Background: Though undernutrition affects academic performance, significant number of Ethiopian school children have undernutrition. To avert nutritional problems the government in collaboration with the world food program has implemented a school feeding program. However, data on the nutritional status of primary school students were scarce in the country. Therefore, this study aimed to assess the relationship between school meals with thinness and stunting among primary school students in Meket Woreda and to identify associated factors. Methods: A school-based comparative cross-sectional study was conducted among 1091 students, from April 1-27, 2015. The study participants were selected using a multi-stage stratified sampling method. A structured questionnaire was used to collect data. Data entry and analysis were done using Statistical Package for the Social Sciences (SPSS) version 20 software. Descriptive statistics and logistic regression analyses were done. Anthropometric data were analyzed using Anthro-plus software. Results: In univariate analysis, thinness was commoner in nonschool feeding program schools (37.5%) compared to school feeding program schools (27.8%) but stunting was less common (48.3% versus 58.5%). However, after adjustment for confounding, there was no difference in stunting levels, but the adjusted odds ratio for thinness in nonschool feeding program schools was 2.6 (95% CI: (1.8, 3.8)) times higher than in school feeding program schools. Other independent risk factors for thinness were: having uneducated mother, being a male and taking meal once daily. Independent risk factors for stunting were ethnicity, having uneducated mother, un-piped water supply, taking meal once daily, type of diet and being a male. Conclusion: The provision of meals seems to offer considerable protection against thinness, though not against stunting. Thus, school meal program should be scaling up into schools in food insecure areas.