The purpose of this study was to assess the psychometrics properties of the Emotional Intelligence Scale and assess the measurement invariance across elite (n = 367), amateur (n = 629), and non-athletes (n = 550). In total, 1,546 participants from various sports completed the emotional intelligence scale. Several competing models were compared through exploratory structural equation modeling. The analyses were performed on the whole sample before subsequent invariance testing between athletic groups. The internal consistency of the scale was tested through Omega for the total scale and relevant subscales, which indicated largely unacceptable levels of stability. Results failed to support the purported unidimensional or four-factor models proposed in the literature. However, a six-factor model provided the best fit to the data. Nonetheless, there was no evidence for weak or strong invariance suggesting that the scale may not be appropriate for use within athletic samples.