Can emotional intelligence be measured?

Emotional intelligence is measured constantly and defined inconsistently, which is why claims about it range from well-supported to indefensible without changing the vocabulary.

There are two distinct traditions, and almost every dispute about EI dissolves once you identify which one is being discussed.

Ability EI

Salovey and Mayer's original 1990 definition treated emotional intelligence as an ability: the capacity to perceive emotion accurately, use it to facilitate thought, understand emotional meaning, and manage emotion in yourself and others.

An ability is measured with a performance test — problems that have better and worse answers. The MSCEIT does this: it asks respondents to identify emotions in faces, judge how emotions blend and progress, and evaluate the effectiveness of regulation strategies.

The persistent difficulty is scoring. There is no objective key for "what is the most effective way to handle this emotional situation", so answers are scored against consensus — the modal response of a large sample, or of experts. Critics reasonably ask whether that measures emotional ability or conformity to a normative view of emotion, and whether an unusually skilled respondent would be penalised for departing from consensus.

Ability EI correlates modestly with general intelligence, which supports its claim to be an ability, and its incremental validity over intelligence and personality is real but small.

Trait EI

The other tradition measures emotional self-efficacy through self-report: how emotionally capable you believe yourself to be. This is what the widely used short instruments capture, and Petrides and Furnham formalised it as a construct located in personality space rather than in cognitive ability space.

This is a legitimate construct. Beliefs about your own emotional competence shape what you attempt, and self-report is the only way to reach a belief.

It is not, however, a measure of emotional ability, and the correlation between trait EI and ability EI is low — commonly reported around 0.2 to 0.3. They are two different things with one name.

The evidence, kept separate

Trait EI predicts job satisfaction, organizational citizenship behaviour, leadership emergence and wellbeing, with modest but replicated effects. The critical finding is the redundancy one: meta-analyses by Van Rooy and Viswesvaran and by Joseph and Newman show substantial overlap with the Five-Factor Model — mainly emotional stability, extraversion and conscientiousness — and limited incremental validity once those and cognitive ability are controlled.

Ability EI has weaker predictive relationships in the workplace but a stronger claim to be measuring a capability rather than a self-view.

The commercial market frequently cites effects that neither tradition supports. Claims that emotional intelligence accounts for the large majority of job success originate in popular writing rather than peer-reviewed research and do not survive contact with the meta-analytic literature.

The structural limitation of self-report here

Self-report cannot detect the person whose emotional self-perception is inaccurate — which is arguably the group of greatest practical interest. Someone poor at reading others who believes they are excellent at it will score high, and there is no item that catches this. It is the same problem self-report faces with mindfulness and with insight generally.

This is not a reason to discard trait EI. It is a reason to read a high score as "I see myself as emotionally capable" rather than "I am emotionally capable", and to treat the two as questions with potentially different answers.

So can it be measured?

Emotional self-perception: yes, reliably, with the caveat that it overlaps heavily with established personality traits.

Emotional ability: partially, with the unresolved scoring problem above.

The thing the term usually promises in commercial contexts — a single number capturing emotional competence, predictive of success, distinct from personality — has not been demonstrated by either tradition.

Where NOESIS publishes emotional intelligence instruments, the test page states which model the instrument implements and whether it measures ability or self-perception, because that difference determines what the score is entitled to say about you.