Yes, with a qualification that matters. Training can improve measured emotional competencies such as recognizing emotion, understanding its effect on thinking, regulating responses, and responding to other people. A 2024 systematic review and meta-analysis of workplace interventions found moderate average gains that remained more than three months after training. The same review found high variation between studies and low methodological quality. That means the evidence supports a modest claim: emotional intelligence training can develop skills that are relevant to work. It does not support the promise that a workshop will automatically produce better leaders, eliminate conflict, or raise performance. A team is more likely to see change when training is tied to a specific behavior, practiced in realistic work, supported by managers, and checked with observations or work-relevant outcomes rather than a post-course score alone.
The claim is stronger than the evidence
The popular version of the claim is easy to recognize: teach emotional intelligence, and communication, leadership, teamwork, and performance will improve. It sounds plausible because work is full of emotional information. A manager gives corrective feedback, a colleague goes quiet in a meeting, a deadline turns a small ambiguity into an accusation, and a repair attempt either lowers the temperature or makes it worse. Skills for noticing and responding to those signals should matter.
The difficulty is that emotional intelligence is not one identical thing across all tests or training programs. In the four-branch ability model associated with Mayer and Salovey, it refers to abilities for perceiving, using, understanding, and managing emotion. Other measures ask people to describe their usual behavior or confidence, while mixed competency models combine emotional skills with broader workplace qualities. An improvement on one measure therefore tells you about that measure and its model. It is not a universal upgrade called a higher EQ.
The strongest recent workplace synthesis is encouraging but restrained. Mehler and colleagues reviewed 50 studies and included 27 controlled trials in a separate analysis. Their average effects on emotional competencies were moderate, and the effects persisted beyond three months. They also reported high heterogeneity, meaning that the studies differed considerably in participants, training, measures, and design, alongside low methodological quality. The result supports training as a reasonable development option. It leaves open how much a particular program will change what people do in a particular workplace.
What training can change
Training becomes easier to judge when it names an action instead of a virtue. Emotion recognition might mean noticing frustration in a terse reply before returning the same tone. Understanding might mean separating disappointment about a missed milestone from a judgment about a colleague's character. Regulation might mean buying enough time to answer the work problem rather than broadcasting irritation. Those are different jobs, and a session that treats them as one quality gives people little to practise.
The workplace review by Mehler and colleagues combined interventions aimed at emotional intelligence, empathy, and emotion regulation. It found moderate average improvement in emotional competencies, including in controlled comparisons, and reported that the average effect remained beyond three months. The studies did not all teach the same thing or use the same measure. In fact, the review notes that most interventions focused on emotional intelligence, only two focused on emotion regulation, and none met its inclusion criteria as a dedicated conflict-management intervention. A team should therefore ask what its chosen course actually rehearses.
The answer also depends on the instrument. A self-report questionnaire describes how someone sees their usual tendencies. An ability test presents emotion-related problems and has a different interpretation. An observer report records behavior as it appears in a particular setting. A change in any one of these can be useful, but it is evidence about that source of information, not a universal rise in EQ.
Why a better score is not the same as better work
A score records a response to a measure. It does not record what happened in a meeting, handoff, or performance conversation. Someone may understand the vocabulary, feel more confident, or answer in a more desirable way without using the skill when status, time pressure, and self-protection are involved. Mehler's review makes this gap especially important: many included studies used self- or peer-reports, and the authors judged 18 of the 50 studies to have critical risk of bias and 28 to have serious risk. The average result is worth taking seriously; it is not a substitute for watching the work.
There is a second gap between emotional skill and job performance. O'Boyle and colleagues found associations between several EI approaches and performance, but an association cannot show that raising a training score causes a performance gain. If unclear authority, chronic overload, or punishment for disagreement is driving a conflict, asking employees to regulate themselves shifts responsibility to the wrong place. Training can help someone describe a broken process; a manager still has to repair the process.
The skill has to appear in a real exchange
Consider a familiar decision after a project handoff. One person says, “This arrived too late for us to review it properly.” The recipient hears blame and begins explaining why the delay was unavoidable. The sender hears defensiveness and becomes more specific, then sharper. A training concept becomes useful only when it changes what happens next.
A practiced response might start with noticing heat, embarrassment, or the urge to prove innocence. The recipient can ask one clarifying question before defending the timeline: “Are you saying the review window was the main problem, or that the contents also missed what your team needed?” The answer may lead to an apology, a correction, or a boundary around an unrealistic deadline. Each option can be heard or seen by another person.
That sequence gives training something concrete to rehearse: pause, check the interpretation, describe impact without assigning motive, make a request, and return to repair if the first attempt goes badly. It does not require permanent calm or agreement. A direct disagreement can be skillful when it is accurate, proportionate, and open to response.
The four-branch model associated with Mayer and Salovey helps keep the actions distinct. Perceiving concerns noticing emotional information; understanding concerns what a feeling may signal and how it may shift; managing concerns influencing the emotional process. EQ Test's self-reflection can prompt a person to examine these moves, while a real exchange supplies the evidence about whether one occurred.
How to test whether learning transferred
Before training begins, write one behavior that can be heard or seen. “Improve communication” cannot guide observation. “In a tense project review, ask one question to check the concern before arguing for a solution” can. The same format works for feedback: describe the impact, give an example, and make a request.
Then follow the behavior through the work. Ask what participants understood, what they could do in rehearsal, what another person noticed later, and whether a relevant process changed. A reaction survey can tell you whether people liked the session. It cannot show that a follow-up happened or that a handoff became clearer. When the stakes justify it, combine a private reflection with an observation and a work record, without turning the result into an employee ranking.
What a responsible team rollout looks like
A rollout should start with a work problem: feedback being avoided, disagreement becoming personal, pressure narrowing listening, or repair arriving too late. The team can then agree on one or two behaviors and decide where they will appear. Team leads belong in that conversation because a new habit cannot survive a meeting culture that punishes it.
Purpose and privacy need to be stated before anyone completes an assessment. EQ Test's browser-based self-reflection is a non-validated development exercise with a local report. It can give an individual a prompt for practice; it should not become a hiring screen, promotion score, team ranking, or employment decision. Participation and individual responses should remain private by default.
The practical decision is whether the team can protect time for rehearsal, conversation, and review. If it can, explore EQ Test for teams at /teams. If it cannot, fix the work condition first. The training case is credible when the team can name the behavior, identify the setting in which it will be visible, and say what evidence will shape the next step.
Questions readers ask
Can emotional intelligence be learned at work?
Evidence from meta-analyses supports improvement in measured emotional competencies after structured training, including in workplace settings. The practical effect depends on what is taught, how it is practiced, and whether the workplace supports the behavior afterward.
Does a higher EQ score prove that workplace skills improved?
No. A higher score shows change on a particular measure. To examine workplace improvement, pair it with specific behavioral observations, participant reflection, and a relevant work-process indicator when one can be interpreted fairly.
What should emotional intelligence training practice?
Practice a defined behavior from real work, such as checking an interpretation before responding to feedback, describing impact without assigning motive, making a clear request, or returning to repair after a difficult exchange.
Should companies use EQ training results to rank employees?
No. EQ training and self-reflection are better suited to development. Using individual results for hiring, promotion, ranking, performance scoring, or team membership can turn a contextual and limited measure into a high-stakes judgment it was not designed to make.
Sources
- Training emotional competencies at the workplace: a systematic review and metaanalysis
Supports the workplace meta-analysis, moderate average effects, persistence beyond three months, heterogeneity, and low methodological quality.
- Can emotional intelligence be trained? A meta-analytical investigation
Supports the synthesis of 58 adult training studies and its moderate positive effect across EI training measures.
- How Efficient Are Emotional Intelligence Trainings: A Meta-Analysis
Supports the earlier meta-analysis of 24 studies and 28 samples, including follow-up effects and model differences.
- The relation between emotional intelligence and job performance: A meta-analysis
Supports the distinction between EI-performance associations and causal evidence that training improves job performance.
- The Science of Emotional Intelligence
Supports the four-branch account of perceiving, using, understanding, and managing emotions.
Apply it to the real situation
See how you respond when work gets emotionally difficult.
From this guide: Choose one behavior from this guide to observe in the next relevant conversation.
Build a private profile across ten emotional-work continuums, then choose one observable behavior to practise.