Undernutrition remains a significant obstacle to children under five years old in Kenya. One in four children under five are stunted, meaning that they have height-for-age Z-scores (HAZ) less than minus two standard deviations below the median of a reference height-for-age standard. This paper employs machine learning techniques to identify significant socioeconomic, demographic, cultural, climatic and environmental indicators of child stunting from the Kenya Demographic and Health Survey (KDHS) 2014 dataset. Consequently, multivariate logistic regression is employed to determine and quantify the likelihood of any significant indicators in explaining stunting incidence. All in all, this paper suggests that a combination of a priori determination and machine learning techniques would be useful in elucidating significant indicators of child undernutrition that are in critical need of, and can respond well to, prioritized policy interventions.
Presented in Session 3. Population, Development, & the Environment; Data & Methods; Applied Demography