This study highlights the importance of national factors in explaining the wide variations in child malnutrition that exists among population groups in the three Central Asian Republics under study, and as an avenue through which the investigation of the processes by which other factors influencing child malnutrition can be executed. This study has been guided by the proposition that the effects of socio-economic, health and the environmental factors vary by province of residence. The suggested theory underlying this proposition is the structural violence theory, which stems from the evidence of differences in malnutrition levels among provinces of residence within the three countries under study. Against the background of this theory, the study explores and compares the effects, on child malnutrition of sixteen covariates of malnutrition within these countries. The study assessed, first, the nutritional status of children below 3 years using the three anthropometric indices, weight-for-age, height-for-age and weight-for-height. Secondly, it examined the confounding factors that influence nutritional status in the Central Asian Republics. The DHS data for Kazakhstan, Kyrgyzstan and Uzbekistan were used. The data of the countries were combined in order to create a more usable dataset for multivariate analyses. This descriptive analysis was done on all relevant variables in order to get a better understanding of the dynamics involved in the data. A binary logistic regression analysis estimating models using the stepwise forward method (based on Wald statistics) on the pooled dataset was carried out controlling socio-economic, health and environmental characteristics. Separate analysis was carried out to study the odds of being underweight, and stunted among children in the study population. Analyses for the odds of being wasted were not performed because the percentage of wasted children was too low. The main results indicate that province of residence within a country, country of residence, number of people in a household, household wealth, birth weight, age of child, knowledge of ORT, maternal education, number of children under five years, and source of drinking water, are strong predictors of child nutritional status in the three Central Asian Republics. Furthermore, it has revealed that chronic malnutrition (stunting), which is long-term undernutrition, is most prevalent in all three countries but at varied levels. An unexpected finding is that fully vaccinated children were highly likely to be malnourished compared to children who were partially vaccinated. Since stunting appears in early ages in Central Asian children many of the children in this study may be stunted before they receive all the recommended vaccinations. This could also be influenced by the poor nutritional status of the mother who cannot provide nutritious breast milk. Another unexpected finding was that breast-feeding especially in children more than six months old had a strong negative association with stunting and underweight. Women who breastfeed longer may be doing so because they lack the resources and nutritional knowledge to provide their children with adequate nutrition. Another explanation for the observed adverse association of breastfeeding and nutritional status beyond infancy is that of reverse causality, where mothers continue to breastfeed children who appear small for their age. In summary, the results from both the descriptive and binary logistic regression analysis are similar in terms of the explanatory variables and the statistical significance in the models. The study recommends more area-specific (province) policies for the development of nutritional intervention programmes including closer attention to childcare and mother care programmes. / Thesis (PhD (Sociology))--University of Pretoria, 2007. / Sociology / PhD (Sociology) / Unrestricted