Social and emotional problems are commonly reported after moderate to severe acquired brain injury (ABI) and pose a significant barrier to rehabilitation. However, progress in assessment of emotional skills has been limited by a lack of validated measurement approaches. This study represents the first formal psychometric evaluation of the use of the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) V2.0 as a tool for assessing skills in perceiving, using, understanding and managing emotions following ABI. The sample consisted of 82 participants aged 18–80 years in the postacute phase of recovery (2 months–7 years) after moderate to severe ABI. Participants completed the MSCEIT V2.0 and measures of cognition and mood. Sociodemographic and clinical variables were collated from participant interview and medical files. Results revealed deficits across all MSCEIT subscales (approximately 1 SD below the normative mean). Internal consistency was adequate at overall, area, and branch levels, and MSCEIT scores correlated in expected ways with key demographic, clinical, cognitive, and mood variables. MSCEIT performance was related to injury severity and clinician-rated functioning after ABI. Confirmatory factor analysis favored a 3-factor model of EI due to statistical redundancy of the Using Emotions branch. Overall, these findings suggest that the MSCEIT V2.0 is sensitive to emotion processing deficits after moderate to severe ABI, and can yield valid and reliable scores in an ABI sample. In terms of theoretical contributions, our findings support a domain-based, 3-factor approach for characterizing emotion-related abilities in brain-injured individuals.