Emotional intelligence at work

What the evidence supports

Studies tend to find modest links between emotional intelligence scores and job performance, and the pattern underneath is more informative than the average: the link is strongest where the work itself is emotional. Roles built on sustained contact with other people’s states, and roles where composure under pressure is the job, show the clearest relationships, and leadership pulls several of these together. Where work is solitary and technical, the link thins toward nothing.

The workplace skills involved are concrete rather than mystical: reading a meeting before speaking into it, staying usable in a tense hour, repairing a working relationship after friction, and self-starting through the flat stretches. Each maps to a domain, and each moves with practice.

The line we draw

None of this makes a consumer questionnaire a hiring instrument. A self-report score records how a person describes themselves in an unsupervised sitting, with a wide margin, as our accuracy page sets out, and it must not be used to decide anyone’s job or place. That is our position on our own report, printed on it. The EU has gone further for a neighbouring technology: systems that infer emotion from faces or voices are prohibited in workplaces there, apart from narrow medical and safety uses. A questionnaire is a different thing on the grounds that matter: it reads no face and no voice, only the words a person chooses to write about themselves. The choosing is the ethics; the modality is the law.

The rest of this family

Each page here covers one part of the territory: the map starts at what emotional intelligence is.