Analysis of 3-Way Interactions in Contingency Table with One Dichotomous Response Category
Njoku O. Ama, D.K. Shangodoyin, K. Setlhare, L. Gabaitiri, K. Sediakgotla
Statistics, University of Botswana, Gaborone, Botswana

This paper is focused on the analysis of categorical data in a 2x c x l contingency table. The theoretical framework of a 2 x c design is extended to 2x c x l contingency table with provision for testing interactions among subsets of either lower or upper columns of the designated table. Chi-square statistics with appropriate degrees of freedom are derived for testing interactions in the different subsets of the table. The applicability of these tests is illustrated with real life data.

Keywords: Contingency table; Chi-square test statistics; interactions; proportions

Biography: Professor Ama had all his University education at the University of Nigeria, Nsukka, Nigeria. He has a Doctor of Philosophy degree (Experimental Design). He has worked at the University of Nigeria (1981-1995), University of Swaziland (1995-1999), University of Namibia (1999-2003) and University of Botswana.

He joined the University of Botswana as a Senior Lecturer in the Department of Statistics in June 2003. His research cover areas of Education, Public Health, Statistical modelling of contingency data, HIV/AIDS, and patent and innovations. He has to his credit 18 published journal articles in international, regional and local journals, one book, one book chapter, two published conference proceedings, 12 unpublished conference papers and 8 research reports.

Born into a Christian family, Professor Ama's greatest delight is serving the Lord and seeing the young people of this age embrace Christ as their personal Lord and Saviour and serving Him in truth and in spirit.