Ethiopia has the highest rate of infant deaths in Eastern and Southern Africa. This studyrnaddresses important issues concerning infant mortality in Ethiopia. The objective of the paper isrnto determine the impact of socioeconomic, demographic and environmental variables on infantrnmortality. The data for this study were obtained from the demographic and health survey (DHS)rnconducted in Ethiopia 2005. The results of Kaplan-Meier estimation show that most of the deathsrnoccurred in the earlier month from birth to one month and then after death declined in the laterrnmonths. The Cox proportional regression model was fitted to select the significant factorsrnaffecting infant mortality in Ethiopia. The model considered provided good fit for the data.rnBased on the result of the Cox proportional regression model, infant mortality was significantlyrnassociated with breast status, mother age at birth, mother's education, birth order, source ofrndrinking water and sex (p < 0.05). This study supports health policy initiatives to stimulate use ofrnfamily planning methods to increase birth spacing. It is hoped that, the results could be used byrnpolicy makers and programme managers in the c,h ,. i ld health sector to formulate appropriaternstrategies to improve the situation of infants in Ethiopia,