Modelling the factors contributing to under-five mortality in Somalia

Abstract

Under-five mortality rate is the key indicator of both child well-being and coverage of child survival interventions factoring social and economic development. This is in line with Millennium Development Goal 4 (MDG 4) projects reduction of under-five mortality rate by two-thirds by 2015. Somalia is one of the countries with the highest mortality rate in the world. This study was conducted to identify the factors contributing to under-five mortality in Somalia using discriminant analysis. The data used was from UNICEF 2006 Multiple Indicator Cluster Survey (MICS). Using discriminant analysis, a stepwise procedure was used to identify only the significant variables which were ranked according to Wilk’s Lambda values. The canonical discriminant function coefficients (unstandardized and standardized) were also calculated for independent variables. Based on this procedure, children ever born, source of drinking water, age of the mother, current marital status of the mother and region of residence were found to be significantly contributing to under-five mortality in Somalia. The classification accuracy of the model was 73.8%. Therefore, the discriminant function constructed was adequate and thus can be used to classify a child into any of the two groups, dead or alive, based on significant factors that are contributing to under-five mortality.