Socio-economic, education, and health variables were examined to determine the primary causes of malnutrition in preschool Haitian children. A survey of 160 women and their child closest to weaning age was conducted in three regions of Haiti in January 1978. Ordinary least squares multiple regression analysis was used to determine the combined power of the socio-economic, education, and health variables and the relative power of each independent variable separately in explaining malnutrition in the sample population. Results showed that the amount of food or food money available had the greatest impact on the child's current nutritional status as, measured by weight/age. Long term malnutrition, as measured by height, age, was most affected by education. Health variables as a group were least effective in explaining malnutrition although the number of illnesses a child had had was highly related to malnutrition. A strong interaction between weaning age and months spent in a nutrition center suggested that children who were weaned early were more likely to become severely malnourished and required longer periods of nutritional rehabilitation than children who were weaned later. Recommendations were made to incorporate the findings into current nutrition center programs in Haiti. / Ph. D.