This study uses an explanatory model of the dimensions of leadership and emotional intelligence according to the methods used in particular teaching environments (universities and other educational institutions). The effect of different kinds of leadership on emotional intelligence dimensions is also established using an explanatory model. A total of 954 teachers participated in this cross-sectional study, teaching in 137 different schools/universities. The instruments used for the data collection were the Multifactor Leadership Questionnaire (MLQ-5) and the Trait Meta Mood Scale (TMMS-24). Data analysis was performed with the software IBM AMOS 23.0. (International Business Machines Corporation, Armonk, NY, USA) using multi-group analysis and structural equations. Results showed that the structural equation model had a good fit. Transformational leadership depends mainly on intellectual stimulation in university teachers, whereas intrinsic motivation is more relevant at the lower educational levels. In relation to transactional leadership, contingency reward has a greater regression weight in non-university education, whereas passive leadership is governed more by passive exception in university teachers. There was a positive and direct relationship between levels of emotional intelligence and transformational leadership in non-university teachers, which reveals the need for effective understanding and management of both one's own and students' emotions in order to act effectively as a leader. Transactional leadership was negatively related to some emotional intelligence dimensions, given the relevance of obtaining power in this dimension.