Abstract Background Uganda is currently host to more than 1·2 million refugees from neighbouring countries. Women and children make up more than 54%. Food is distributed to address the nutritional needs of the refugees. Recently, unconditional cash transfers have also been provided to some refugee households to help in diet diversification. It is unclear whether cash transfers offer better nutritional outcomes than food rations. This study compared the nutritional status of children aged 6–59 months between households receiving cash transfers and those receiving dry food rations in Rwamwanja refugee settlement. Methods This cross-sectional study was done in May, 2017, and included 497 children aged 6–59 months (247 [50%] boys and 250 [50%] girls) from 316 households. Structured questionnaires and key informant interviews were conducted among caregivers and focal people from different organisations in the refugee settlement. Nutritional status was assessed using anthropometric measurement. ENA for SMART version 13 software was used to convert nutritional data into Z-score indices of weight-for-age (underweight), height-for-age (stunting), and weight-for-height (wasting), which are less than −2SD of the WHO Growth Standard chart. Data were analysed using Stata version 13 to obtain odds ratio (OR) and prevalence rate ratio (PRR). Findings The prevalence of stunting among children 6–59 months was 46·7% (95% CI 40·4–53·2) in households receiving cash and 49·4% (44·0–54·8) in households receiving dry food rations. The prevalence of underweight was 19·6% (95% CI 15·0–25·1) in households receiving cash and 25·3% (21·0–23.0) in households receiving food. The prevalence of wasting was 2·1% (95% CI 0·9–4·9) in households receiving cash and 3·5% (2·0–6·1) in households receiving food. Prevalence of stunting among recipients of dry food was two times higher in the 12–23 month age group (adjusted PRR 2·37, 95% CI 1·24–4·54) and two times higher in the 24–35 month age group (2·30, 1·17–4·51) than in the 6–11 month age group. Children from households with female caretakers were less likely to be underweight than those with male caretakers (adjusted PRR 0·47, 95% CI 0·02–0·99). Among recipients of cash transfer, children who were not immunised for measles vaccine were five times more likely to be wasted than children who were immunised (adjusted OR 4·50, 95% CI 1·08–18·83). Interpretation Prevalence of stunting was high in the refugee settlement, and nutritional status of children in households receiving cash was not better than those receiving dry food rations. Cash transfer was preferred by most households. From this study, we recommend promotion of cash transfers in refugee settlements in Uganda. Funding None.