engleski

On the discriminant validity evidence of the ability- based emotional intelligence measures

The study aimed to provide a set of validity data on the performance-based instruments intended to measure three branches of emotional intelligence (EI). The validity evidence will be demonstrated on the composite measures designed to assess perception, understanding and regulation of emotions, as generally recognized EI dimensions within ability- based EI models. In view of the previously obtained findings on relatively low discriminant validity of the instruments when administered on early adolescent samples, the present study focuses on the age related differences in the latent relationships among the EI dimensions, as well as between EI and selected cognitive ability and personality measures. The analyses are based on the sets of both cross- sectional and panel data collected in two time points with adolescent samples of primary school students of both genders (N=523). The structure of relationships among the specified EI, cognitive reasoning and personality latent variables was examined under the latent state-trait framework (LST ; Steyer at al., 2015). Multi-group structural equation analyses showed that the structure of observed relationships examined was contingent upon the cohorts or age of students. However, the results could not be interpreted in terms of differentiation of abilities hypothesis, at least when speaking of emotion understanding latent dimension. The implications of the results in terms of attaining more accurate estimates of the relationships among the constructs of interest and thus more convincing convergent and discriminant validity evidence will be discussed.

performance-based EI tests, construct validity, measurement invariance, early adolescents

nije evidentirano