Does EQ or IQ predict success better? We break down the meta-analyses on both, the 80% myth, and which skill is worth training first — backed by research.
Ask around and you'll hear two confident, contradictory answers. One camp insists that raw brainpower — IQ — decides everything: grades, careers, income. The other quotes a line from Daniel Goleman's 1995 bestseller: emotional intelligence "can matter more than IQ." Somewhere along the way, that turned into a piece of pop-psychology folklore: EQ accounts for 80% of success, IQ for a measly 20%.
Neither camp is right. The actual research, built on decades of meta-analyses, tells a more interesting story — one where both capacities matter, but in different arenas, and where the honest answer to "which one predicts success?" is: it depends entirely on which success you're measuring. Let's walk through the evidence.
What IQ Actually Predicts (And How Well)
IQ's predictive power is one of the most replicated findings in all of psychology. Correlations between iq scores and academic achievement sit around 0.50 to 0.60 — very strong by social-science standards. For job performance, meta-analyses consistently find correlations of roughly 0.40–0.55 for complex work (engineering, science, finance, programming) and 0.20–0.30 for simpler roles. IQ also quietly predicts things most people never associate with it: income, occupational status, even health and longevity.
But there's a ceiling, and it matters. Above a threshold of roughly 115–120, extra IQ points buy diminishing returns for real-world outcomes. Once you're smart enough to do the job, being smarter stops separating the stars from everyone else. Researchers call this the threshold effect, and it's the single best starting point for understanding the whole IQ-versus-EQ debate.
One more property worth knowing: IQ is remarkably stable. Longitudinal studies following people from childhood into their seventies find correlations around 0.73 between early-life scores and late-life scores. Your cognitive foundation is set early and doesn't move much — which is exactly why the trainability of EQ (more on that below) is such a big deal.
What EQ Predicts — and the Measurement Problem Nobody Mentions
Here's where things get murky, and where most popular articles quietly mislead you. "EQ" is measured in two very different ways, and they produce very different evidence.
Ability-based tests, like the MSCEIT, treat emotional intelligence as a skill you either demonstrate or don't: How accurately do you read emotions in faces? Can you use emotional information to solve problems? These tests show modest but genuine predictive power — correlations of roughly 0.18–0.28 with job performance, strongest in roles with heavy emotional demands like sales, management, nursing, and teaching.
Self-report inventories, like the EQ-i, just ask people to rate themselves. These show flashier correlations with success — but they also overlap heavily with known personality traits: agreeableness, conscientiousness, emotional stability. When researchers statistically control for personality, the "unique" contribution of self-reported EQ often shrinks dramatically. Much of the eye-popping evidence behind "EQ beats IQ" headlines comes from these self-report measures.
That doesn't mean EQ is fake. The cleanest findings are consistent: emotionally intelligent people tend to have better relationships, lower stress and anxiety, stronger leadership ratings, and better outcomes in people-facing jobs. A widely cited meta-analysis by Joseph and Newman (2011) found that ability-based EQ predicted job performance beyond IQ and personality — but almost exclusively in jobs with high emotional labor. For technical roles, the incremental prediction was close to zero.
And the famous "58% of job performance" statistic from TalentSmart? It comes from a proprietary self-report assessment across hundreds of thousands of workers. It may be directionally true, but no independent meta-analysis reproduces a number that large with ability-based measures. Treat it as marketing-adjacent, not settled science.
Where the "EQ Matters 80%" Myth Came From
Goleman never actually claimed 80%. What he reported, based on competency studies across hundreds of companies, was that when you compare star performers with average ones in the same role, the majority of distinguishing competencies are emotional and social rather than purely cognitive. His argument was about differentiation at the top — among people whose IQ and technical skills are already similar — not about predicting success from scratch.
Over thirty years of retelling, that nuance eroded into "80% of success is EQ," which no peer-reviewed analysis supports. Meanwhile, the most careful direct comparison on record — a seven-year study of general managers by Dulewicz and Higgs (1998) — found emotional competencies accounted for 36% of the variance in career advancement versus 27% for cognitive measures. A real edge for EQ, yes, but nothing like 80-to-20.
There's also a mental-health dimension where the asymmetry flips entirely. Studies consistently find that higher emotional intelligence correlates with lower stress, anxiety, and depression, while IQ shows no protective effect against any of them. Some research even suggests high-IQ individuals are more prone to overthinking and rumination when they lack the emotional skills to manage it. If "success" includes simply being okay, EQ isn't the runner-up — it's the whole race.
The myth survives because it contains a grain of truth wrapped around a false frame.
The Verdict: It Depends on What "Success" Means
Strip away the hype and the research supports a clean domain-split framework:
IQ wins in analytical arenas. Academic performance, technical mastery, learning speed, and performance in engineering, science, law, and programming remain IQ's home turf. No amount of charm substitutes for the cognitive horsepower to solve hard problems.
EQ wins in interpersonal arenas. Leadership effectiveness, relationship quality, conflict management, sales, mental health and resilience — here cognitive ability predicts surprisingly little. Being smarter does not help you regulate your emotions, read a room, or repair a friendship.
Two more findings sharpen the picture. First, IQ and ability-based EQ correlate at only 0.20–0.35 — they're largely independent traits. High IQ doesn't buy you EQ, and the brilliant-analyst-who-can't-run-a-meeting is a real archetype, not a stereotype. Second, the two interact. A study in the Journal of Organizational Behavior found that people high on both dimensions outperformed those high on just one by roughly 20–30% — the capacities compound rather than compete.
So the fairest one-line answer: IQ determines whether you can do the work; EQ largely determines what happens once other people are involved.
A concrete example makes this vivid. Picture two data scientists with identical modeling skills. One sends her analysis, gets ignored, and assumes the work speaks for itself. The other spends twenty minutes learning who needs the insight, translates the model into three sentences the sales VP actually cares about, and flags one risk before it becomes a fire drill. Same IQ, same technical output — completely different careers. Google's Project Oxygen, which mined years of internal hiring and performance data to find what makes managers effective, landed on almost exactly this: the top differentiators of great managers were coaching, communicating, and supporting their people, not technical depth. Meanwhile, Google's most productive individual contributors were, as you'd expect, heavy on cognitive ability and technical skill. The pattern repeats across every knowledge industry: IQ gets you into the game, EQ decides how far up the field you play.
The Good News: EQ Is Trainable at Any Age
Given IQ's stability, this asymmetry is the most practically important fact in the entire debate. Meta-analyses of emotional-intelligence training programs — including reviews covering 28 and 50 separate interventions — find moderate effect sizes around 0.51 to 0.61, roughly the equivalent of moving from the 50th to the 69th percentile. Follow-up measurements show the gains don't just persist; in some analyses they keep growing after training ends, because emotional skills improve through real-world practice.
Four practices with the best evidence behind them:
- Build your emotion vocabulary. Most people run on "happy, sad, angry, stressed." Learning to distinguish frustrated from disappointed from resentful measurably improves regulation — naming an emotion precisely is the first step to managing it.
- Keep a two-minute emotional journal. Note what triggered you, how intense it was, and how you responded. Patterns emerge within weeks, and pattern-recognition is the engine of self-awareness.
- Ask for feedback — specifically. "How did that land?" after a difficult conversation surfaces blind spots that introspection alone never will.
- Practice the pause. A deliberate breath between trigger and response is the simplest self-regulation intervention there is, and it's exactly what impulse-control research recommends.
Curious where you stand on both dimensions? Take the free iq test and eq test on iQiQGame — and while you're there, the puzzle games double as a cognitive warm-up.
FAQ
Can you have a high IQ and a low EQ? Yes — and the reverse. Since the two correlate only weakly (0.20–0.35), all four combinations exist in the wild. High-IQ/low-EQ individuals often excel technically but struggle with leadership and conflict; neither score predicts the other.
Does EQ really predict salary? The often-quoted "$29,000 annual premium" comes from proprietary industry research and should be taken loosely. What's solid: EQ correlates with promotion and leadership advancement, and compensation tracks seniority in people-facing roles. The salary link is real but indirect.
Which should I work on first? Whichever is your weaker suit — but if you're already in (or aiming for) a technical career, EQ training usually offers more marginal return, since your cognitive skills are likely already doing their job. And unlike IQ, it's a skill you can genuinely build.
Is EQ just personality by another name? No, but self-report EQ tests blur into personality. Ability-based measures like the MSCEIT capture something distinct — and it's that measured ability, not self-image, that carries the strongest independent evidence.
