Emotional Intelligence (EI) is a theme that is gaining more and more space in the corporate world, especially with regard to its evaluation among professionals who occupy leadership positions. However, there are currently no properly validated instruments for this context that specifically assess the EI of leaders. Additionally, considering that this group composes a high-stake sample (i.e. samples that are naturally more focused on having their responses contaminated in order to please the evaluator), the need for an instrument that can have a more precise control of the social desirability bias arises. Therefore, the study aimed to construct and seek initial evidence of validity of the Evaluation of Emotional Intelligence of Leaders (AIEL) scale, a measure that aims to assess four emotional intelligence skills based on the forced choice method. Studies were conducted to search for evidence of contente validity and internal structure through the use of item agreement indices, exploratory factor analysis, confirmatory factor analysis and analyses based on the Thurstonian Item Response Theory (TRI-T). The final version of the instrument was answered by a sample of 215 leaders. After all the analyses performed, 36 items were selected (sociability = 11, well-being = 6, selfcontrol = 10 and emotionality = 9) with factor loadings ranging from 0.387 to 0.902 and an average value of 0.64. The model also presented a good fit. The accuracy of factorial scores was high in all factors, ranging from 0.97 to 0.98. Despite a good internal structure and possibility of use in scientific research, it is worth highlighting some limitations of the AIEL: the sample used was small and there was a high discrepancy of the mean values of social desirability among the items of the triplet blocks. It is suggested to create new items with greater balancing and in quadruple format. Additionally, it is necessary to apply it to larger and more diverse samples, with respondents with different demographic characteristics, to carry out studies about their standardization. It is also suggested to conduct validity studies based on relations with external measures. Finally, it is expected that the present study will contribute to the development of EI measures in the forced choice format specific to the organizational context, thus contributing to the work of the organizational psychologist and the advancement of the Brazilian field of organizational and work psychology, with regard to the creation of more properly validated and suitable instruments for this activity área.