Abstract:
This study was done to find the factors related to malnutrition and the relationship between response variable and explanatory variables. I have used Demographic Health Survey 2000 secondary data set for my analysis. 314 families and 641chiidren (aged 3-59 months) were selected from the Demographic Health Survey 2000 main data set. This sample included 333 males and 308 females. Age and anthropometrical measurements (weight and height) were used to assess the nutritional status of children. The nutritional status can be categorised in to four groups (Normal. underweight. Wasting and Stunting) using U.S. National Centre for Health Statistics (NCHS). and the Centre for Disease Control (CDC) reference value of the standard population. Using the nutritional status of the child. the binary response variable was created and 30 explanatory variables were selected from main DHS data set which explained the health condition of child. Basically, in my analysis the association of response variable and explanatory variables were checked using cross - tabulation and Chi - squared test. Binary Logistic Regression models were fitted according to the theories of Generalized Linear Model and checked the significance of explanatory variables. Model significance was also checked using Goodness of Fit test. The final models of binary logistic regression revealed that malnutrition of children was negatively related with income score of family, birth weight, birth interval (age gap in months)and race(Moor). It is also revealed that malnutrition of children was positively associated with the total number of the children in the family (Parity) and sex(females)
