Introduction: The study assessed dietary diversity score (DDS) of rural under-five children and the relationship with their nutritional status and socio-demographic characteristics of their mothers/caregivers. Methods: The study was a descriptive cross-sectional survey involving 226 mothers and their under-five children selected at random from rural communities in Imo state, Nigeria. A pre-tested questionnaire was used to obtain information on socio-demographics. Qualitative recall of the child’s food consumption during the previous 24-h was used to calculate individual dietary diversity score (DDS=representing the number of food groups, based on a scale of 12 groups) and the scores were divided into terciles low= =4, medium=5 - 8, and high= 9 - 12). Weight-for-age (WAZ), height-for-age (HAZ) and weight-for-height (WHZ) Z-scores were used to determine nutritional status of the children. Descriptive statistics, Chi square and linear regression analyses were performed. Results: Average age of the children was 4.2±0.7years, with 51.8% being males, and 48.2% females. Stunting, wasting and underweight were 11.5%, 5.4% and 2.7%, respectively. Mean DDS for all food groups was 6.04±4.18. Cereals and vegetables had higher mean values (0.78±0.29 and 0.78±0.30, respectively), while eggs had the least value (0.15±0.25). The prevalence of low, medium and high DDS (in terciles) was 73.5%, 25.2% and 1.3%, respectively. DDS was significantly associated with HAZ (?2 =10.63;p=0.03), while total family income remained significantly and positively associated with dietary diversity score (p<0.05) in the linear regression model. Conclusion: Children with low DDS were more likely to be stunted. Therefore, efforts aimed at increasing diversity in meals, both at home and in schools, are likely to benefit children at nutritional risk.