A DISCOURSE ON CHILD MALNUTRITION: ANTHROPOMETRY, EMERGENT THEMES, QUALITY CONTROL MAXIMS, AND CLIMATIC AND ECONOMIC DETERMINANTS

Abstract

Malnutrition is a detrimental and significant plight for young children, responsible for 45% of all deaths among children worldwide. The aim of my dissertation is to assess the history of the science of anthropometry, synthesize the cumulative findings within the contemporary child malnutrition literature, dispute certain quality control maxims of anthropometric child-health surveys, and quantify the responsible latent factors of child malnutrition. These efforts are in service of a better characterization of malnutrition, a more reliable estimate of how many children are malnourished, and a better understanding of the geographical distribution and dynamic stochastic characteristics of malnutrition. It is essential to better understand malnutrition and its causes to suggest appropriate corrective policy. This dissertation consists of four principal essays, each from a unique conceptual perspective. The first essay is a historical and epistemological perspective of the science of anthropometry. I contextualize the legacy of child malnutrition efforts, including the link between eugenics and contemporary notions of “normal” child growth, the institutional power-struggle for child growth chart superiority, the obfuscated distinction between growth references and standards of growth, and the consequences of universal standards that do not reflect observable populations. The second essay is a systematic review of the literature, the largest of its kind to date. I synthesize 184 disaggregate empirical studies of the determinants of child malnutrition in Africa published since 1990. I find numerous opportunities for development within this corpus, in particular opportunities to enrich the scope, scale, and quantification of the field. The third essay is an analytical perspective on the quality control mechanisms applied to anthropometric surveys. I challenge the practice of rejecting datasets based on overlarge z-score standard deviation values and offer an alternative approach. The fourth essay is an econometric empirical analysis in Kenya and Nigeria of child malnutrition determinants. I use spatial Bayesian kriging and four-level random intercept hierarchical logit models to show the spatial heterogeneity of malnutrition prevalence, and to quantify various socio-economic and climatic determinants of child malnutrition. I find significant spatial and hierarchical relationships and determinants, which can move malnutrition rates by over 50%.