CATANOVA Analysis of Knowledge and Control Practices of Hepatitis B Virus Infection amongst Tertiary University Students

Abstract

Frequency data, having no underlying metric, are frequently encountered in real-life situations. The analysis of such data is usually difficult as nominal data are inherently less informative than quantitative data and decisions are taken erroneously using such results.
 The objective of the research was to study the significant effect of gender, faculties and interaction using categorical data in a two-way cross classification. 
 Materials and Methods. The study applied a cross sectional study with a total sample size of 434. Multistage sampling was adopted. Categorical analysis of variance (CATANOVA) technique was used for analysis. This is suitable as it uses a two-way ANOVA with quantal responses as equivalent of a three-way contingency table in which one of the classifications is treated as responses to the other two. The study considered frequency data involving response scores of students using a scale of good, fair and poor.
 Results. Numerical results revealed poor level of student’s knowledge and control practices of hepatitis B virus infection. And it is significantly (p>0.05) the same in the study Universities. Moreover, gender and faculties, as well as interaction have no significant (p>0.05) effect on student’s knowledge and control practices of hepatitis B virus infection.

Categorical Analysis of Variance on Knowledge, Compliance and Impact of Hand Hygiene among Healthcare Professionals during COVID-19 Outbreak in South-East, Nigeria

“…The categorical analysis of variance (CATANOVA) is a technique designed to identify the variation between treatments of interest to the researcher. This CATANOVA is used to solve problem in analysis of variance when the observations are nominal without any underlying metric and it was also formulated to solve erroneous analysis of nominal data by using chi-square test [26,27]. In addition, there are several methods for analyzing categorical data in which some of these methods use data transformation before proceeding to analyse the data.…”