Exploratory Data Analysis of Infant Mortality in North-East Nigeria Using Multivariate Statistical Techniques

This study conducts an exploratory multivariate statistical analysis of infant mortality data across the North East geopolitical zone of Nigeria, encompassing Adamawa, Bauchi, Borno, Gombe, Taraba, and Yobe states, with the objective of identifying the key socioeconomic, nutritional, and healthcare access determinants of infant death in the region. The North East has consistently reported among the highest infant mortality rates in Nigeria, a crisis exacerbated by the humanitarian consequences of the Boko Haram insurgency and related displacement. Cross-sectional data on infant mortality rates, exclusive breastfeeding rates, immunisation coverage, diarrhoea prevalence, maternal education levels, household wealth quintile, and access to clean water are drawn from the 2021 Nigeria Demographic and Health Survey and UNICEF Multiple Indicator Cluster Survey datasets. Principal component analysis is applied to reduce the dimensionality of the predictor space and identify underlying latent factors driving infant death risk. Cluster analysis using k-means and hierarchical methods is used to group local government areas into homogeneous risk strata. Logistic regression is subsequently applied to estimate the independent effects of key identified determinants on the probability of infant death before age one. Results highlight immunisation coverage and access to oral rehydration therapy as the most actionable intervention targets across all identified risk clusters. Keywords: infant mortality, exploratory data analysis, principal component analysis, North East Nigeria, demographic health survey