Emotional Intelligence vs. IQ: What Actually Predicts Success

Short answer: neither IQ nor emotional intelligence alone decides who succeeds. Decades of workplace research point to general cognitive ability — the thing an IQ test measures — as the single strongest predictor of job performance, while emotional intelligence adds a smaller but real edge, especially in roles built around people. The useful question is not "which one wins," but "which one matters for the specific thing you are trying to do."

Below is what the evidence says, stripped of the self-help mythology that has grown up around both ideas.

What each term actually measures

"IQ" is shorthand for general mental ability (often abbreviated g or GMA): the capacity to reason, spot patterns, learn quickly, and hold and manipulate information. It is the trait a well-built cognitive test is designed to capture, and it is remarkably stable across the kinds of problems people face.

"Emotional intelligence" (EI, sometimes EQ) is a looser umbrella. In the research literature it usually means some mix of perceiving emotions accurately, understanding what drives them, managing your own reactions, and using that awareness to guide behavior. Psychologists distinguish ability EI (measured with performance tasks, like reading facial expressions) from trait or mixed EI (measured with self-report questionnaires that blend emotion skills with personality and motivation). That distinction matters, because the two kinds behave very differently in studies.

Why cognitive ability keeps winning the horse race

When researchers ask what best predicts how well someone will do a job, general mental ability has come out on top for decades. The landmark synthesis by Frank Schmidt and John Hunter, published in Psychological Bulletin in 1998, reviewed 85 years of personnel-selection data and concluded that GMA is the most valid single predictor of job performance across occupations, with validity rising as jobs get more complex. Later re-analyses have debated the exact size of the number, but not the basic ranking: raw reasoning ability travels well from one task to the next.

The reason is intuitive. Almost every job involves learning new information, solving unfamiliar problems, and adapting when the routine breaks. Those are exactly the demands a cognitive test samples. If you want a rough sense of where your own reasoning sits, a structured measure like the Real World IQ test is built around those same everyday problem-solving skills rather than trivia.

Where emotional intelligence earns its keep

Emotional intelligence is not a myth — it just does a more specific job than the bestseller version promises. A widely cited meta-analysis by Ernest O'Boyle and colleagues, published in the Journal of Organizational Behavior in 2011, found that EI is meaningfully related to job performance and, importantly, adds predictive power on top of cognitive ability and personality. One synthesis of 99 studies covering nearly 18,000 employees put EI's operational validity for job performance at roughly 0.45, with self-report ability measures faring best.

The catch is the word "incremental." Once you already know someone's cognitive ability and basic personality, EI typically explains a modest additional slice of the variance — on the order of a few percentage points for outcomes like job satisfaction and commitment in the meta-analytic work of Chao Miao and colleagues (2017). Modest is not nothing, but it is a long way from "EQ matters twice as much as IQ," a claim the research has never supported.

The one place EI pulls ahead: emotional labor

The most useful finding is about fit. In the cascading model proposed by Dana Joseph and Daniel Newman (2010), emotional intelligence predicts performance most strongly in jobs with high "emotional labor" — nursing, sales, teaching, customer service, management — where reading a room and regulating your own reactions is the work. In a data-entry role, EI barely moves the needle. So the honest statement is not "EI predicts success," but "EI predicts success in people-heavy roles, and much less elsewhere."

The marshmallow test, and why the famous version oversells it

Self-control is often folded into emotional intelligence, and no study is quoted more often than Walter Mischel's marshmallow experiment, in which preschoolers who resisted a treat supposedly went on to better outcomes years later. The story is tidy. The replication is messier.

In 2018, Tyler Watts, Greg Duncan, and Haonan Quan revisited the design with a larger, more diverse sample in Psychological Science. They still found a link between a four-year-old's ability to wait and later achievement — but the effect was about half the size reported in the original studies, and it shrank by roughly two-thirds once they controlled for family background, early cognitive ability, and home environment. The lesson is not that self-regulation is worthless. It is that much of what looks like willpower is entangled with the circumstances a child grows up in. Single-trait explanations of success almost always overpromise.

Why "IQ vs EQ" is the wrong frame

High cognitive ability is an advantage, but it is not a guarantee. Lewis Terman's decades-long study of high-IQ children — the so-called "Termites" — is a useful corrective: the group did well on average, yet raw IQ did not neatly sort the eminent from the ordinary. Persistence, opportunity, and social skill all mattered.

The productive way to think about it is that intelligence and emotional skill are different tools for different problems. Fluid reasoning helps you crack a novel problem; emotional skill helps you get a team to act on the solution. Careers reward the combination, weighted by what the role actually demands. Framing them as rivals mostly sells books.

How to read your own scores without fooling yourself

A few honest guidelines follow from the evidence. Treat any single number — IQ or EI — as one data point, not a verdict. Prefer measures that ask you to solve problems over ones that only ask you to rate yourself, because self-report EI is easy to inflate. And match the trait to the target: if your goal is a role that lives or dies on relationships, emotional skill deserves real attention; if it is a role that lives or dies on analysis, reasoning ability does. If you want a baseline for the reasoning side, you can measure your own real-world problem-solving and treat the result as a starting line rather than a label.

FAQ

Is emotional intelligence more important than IQ?

Not according to the research. Cognitive ability is the stronger single predictor of job performance overall. Emotional intelligence adds a smaller, genuine boost that is largest in people-facing roles. Both matter; neither replaces the other.

Can you improve your emotional intelligence?

The emotion-management side of EI appears more trainable than raw cognitive ability, particularly the habits of noticing your reactions and pausing before responding. Gains tend to be gradual, and self-report questionnaires can overstate progress, so external feedback is useful.

Does a high IQ guarantee success?

No. High cognitive ability raises the odds of learning quickly and handling complex work, but outcomes also depend on motivation, opportunity, health, and social skill. Terman's long-running study of gifted children showed high IQ did not, by itself, sort out who became eminent.

What is the difference between ability EI and trait EI?

Ability EI is measured with performance tasks that have better-and-worse answers, such as identifying an emotion in a face. Trait or mixed EI is measured with self-report questionnaires that blend emotion skills with personality. Ability measures tend to relate more cleanly to cognitive ability; self-report measures are easier to inflate.

Which one should I focus on developing?

Match it to your goal. For analysis-heavy work, invest in reasoning and domain knowledge. For relationship-heavy work, the emotion-management skills inside EI carry more weight. Most careers reward a blend, so the answer is rarely "only one."

Sources: Schmidt & Hunter, Psychological Bulletin (1998); O'Boyle et al., Journal of Organizational Behavior (2011); Joseph & Newman, Journal of Applied Psychology (2010); Miao et al., Journal of Occupational and Organizational Psychology (2017); Watts, Duncan & Quan, Psychological Science (2018).