Emotional Intelligence in Children: What the Research Actually Shows

A nine-year-old notices his teammate is upset after a bad play and says something kind instead of piling on. A twelve-year-old is frustrated by a difficult homework problem, recognizes the frustration, and takes a break before it turns into a meltdown. A seven-year-old manages to share a toy she doesn’t want to share because she can read what her friend needs and regulate her own impulse well enough to act on that reading. These moments look unremarkable in a happy childhood. The research says they’re actually predictive of outcomes that IQ scores miss.

Key Takeaways

• Emotional intelligence (EQ) is a genuine and measurable psychological construct, distinct from both IQ and from general personality, though its measurement remains contested among researchers.
• Well-designed studies show EQ predicts relationship quality, some mental health outcomes, and certain kinds of academic performance — particularly in collaborative and high-stress contexts.
• The popular version of EQ (from Goleman’s 1995 book) overclaims substantially; the scientific version (from Mayer and Salovey’s ability-based model) is more modest and more rigorous.
• EQ can be developed through deliberate, consistent parenting practices — emotion coaching, modeling, and naming — though it isn’t a simple skill that can be taught in a curriculum and then assessed.
• Social-emotional learning programs in schools show genuine outcomes for behavior, relationships, and some academic measures, but vary enormously in quality and implementation.

The Problem: Popular EQ Theory vs. the Science

In 1995, Daniel Goleman published Emotional Intelligence: Why It Can Matter More Than IQ. The book sold millions of copies, launched a corporate training industry, reshaped how schools thought about social development, and introduced EQ into everyday vocabulary. It also, in the estimation of many psychologists who study emotion and personality, overclaimed significantly.

Goleman’s version of EQ was expansive. In his framing, emotional intelligence encompassed motivation, persistence, empathy, social skill, and self-awareness — a grab-bag of traits that together amounted to something like “everything good about a person’s character that IQ doesn’t measure.” He cited research suggesting EQ accounted for more of life success than IQ, that it predicted outcomes across virtually every domain, that it was the real driver of achievement and happiness.

The problem is that this version of EQ doesn’t hold up under rigorous scrutiny. When personality psychologists examined Goleman’s claims against the measurement literature, they found two core problems: first, much of what he called EQ was already captured by existing personality dimensions (particularly conscientiousness and agreeableness from the Big Five); second, the instruments used to measure EQ in his cited studies often weren’t measuring a distinct cognitive ability — they were measuring personality self-reports that overlapped heavily with existing measures.

This critique doesn’t mean emotional intelligence isn’t real or important. It means the honest version of EQ research is more narrow and more carefully specified than the bestseller suggested. The scientific model developed by Peter Salovey and John Mayer — which preceded Goleman’s popularization — is both more rigorous and more modest. It’s also more useful for parents.

What the Research Actually Says

Mayer and Salovey’s Ability Model (1990, 1997)

The Salovey-Mayer model defines emotional intelligence as a set of four cognitive abilities arranged in a hierarchy: perceiving emotions accurately (in faces, voices, images), using emotions to facilitate thought, understanding how emotions develop and change over time, and managing emotions in oneself and others. Crucially, this model treats EQ as an ability — something measured by how well you actually perform on emotion-related tasks, not by how emotionally intelligent you report yourself to be.

This distinction matters enormously. Self-report EQ measures — “I am good at understanding others’ feelings” — largely measure personality and self-concept, not ability. Ability-based EQ measures, like the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT), actually present emotion-related problems and score performance against consensual or expert-defined correct answers. Studies using ability-based measures consistently find smaller but more robust effects than studies using self-report measures — and find EQ to be more distinct from existing personality measures.

A 1999 study by Mayer, Salovey, and Caruso found that ability-based EQ predicted supervisor performance ratings, social competence measures, and positive relationships, even after controlling for verbal intelligence and the Big Five personality traits. The effects were modest — EQ wasn’t explaining enormous additional variance — but they were real and specific to EQ as distinct from general personality.

Brackett and Mayer (2003) — EQ and Life Outcomes in Young Adults

Marc Brackett and John Mayer’s 2003 study followed a college-age sample and measured ability-based EQ against a broad set of outcomes including academic performance, quality of social relationships, illegal drug and alcohol use, and behavior. Men with higher ability-based EQ showed less drug and alcohol use and higher quality relationships. Women with higher EQ showed better academic performance and more positive social behavior. Both findings held after controlling for personality.

The drug-and-alcohol finding was particularly striking because it suggests EQ is functioning partly as a stress regulation resource — people with higher capacity to identify, understand, and manage their emotional states may be less likely to turn to external regulation strategies like substance use. This is one of the more practically important findings in the EQ literature and one that rarely makes it into the popular coverage.

CASEL Meta-Analysis on Social-Emotional Learning (Durlak et al., 2011)

The largest and most rigorous synthesis of school-based social-emotional learning (SEL) programs — the domain most directly relevant to parents who want to support children’s EQ development — was published by Durlak, Weissberg, Dymnicki, Taylor, and Schellinger in 2011 in Child Development. Analyzing 213 school-based SEL programs covering 270,000 students from kindergarten through high school, they found:

• Social and emotional skills improved significantly, with an average effect size of 0.57.
• Attitudes toward school and self improved with an effect size of 0.23.
• Prosocial behavior improved with an effect size of 0.24.
• Behavior problems decreased with an effect size of -0.22.
• Academic achievement improved with an effect size of 0.27 — the equivalent of approximately 11 percentile points.

This is one of the stronger effect sizes in the education intervention literature, though it’s important to note that the quality of included studies varied, and the academic achievement finding in particular has been harder to replicate in more recent high-design studies. The behavioral and social skill findings are more consistent across study quality levels.

The Durlak meta-analysis also identified the SAFE model (Sequenced, Active, Focused, Explicit) as the design feature most predictive of positive outcomes — the same model found effective in after-school program research. Programs that didn’t use this approach showed dramatically smaller effects.

Goleman’s Claims vs. the Data: A Necessary Correction

Goleman’s central claim — that EQ “matters more than IQ” — is not supported by the rigorous measurement literature. Studies using ability-based EQ measures find that EQ explains modest incremental variance in outcomes after controlling for IQ and personality — meaningful but not dominant. A 2006 review by Locke published in Industrial and Organizational Psychology found that Goleman’s mixed model of EQ (combining ability and personality) was conceptually incoherent and measured largely the same things as existing personality instruments. Studies using Goleman-derived measures tend to report larger EQ effects, but these effects are inflated by the personality overlap.

What the data does support is that EQ, carefully measured, predicts outcomes that IQ doesn’t — particularly in social domains, relationship quality, and stress regulation. IQ predicts outcomes that EQ doesn’t — particularly in academic achievement in knowledge-based domains. They’re measuring different things, and life outcomes depend on both.

Denham et al. (2003) — Emotional Competence in Early Childhood

Susanne Denham’s longitudinal research on emotional competence in preschool children followed children from age 3 through kindergarten and found that emotional knowledge — the ability to identify and understand emotions in oneself and others — at age 3 significantly predicted social competence and teacher-rated adjustment in kindergarten, even after controlling for verbal ability. Children who could accurately identify and label emotions at age 3 were better liked by peers, had more positive interactions with teachers, and had fewer behavioral problems at the start of formal schooling.

This is important for parents of young children because it suggests emotion knowledge is both meaningful and developable at very young ages — and that the window for foundational emotional skill building starts before kindergarten, not in a school SEL curriculum.

EQ vs. IQ in Predicting Outcomes

Outcome Domain	IQ Predicts?	EQ Predicts?	Relative Strength	Research Quality
Academic achievement (knowledge-based)	Strong	Weak-Moderate	IQ dominates	High (many large studies)
Academic achievement (collaborative/project work)	Moderate	Moderate	Roughly equal	Moderate
Quality of close relationships	Weak	Moderate-Strong	EQ dominates	Moderate
Job performance (technical roles)	Strong	Weak	IQ dominates	High
Job performance (leadership/social roles)	Moderate	Moderate-Strong	Roughly equal, EQ edge	Moderate
Mental health outcomes (anxiety, depression)	Weak	Moderate	EQ dominates	Moderate
Substance use risk	Weak	Moderate	EQ dominates	Moderate
Behavioral problems in school	Weak	Moderate-Strong	EQ dominates	Moderate-High
Peer acceptance and social competence	Weak	Moderate-Strong	EQ dominates	Moderate-High
Life satisfaction (self-reported)	Weak	Moderate	EQ dominates	Moderate

Note: “Predicts” means independent prediction after controlling for the other construct and basic personality. All effects are population-level averages; individual variation is large.

What to Actually Do

The SEL and parenting research on EQ development converges on a set of specific practices. None of them require a curriculum or a program. They require consistent, deliberate attention to how emotional content gets handled in daily interactions.

Practice Emotion Coaching, Not Emotion Dismissing

John Gottman’s research on family interactions identified emotion coaching — acknowledging and validating a child’s feelings before moving to problem-solving or correction — as one of the most important parenting behaviors predicting children’s emotional competence. The contrast is “emotion dismissing,” which treats feelings as problems to be fixed or ignored: “Stop crying. It’s not a big deal.” Emotion coaching treats feelings as valid and important information: “I can see you’re really frustrated right now. That makes sense.”

This isn’t about validating every behavior (she can be frustrated without being allowed to throw things) — it’s about validating the feeling before addressing the behavior. Gottman’s longitudinal research found children of emotion-coaching parents showed better emotional regulation, fewer behavioral problems, stronger peer relationships, and better academic achievement — a finding that has been replicated across multiple cultural contexts.

Name Emotions With Specificity

The vocabulary of emotion matters. A child who can distinguish between “frustrated,” “disappointed,” “embarrassed,” and “overwhelmed” is better equipped to understand and manage their emotional experience than a child whose entire negative emotional vocabulary is “mad” or “bad.” This is sometimes called “emotional granularity” — the ability to make fine-grained distinctions among emotional states.

Lisa Feldman Barrett’s research on emotional granularity found that individuals with higher granularity show better emotion regulation, less intense negative emotional experiences, and more adaptive responses to stress. Children develop emotional granularity through exposure to emotion vocabulary — which means parents who regularly use specific, varied emotion language while naming what they’re observing (“you look disappointed that we couldn’t stay longer”) are actively building this capacity.

Start simple with young children — happy, sad, scared, angry, surprised — and gradually introduce more nuanced vocabulary as children develop. Reading books with emotionally complex characters and discussing their emotional states is one of the most accessible ways to build this vocabulary in context.

Model the Emotional Regulation You Want to See

Children learn emotion regulation from observation at least as much as from explicit instruction. A parent who names their own emotional states out loud — “I’m feeling pretty stressed right now, so I’m going to take five minutes to calm down before we figure this out together” — is demonstrating both emotional awareness and a regulatory strategy simultaneously. This modeling is more powerful than any curriculum.

Conversely, parents who regularly display dysregulated responses to stress, who dismiss their own emotional experiences, or who treat emotional expression as weakness are modeling those approaches as well. The literature on intergenerational transmission of emotional competence is consistent: parental emotional competence is the strongest predictor of child emotional competence, through both genetic and socialization channels.

Don’t Force Emotional Display or Sharing

One of the most counterproductive practices in parenting for emotional development is forcing emotional performance — making a child hug someone they don’t want to hug, demanding apologies that aren’t genuine, requiring “happy” behavior at family events. These practices teach children that emotional experience is less important than social performance, which is precisely the opposite of the emotional awareness and authenticity that EQ development requires.

Children develop genuine emotional understanding through real emotional experience, not through performing expected emotional responses. Giving children agency over their own emotional expressions — within the limits of appropriate behavior — supports authentic development.

Look for SEL Programs With Evidence-Based Design

If your child’s school uses an SEL curriculum, it’s worth knowing which one and whether it meets the CASEL SAFE criteria. Programs like Second Step, RULER (developed at the Yale Center for Emotional Intelligence), and PATHS (Promoting Alternative Thinking Strategies) have been evaluated in rigorous studies and show consistent positive outcomes. Generic social skills lessons that don’t follow a structured curriculum with evidence behind it are much harder to evaluate and have weaker research support.

Parents can ask schools which SEL curriculum they use and whether it’s been evaluated. The CASEL website maintains a guide to evidence-based SEL programs that’s publicly accessible. This isn’t a small issue: a well-implemented evidence-based SEL program can produce measurable behavioral and academic benefits; an improvised approach may produce no measurable effect.

What to Watch for Over the Next 3 Months

Emotional intelligence develops slowly and doesn’t produce dramatic before-after transformations in a semester. But there are observable behavioral markers worth watching.

Week 4: Notice how your child handles frustration — not meltdowns (which are normal), but the recovery from frustration. Is she able to articulate what she’s feeling after she’s calmed down? Is she beginning to use any of the emotion vocabulary you’re introducing? Small shifts in labeling are the earliest behavioral signal that emotional awareness is developing.

Month 2: Look for evidence that your child is starting to notice emotions in others. Does she comment on how a sibling, friend, or character in a book might be feeling? Unprompted emotional observation is a meaningful developmental marker. You can support this by asking open-ended questions after social situations: “How do you think your friend felt when that happened?”

Month 3: Check in on peer relationship quality. EQ’s most consistent short-term behavioral outcome is improved quality of peer interactions — fewer conflicts that escalate, more repair behaviors after conflicts, wider social acceptance. This is hard to measure formally, but you can get a qualitative read from your child’s descriptions of her school day, from teacher feedback at conferences, and from your own observations of how she handles social friction.

Frequently Asked Questions

Is emotional intelligence something kids are born with or can it be learned?

Both. Twin studies suggest emotional intelligence has a moderate genetic component — roughly 30–40% of variance in ability-based EQ measures appears heritable. But the environment — particularly parenting practices, school environment, and emotional experience — shapes the remaining variance substantially. EQ can be meaningfully developed, particularly in early childhood, and parenting practices have a well-documented effect on children’s emotional competence.

At what age should I start thinking about my child’s emotional intelligence?

Earlier than most parents expect. Denham’s research shows that emotional knowledge at age 3 predicts social and academic outcomes at kindergarten entry. The foundational skills — identifying basic emotions, beginning to understand that others have different feelings, early regulation of emotional responses — are developing in toddlerhood and preschool. The practices that support EQ development (emotion coaching, emotion naming, modeling regulation) are appropriate and beneficial from infancy.

Can emotional intelligence be taught in a classroom?

It can be supported and developed in structured ways, but “taught” in the same way math is taught overstates the directness of the instruction. School-based SEL programs that use evidence-based structured curricula show real benefits for emotional and social skill development. But the evidence suggests that high-quality relationships with emotionally attuned adults — parents and teachers — are more predictive of EQ development than any curriculum. Programs supplement relationship; they don’t replace it.

My child scores high on IQ tests but struggles socially. Does EQ explain this?

Possibly in part, but social struggles have multiple potential causes beyond EQ — including temperament, anxiety, autism spectrum traits, ADHD, and environmental factors. High IQ and lower EQ coexist in many people and isn’t inherently pathological. If a child’s social struggles are causing distress or functional impairment, that’s worth evaluating with a psychologist who can distinguish between the relevant contributing factors — not just assuming it’s an EQ issue.

Are boys less emotionally intelligent than girls?

The research finds small gender differences in ability-based EQ, with girls generally scoring modestly higher on emotion perception and emotion management subscales. These differences are population-level averages and there is enormous overlap between distributions — knowing a child is male tells you almost nothing about his individual EQ level. What research does find is that emotional expression is more socially sanctioned for girls than boys in most Western cultures, which affects both the development and the expression of emotional competence without necessarily reflecting underlying ability differences.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

Sources

1. Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D. Sluyter (Eds.), Emotional Development and Emotional Intelligence: Educational Implications (pp. 3–31). Basic Books.
2. Goleman, D. (1995). Emotional Intelligence: Why It Can Matter More Than IQ. Bantam Books.
3. Brackett, M. A., & Mayer, J. D. (2003). Convergent, discriminant, and incremental validity of competing measures of emotional intelligence. Personality and Social Psychology Bulletin, 29(9), 1147–1158.
4. Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The impact of enhancing students’ social and emotional learning: A meta-analysis of school-based universal interventions. Child Development, 82(1), 405–432.
5. Denham, S. A., Blair, K. A., DeMulder, E., Levitas, J., Sawyer, K., Auerbach-Major, S., & Queenan, P. (2003). Preschool emotional competence: Pathway to social competence? Child Development, 74(1), 238–256.
6. Gottman, J. M., Katz, L. F., & Hooven, C. (1996). Parental meta-emotion philosophy and the emotional life of families. Journal of Family Psychology, 10(3), 243–268.
7. Barrett, L. F., Gross, J., Christensen, T. C., & Benvenuto, M. (2001). Knowing what you’re feeling and knowing what to do about it: Mapping the relation between emotion differentiation and emotion regulation. Cognition and Emotion, 15(6), 713–724.
8. Locke, E. A. (2005). Why emotional intelligence is an invalid concept. Journal of Organizational Behavior, 26(4), 425–431.
9. Mayer, J. D., Salovey, P., & Caruso, D. R. (2000). Models of emotional intelligence. In R. J. Sternberg (Ed.), Handbook of Intelligence (pp. 396–420). Cambridge University Press.