Key Takeaways
- A meta-analysis of 58 controlled studies confirms EI training reliably raises measured scores, with pre-post effect sizes of d approximately 0.61 and treatment-control effect sizes of Hedges’ g approximately 0.46.
- Score gains and behavior change are not the same outcome. The research base has largely measured the first without confirming the second.
- Programs that include active practice and constructive feedback consistently outperform lecture-based formats. Interactivity is the primary design variable separating score improvement from behavioral transfer.
- Manager reinforcement after training determines whether behavioral gains hold. Without structured check-ins and coaching prompts in the weeks after training, even well-designed programs produce diminishing returns over time.
- L&D leaders who build independent behavioral observation and longer follow-up windows into their evaluation designs are producing organizational evidence the broader research base currently lacks.
An L&D team at a mid-size professional services firm renews its emotional intelligence training contract for the third consecutive year. Satisfaction scores are high, completion rates are solid, and self-assessments show measurable gains. The question no one has asked is whether any of that changed how people work. A large-scale review of 58 studies confirms that EI training reliably raises measured scores. What happens to workplace behavior after training ends is a separate question, and the research is far less settled on that point than most program proposals let on.
Emotional Intelligence Training: Score Gains Are Proven, Behavior Change Is Not
What 58 Studies Found, and What They Measured
The most comprehensive evidence base for emotional intelligence training comes from a 2019 meta-analysis by Mattingly and Kraiger, examining 58 controlled studies. The researchers found treatment-control effect sizes of Hedges’ g approximately 0.46 and pre-post effect sizes of d approximately 0.61, both falling in the moderate range by conventional benchmarks. The QIC-WD umbrella summary of that research frames the finding plainly. EI training reliably improves measured EI.
These figures are worth knowing when you are building the business case for emotional intelligence across HR and management development, and L&D leaders rely on them for that very reason. The effect sizes are consistent and replicated. But that same consistency makes the follow-up question harder to ignore: when scores go up, does the way people work in your organization change in practice? The research on that question is thinner than the score data suggests, and most programs have not updated their design since 2015.
What Those Numbers Do Not Tell You
Those numbers measure EI scores, assessed by self-report instruments before and after training, across studies published between 2000 and 2019. They do not measure whether a manager in your organization handles a difficult feedback conversation differently six months later, whether team relationships shift, or whether interpersonal conflict decreases. If you are running EI training because you want changed workplace behavior, you are relying on score gains to stand in for something the research has not yet reliably confirmed.
Why EI Scores Rise While Workplace Behavior Stays the Same
The Part That Works: Score Gains Are Real
Participants who complete EI programs score meaningfully higher on assessments than those who do not. A 2025 simulation-based intervention study with 106 primary care physicians and nurses in Spain found rank-biserial effect sizes of 0.35 to 0.40 on intrapersonal and interpersonal dimensions at immediate post-intervention assessment, with interpersonal gains reaching 0.59 in the nurses subsample. The score gains confirmed by the broader 58-study meta-analysis are real, consistent, and replicable across program types.
The Part Less Documented: Behavior After Training Ends
What the research tracks far less carefully is what happens in the months after training ends. Most studies measure outcomes at program completion or within a few weeks. Independent behavioral observation (whether your direct reports notice changed feedback patterns, whether team psychological safety shifts, whether interpersonal conflict decreases) appears in only a fraction of studies. The 2025 simulation-based intervention study measured EI scores immediately post-intervention with no follow-up window. Behavioral change in clinical practice was not tracked. Score improvement and behavior change are not the same outcome, and the research base has largely treated them as interchangeable by measuring the first and assuming the second.
The Design Decision That Determines Whether EI Training Changes Behavior
What Lecture-Based Programs Give You
The distinction between what lecture-based formats deliver and what they leave unaddressed breaks into two separate questions worth examining before choosing a program design.
What Scores Your Program Will Move
Lecture-based programs do produce score gains, and that is not a trivial result. Your participants will finish the program understanding what emotional regulation is, how empathy works as a skill, and how self-awareness can be applied. EI assessment instruments are built to capture that kind of conceptual knowledge, so scores rise. The research confirms this across diverse program formats.
What Behavior Your Program Is Not Designed to Change
The gap opens when you ask whether those score gains show up in how people work. Understanding that empathy matters does not create the habit of pausing before responding in a difficult meeting. Behavioral skills require deliberate practice to develop, and lecture-based programs do not build it in. Without structured rehearsal and corrective feedback, the gap between knowing and doing persists after participants leave the room.
What Practice-and-Feedback Programs Give You Instead
Mattingly and Kraiger’s design-variable finding is clear. Programs that include active practice and constructive feedback outperform lecture-based formats across the full 58-study dataset. The moderating effect of interactivity holds consistently regardless of study design. Programs built around behavioral rehearsal, feedback loops, and spaced repetition produce larger effects than those built around content delivery alone. For L&D teams designing or selecting emotional intelligence training, this is the most actionable finding in the research base.
Why Spending Keeps Growing Without Accountability for Transfer
Why Budget Keeps Flowing Into Programs That Stop at Awareness
Your organization is not alone in this pattern. Average direct learning spend in 2024 was $1,254 per employee, per ATD data, and EI and soft-skills programs make up a meaningful share of leadership training for managers’ budgets across the sector. The spend continues not because L&D functions are indifferent to results, but because most measurement frameworks do not distinguish between awareness gain and behavioral change. When your satisfaction scores are solid and completion rates are high, the program looks successful by every metric you have. Whether behavior changed is a question those metrics were never designed to answer.
The Problem Training Design Cannot Solve on Its Own
Even well-designed programs with practice and feedback components face one structural constraint training design cannot resolve on its own. What happens in the weeks and months after training ends is equally important. Without manager reinforcement (coaching moments, application prompts, observed practice in context), participants who leave training with genuine behavioral gains tend to revert toward baseline patterns over time. The constraint is a transfer architecture problem, operating largely outside the scope of the program itself. Research on learning transfer dating to Baldwin and Ford’s foundational 1988 review consistently documents this pattern across skill domains.
Browse practice-based courses on EI, leadership communication, and behavioral skill development across the KnowledgeCity course library.
What the Research on EI Training Still Cannot Confirm
Two Gaps the Research Base Has Not Closed
Gap 1: The Research Mostly Measures What People Say About Themselves
The most pervasive measurement limitation in EI training research is outcome type. The majority of studies in the 58-study meta-analysis used self-report instruments as the primary outcome measure. Participants rate their own emotional awareness and regulation before and after training. This captures something real, but self-ratings are sensitive to the framing participants have just received in training, a phenomenon sometimes called response shift. An independent observer rating the same person’s behavior before and after the program would be a more demanding and more informative test of what the training changed.
Gap 2: Most Studies Stop Tracking Too Soon
The 2025 simulation-based intervention study assessed outcomes immediately post-intervention, with no follow-up window. That immediate-assessment approach is not an outlier but the norm across the research base. Short follow-up windows capture the immediate effect of training but cannot tell you whether behavior holds up under pressure, in high-stakes conversations, or six months into daily work. Six-month behavioral observation is expensive, which is why most studies skip it. The research cannot currently confirm how durable EI training results are, and almost none of the studies have tracked outcomes past the first few weeks.
What Research on Active Listening Tells You About EI Program Design
Research on active listening, a component competency within broader EI development with dedicated studies dating to 2012, runs into the same wall. Structured practice and feedback improve observable listening behaviors in the short term, but durability without reinforcement is less documented. Where this research is more useful than most EI studies is in its measurement approach, which relies on behavioral checklists and observer ratings rather than self-report alone. If you are designing active listening and interpersonal communication skills programs and building in observer-rated measures, you are already applying a more rigorous framework than most EI programs use. That same approach translates directly to broader EI program design.
How to Build Emotional Intelligence Training That Changes Behavior
Build for Behavioral Transfer, Not Just Completion
The design implication from the research is specific: build behavioral practice into the program itself, not as an optional post-training exercise. Participants need structured rehearsal with immediate corrective feedback and spaced repetition across multiple sessions rather than a single training event. Scenario-based practice tied to situations they encounter in their actual roles outperforms generic EI exercises. Research on how to improve emotional intelligence points consistently to practice and feedback as the primary design variable separating programs that move scores from those that change observable behavior.
Treat Manager Reinforcement as a Design Requirement, Not a Post-Launch Task
Program design handles one part of the transfer problem. What managers do in the weeks after training handles the other. CIPD data from 2023 shows that perceptions of line manager support for learning transfer are generally low across organizations. If your participants return from training to a manager who has not been briefed on what to reinforce, the behavioral gains from even a well-designed program are at risk. Guidance on how L&D should train managers to coach effectively is directly relevant here. Build these reinforcement practices into the program design before launch:
- Pre-training briefing for managers covering what participants will practice and how to reinforce it in day-to-day team interactions
- Structured check-in prompts at 30, 60, and 90 days post-training, tied to specific behavioral objectives from the program
- Coaching conversation frameworks that connect EI competencies to observed team behaviors rather than abstract trait descriptions
- Measurement frameworks that include manager-rated behavioral observations alongside, not instead of, learner self-assessments
Where Emotional Intelligence Training Is Heading
Where the Best Programs Are Already Heading
The shift is already underway in how the best programs are being designed. Single-event awareness sessions are giving way to practice-intensive, spaced, scenario-based formats. The 2025 simulation-based intervention study points toward a meaningful step forward. Programs that allow repeated behavioral rehearsal in realistic environments can address both the design problem and the measurement problem at the same time. If that model gets replicated with longer follow-up windows, the evidence base will be considerably stronger.
What Would Make This Research More Useful to You
The research base will advance when two specific changes become more common. Independent behavioral observation needs to replace self-report as the primary outcome measure. Longer follow-up windows, including data at three, six, and twelve months post-training, are equally necessary. L&D leaders who build these measurement features into their program evaluation designs are not waiting for academic research to catch up. They are producing the organizational evidence base their programs need to justify continued investment. Understanding how to measure the effectiveness of soft skills is an increasingly important competency for the function as a whole.
What Your EI Training Program Needs to Do Differently
What the 58-Study Dataset Can and Cannot Confirm
The 58-study meta-analysis confirms that EI training raises measured scores reliably and consistently. What the research has not confirmed is that score gains automatically translate into changed workplace behavior. Whether your program closes that gap depends on whether it is designed to build behavioral practice rather than just awareness, whether your organization creates conditions for on-the-job application, and whether your measurement approach can detect what changed in the months after training ends.
The Design Decisions That Determine Whether Transfer Happens
Lecture-based programs reliably produce knowledge gains and score improvements. They do not reliably produce the behavioral practice required for transfer to daily work. Practice-and-feedback designs, supported across the 58-study dataset, produce stronger results, and manager reinforcement in the weeks after training determines whether those results persist. When program design, delivery format, and post-training reinforcement are treated as separate decisions made by different parts of the organization, the behavior gap stays open regardless of how good the training content is.
The Question Your Program Design Has to Answer
The evidence for EI training is solid enough to justify continued investment. What it does not justify is treating score gains as a proxy for behavior change or satisfaction surveys as evidence of transfer. L&D leaders who ask whether a program is built to change behavior, not just improve how participants talk about emotional intelligence, are asking the question the research has not yet answered well enough for anyone else to answer it for them.
KnowledgeCity’s workforce development platform includes practice-based courses on emotional intelligence, active listening, leadership communication, and managerial coaching, designed to move beyond awareness and build the behavioral skills that transfer to the job.
Frequently Asked Questions
1. Can emotional intelligence be improved through training?
- Yes. Meta-analytic evidence from 58 studies shows consistent moderate improvement in measured EI following training, with pre-post effect sizes of d approximately 0.61 and treatment-control effect sizes of Hedges’ g approximately 0.46. These findings come from the Mattingly and Kraiger (2019) meta-analysis, summarized by QIC-WD (2021).
2. How long do gains from emotional intelligence training last?
- The current research base has limited long-term follow-up data. Most studies assess outcomes at program completion or within six weeks post-training. Whether EI score gains and any associated behavioral changes persist without ongoing manager reinforcement is not well-documented in the existing evidence base.
3. What design features make emotional intelligence training more effective?
Programs that include active practice and constructive feedback consistently outperform lecture-based formats. Spaced repetition, scenario-based rehearsal, and immediate behavioral feedback are the strongest design moderators identified in the Mattingly and Kraiger (2019) meta-analysis. Interactivity, not content coverage, is the primary differentiator.
4. How should L&D leaders evaluate emotional intelligence training outcomes?
Evaluation frameworks combining self-report measures with independent behavioral observation produce more reliable data on what the training changed. Short follow-up windows under six weeks capture initial impact but not sustained behavioral change. Including manager-rated behavioral observations at 30, 60, and 90 days post-training provides a more complete picture of transfer.
5. What is the relationship between active listening training and emotional intelligence training?
Active listening is a component competency within the broader EI skill set. Programs that address active listening as a distinct behavioral objective, using behavioral checklists and observer ratings rather than self-report only, tend to produce more transfer-oriented outcomes. The measurement practices from active listening research apply directly to broader EI program evaluation design.
References
- Mattingly, V., & Kraiger, K. (2019). Can Emotional Intelligence Be Trained? A Meta-Analytical Investigation. Human Resource Management Review. Summarized in: QIC-WD Umbrella Summary: Emotional Intelligence Training. Quality Improvement Center for Workforce Development, October 2021.
- Association for Talent Development (ATD). (2025). 2025 State of the Industry: Talent Development Benchmarks and Trends (2024 data). ATD Press.
- CIPD. (2023). Learning at Work Survey Report 2023. Chartered Institute of Personnel and Development.
- Pérez-Fuentes, M.C., Molero Jurado, M.M., et al. (2025). Simulation-based training in emotional intelligence and self-esteem: enhancing effectiveness and wellbeing in healthcare. Frontiers in Public Health, 13:1667192.
