Background : Childhood malnutrition is a major concern in developing countries due to high morbidity and mortality rate. To estimate the prevalence of malnutrition among school children, Kassala State, A cross-sectional study was conducted randomly among 2638 children. Height for age and body mass index for age z scores were calculated using WHO Anthro Plus software as indicators of stunting and thinness respectively. A multinomial logistic regression model was used to determine the associated factors. Results: The overall prevalence of stunted and thinness was 22.1% and 32.3% respectively. Stunting was significantly higher among children in age group 13-15 years old (OR 3.788; 95% CI: 2.802- 5.123), boys (OR 1.394; 95% CI: 1.114-1.743) and children from rural areas (OR 2.406; 95% CI: 1.768- 3.274). Likewise thinness was significantly higher among boys (OR 1.717; 95% CI: 1.436- 2.052), Rural residence (OR 1.885; 95% CI: 1.487- 2.389), child whose families depend on unsafe source of drinking water (OR 2.205; 95% CI: 1.484- 3.276). Moreover, children who skipped meal during the school day or bring their food from their houses OR 2.744; 95% CI: 2.131- 3.532, 1.752(95% CI: 1.441- 2.129) respectively). Maternal education and occupation, in addition to family size and child ordered had no effect on the prevalence of malnutrition. Conclusions: In the present study, different forms of malnutrition was documented and affected by child age, gender, residence, source of drinking water. Therefore, more attention is needed to focus on nutrition status, improving the school environment, maternal education, and regular health assessment.