On Improving Precision of A Repeated Measures Analysis of Variance

Abstract

This paper investigates the use of repeated measures analysis of variance (RMANOVA) in analyzing repeated measurements. We focus on how serial auto correlation of repeated measures data can be removed in order to improve the validity of decision on repeated measures since this is widely used tool of applied statistician because its users are confronted with what seems to be a myriad of decisions, even in its simple application. In this study, the higher order interactions are used as error1 and error2 in order to improve the precision of our result. Ten (10) data sets were used to justify this paper and the computation was done with the aid of Statistical Package for Social Sciences (SPSS). The test on the overall estimates was significant at 95%