Nowadays, educators pay attention to emotional intelligence which is defined as the ability to monitor and explain one's own and other's emotional experience and feelings to differentiate between them as well as applying necessary information for determining thoughts and actions. The goal of this study was to determine emotional intelligence of medical residents of Tehran University of Medical Sciences. By means of two stage cluster sampling, 98 medical residents of Tehran University of Medical Sciences were selected. Participants were asked to fill valid and reliable Persian version of Emotional Quotient inventory (EQ-i) questionnaire which had been developed due to Bar-On model. Seventy two filled-up questionnaires were returned (RR=73%). Mean EI score of all participants was 319.94 ± 32.4. Mean EI score was not significantly different between male and female also, single and married participants. EI did not differ significantly in residents in respect to their discipline. Mean responsibility subscale differ significantly between male and female participants (P=0.008). Multiple regression analysis showed that happiness subscale is a predictive factor for total EI score (B=-0.32, P=0.009). Responsibility subscale differed significantly between men and women participants and happiness subscale was a good predictor for emotional intelligence score. These factors should be considered in education of medical residents.