Despite over a decade of recognition in popular psychology, the construct known as emotional intelligence (EI) remains elusive in many aspects of empirical research. Self-other reporting as well as multiple measurement methods were used to contribute towards the construct validity of EI. Self- and other-ratings of EI were expected to positively correlate. In addition, this correlation was expected to be moderated by the level of each partner’s EI. Four different measures of ability EI were also expected to positively correlate with one another, and the aggregation of these measures into one ability score was expected to enhance the correlation between the performance-based and self-report measures. A sample of adults (N=83) completed the self-report Trait Emotional Intelligence Questionnaire (TEIQue) as well as four measures of ability EI: the Ekman-60 faces test (Faces), the Situational Test of Emotion Management (STEM), the Situational Test of Emotional Understanding (STEU), and the Reading the Mind in the Eyes test (Eyes). A subset of (M=28) couples also rated their partner on the TEIQue. Evidence was found for strong associations between self- and partner-rated trait EI indicating construct validity for the TEIQue. The four ability measures yielded strong interrelationships among them showing convergent validity with one another. Levels of EI as well as multiple demographic variables were found to moderate the ability of partners to rate one another, again supporting the construct validity of EI. Lastly, even when all four ability measures were aggregated to form one score, no correlations emerged between trait and ability EI measures, suggesting a need for further research into the conceptualization of EI. Implications of these results for future work on trait and ability EI are discussed.