Title : A Psychometric Investigation into the Construct of Emotional Intelligence

This thesis investigates the construct of Emotional Intelligence (EI), that is, the notion that individuals differ in the extent to which they attend to, process, and utilize affect-laden information. The thesis proposes a distinction between two types of EI based on the method of measurement, viz., self-report versus maximum-performance. ‘Trait EI’ refers to a constellation of emotion-related dispositions and self-perceived abilities and is assessed with self-report measures, whereas ‘ability EI’ refers to the actual abilities to attend to, process, and utilize affect-laden information and ought to be assessed with maximum-performance tests. It is argued that the latter type of EI cannot be operationalized along typical cognitive ability lines because it is not possible to devise items with objectively correct responses, covering its sampling domain comprehensively. Thus, the empirical component of the thesis largely focuses on the construct of trait EI. The psychometric properties of two extant trait EI measures are scrutinized. The construct is also explored with reference to established trait taxonomies through a series of joint factor analyses and it is concluded that it can claim a place at the lower level of established trait hierarchies. Because extant trait EI measures do not cover the construct’s sampling domain comprehensively, two new trait EI inventories are developed. The construct validity of trait EI is examined across a series of correlational and experimental investigations based on samples of adolescents, university students, and adult employees. It is shown that trait EI relates consistently to measures of coping styles, life satisfaction, depression, academic achievement, truancy, and exclusions from school. Experimental evidence indicates that high trait EI individuals are faster at identifying emotional expressions and more sensitive to mood induction procedures than their low trait EI counterparts. Most of these relationships persist even after personality variance is statistically controlled. Last, the thesis investigates gender differences in trait EI, correlations between measured and self-estimated trait EI scores, and the relationship between self-estimates of IQ and EI.