Early childhood developmental delays contribute to high burden of social support, medical and educational needs for families. While child survival rates have increased due to emphasis on child survival, among these survivor groups, there is an increasing population of children with developmental delays. In developing countries, the prevalence of development delays among survivors is not known because of lack of routine screening. Also, the sociol-economic determinants of developmental delays are poorly understood. Therefore, a cross-sectional study was conducted to determine the prevalence of early childhood delays, identify the socioeconomic factors that are associated with early childhood delays and understand how socioeconomic factors predict the development of fine motor, gross motor, receptive language, expressive language in Ugunja Sub-location, western Kenya. The study involved randomly selected 449 children under the age of five years with their respective caregivers. Structured questionnaire was used to collect data. The prevalence of global developmental delay defined as delay in two or more domains was 15.96%. Prevalence of respective domains of development such as expressive language, receptive language, fine motor, gross motor and socio-emotional; were 10.56%, 12.36%, 12.13%, 13.48% and 13.71%, respectively. Households with higher number of children were at risk of global delay with Odds Ratio (OR) = 1.2, 95%CI (1.08-1.36), P = 0.0001. However, the maternal education was with OR = 0.9 [CI; 0.8, 0.9], P = 0.002; number of Antenatal Care (ANC) visits was with OR = 0.6 [CI; 0.4, 0.8], P = 0.002; and Higher SES OR = 0.44 [CI; 0.29, 0.70] P < 0.001 were beneficial to global delay. The probability of global delay in lower Socio Economic Status (SES) was 0.2 [95% CI; 0.14, 0.26], medium SES, 0.10 [95% CI; 0.07, 0.13] and high SES was 0.05 [95% CI; 0.02, 0.08]. Years in school seem to provide good prediction of child delay, while maternal education was shown to have statistically significant effects over a wide range of outcomes including receptive language, fine motor, gross motor, and global delay. Generally, children whose mothers had under 8 years of school had probability of delay. In conclusion, this study reports high prevalence of developmental delay in this population and further points out that improved ANC visits and maternal education can alter the developmental delay outcome, in particular low SES group should be targeted with interventions.