over time in the dependent variable due to either intervening treatments or subject development. Winer (1971) calls this a repeated measures design. Because the same subject is used for each level of the repeated factor the assumption of independence of levels is violated, and the relationship, or correlation, between levels must be taken into consideration. Several statistical techniques have been developed for analysis of repeated measures data. Of most prominent use is the mixed model univariate analysis of variance. This method assumes that a non-zero correlation exists between the repeated measurements and further, that the correlation is constant between all measurements. In educational and psychological research the second phase of the univariate assumption is frequently violated. The more common pattern of correlation is a descending relationship between time periods, known as a simplex pattern. When this pattern occurs, the correlation between time periods one and two will be greater than the correlation of periods one and three.
Subject Area
Educational tests & measurements|Educational evaluation
Recommended Citation
NEWMAN, DIANNA LEE, "MULTIVARIATE COVARIANCE WITH REPEATED MEASURES--A MONTE CARLO STUDY." (1978). ETD collection for University of Nebraska-Lincoln. AAI7809161.
https://digitalcommons.unl.edu/dissertations/AAI7809161