Campus Access Only

All rights reserved. This publication is intended for use solely by faculty, students, and staff of University of the Pacific. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, now known or later developed, including but not limited to photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author or the publisher.

Date of Award


Document Type


Degree Name

Doctor of Education (Ed.D.)


Graduate Studies

First Advisor

Roger L. Reimer

First Committee Member

Robert D. Morrow

Second Committee Member

Bobby R. Hopkins

Third Committee Member

Lee Fennell

Fourth Committee Member

Carolyn M. Fowle


PROBLEM: In the planning phase of developing quality educational research there are several critical decision points. Some of these decisions are made to en sure that the research will have adequate statistical power, that is, the capability to detect meaningful differences, if they do exist. In an unpublished article, K. D. Hopkins from the Laboratory of Educational Research, University of Colorado and B. R. Hopkins from the University of the Pacific commented that all theoretical and empirical options for increasing statistical power had been investigated except for the reliability of the dependent variable. In this article, they derived a method for estimating the statistical power of the analysis of variance (ANOVA) and the analysis of covariance (ANCOVA) which factored out the influence of the reliability of the dependent variable. This method involved modifying the non-centrality parameters used in estimating the power of the ANOVA and ANCOVA tests. This modification required some untested mathematical assumptions and therefore needed to be empirically verified.



Included in

Education Commons