Simplified And Alternative Approach To Modified Ridit Analysis
Matthew Chukwuma Michael
Department of Mathematics and Statistics, Faculty of Science, University of Delta, Agbor State, Delta State, Nigeria
Oyeka, I. C. A
Department of Statistics, Faculty of Physical Sciences, Nnamdi Azkiwe University, Awka, Anambra State, Nigeria.
Eriobu Nkiru Obioma
Department of Statistics, Faculty of Physical Sciences, Nnamdi Azkiwe University, Awka, Anambra State, Nigeria.
Ehiwario, Jacob Chinedum
Department of Mathematics and Statistics, Faculty of Science, University of Delta, Agbor State, Delta State, Nigeria
Keywords: Simplified, Modified, Alternative, Ridit Analysis, natural ordering, qualitative variable
Abstract
This paper proposes a simplified and alternative modified method of Ridit Analysis for qualitatively ordered groups of sample data. The proposed method enables the estimation of the probabilities that a randomly selected subject from one sampled population is at a more serious (higher, worse), the same (equal) or less serious (lower, better) condition or state than a randomly selected subject from another sampled population. Test statistics were developed which enables the testing not only of the null hypothesis, if desired, that subjects at a given condition of seriousness in one population is at a more serious (higher, worse) condition than subjects at all levels or conditions of seriousness in the other population, but also the null hypothesis that subjects from one population are on the average in a more serious (higher, worse) condition than subjects from the other population. The method was illustrated with some sample data and the proposed test statistic was shown to be relatively more efficient and hence likely to be more powerful than the corresponding earlier modified test statistic.