Background: Undernutrition is a problem of public health significant in developing countries. This study examined the factors associated with stunting in children in Dodoma Region in Tanzania.

Methods: Multistage stratified sampling procedure was used to select respondents. Height and weight were measured and nutritional status of children was determined. Stunting, underweight, and wasting in children were defined using these criteria, which were compared to WHO norms and standards. Logistic regression models were used to establish relationships between stunting and multiple variables.

Results: The overall prevalence of stunting was 33%. Mzula village had the highest prevalence of severely underweight children at 3.6%. There were no significant differences in weight and height among children between villages. Important factors associated with stunting in a univariate logistic regression were mother’s education, material used to build house, age of the child and distance to water source. In the multivariate analysis distance to water source and mother’s education, maternal education, and child’s age were found to independently predict stunting. The univariate logistic regression picked the father’s education, age of the mother and distance to water source as main determinants of being underweight.

Conclusion: The current study stresses the importance of implementing region specific and context-relevant treatments to prevent malnutrition in this and other similar contexts in Tanzania.