Application of Categorical Data-nested Design of Knowledge & Control Practices of HBV Infection

Abstract

In real-life, most experimental data are presented in frequencies with no underlying metric probably because of some reasons such as less susceptibility to observational errors. Unfortunately, some of these data have been erroneously analyzed resulting to either type I or type II error. The significance of main factor (University) and sub-factor (Faculty) are studied using categorical data in nested classification. The CATANOVA technique used is suitable for mixed design, having some factors crossed and others nested. The study considered frequency data involving response scores of student’s knowledge and control practices of HBV infection using a scale of good, fair and poor. Numerical results revealed that the main factor, University and the sub-factor, Faculty are not significant (p>0.05) in each case. More so, there was poor level of student’s knowledge and control practices of HBV infection which were also found to be significantly (p>0.05) same in Universities.