EQ vs IQ: what is the difference?

Short answer: IQ and EQ measure different things. An IQ score comes from an intelligence test of verbal and nonverbal tasks with scored answers, scaled to an average of 100 and a standard deviation of 15 (APA task force report, 1996). EQ, or emotional intelligence, is skill with emotions: perceiving, using, understanding and managing them (Mayer and Salovey's four-branch model, 1997). It is measured by an ability test or by a self-report questionnaire, and the two kinds agree weakly at best. For school grades, a 2020 meta-analysis ranked emotional intelligence third, after intelligence and conscientiousness (MacCann and colleagues).

EQ vs IQ at a glance

• What it is. IQ: an overall score on an intelligence test (APA task force report, 1996). EQ: either a set of emotional abilities or a mix of self-rated traits, depending on the researcher (Mayer, Salovey and Caruso, 2008).
• How it is measured. IQ: verbal and nonverbal items, scored to give subscores and an overall score (APA task force report, 1996), on tests such as the WAIS-5, from Pearson. EQ: an ability test such as the MSCEIT 2, or a questionnaire such as the EQ-i 2.0 or the TEIQue, per MHS and O'Connor and colleagues (2019).
• Score format. IQ: average 100, standard deviation 15 (APA task force report, 1996). EQ: the MSCEIT 2 and the EQ-i 2.0 use the same 100 and 15 (Mayer and colleagues, 2025; Shahid and Adams, 2020), but the numbers are not interchangeable with IQ.
• How reliable. IQ: the WAIS-5 FSIQ has a mean reliability of 0.97 (Winter, Trudel and Kaufman, 2024). EQ: 0.88 for the MSCEIT 2 total in its makers' own study (Mayer and colleagues, 2025). The EQ-i questionnaire reaches .97 (Bar-On, 2006), but a consistent self-rating is not tested ability. Self-report questionnaires can also be faked, as a study of the TEIQue short form showed (Walker and MacCann, 2024).
• Overlap. IQ and tested EQ correlate about 0.25 to 0.35 across meta-analyses, from 0.25 (Joseph and Newman, 2010) to 0.29 and 0.35 in a 2021 meta-analysis (Bryan and Mayer). Self-reported scores barely overlap with IQ, by Bar-On's account (2006).
• Stability and change. IQ: people keep roughly their rank over five years in adulthood (Breit and colleagues, 2024). EQ: training meta-analyses report moderate gains (Mattingly and Kraiger, 2019), though a 2024 healthcare review found a high risk of bias in many studies (Powell and colleagues).
• What it predicts. IQ: school grades and job performance, with the job estimate revised down in 2022 (Sackett and colleagues). EQ: grades a little beyond IQ and personality (MacCann and colleagues, 2020). For tested EQ, job performance mainly in emotionally demanding jobs (Joseph and Newman, 2010).

What IQ is and how it is measured

IQ stands for intelligence quotient. The overall score on an intelligence test is conventionally "converted to a scale in which the mean is 100 and the standard deviation is 15" (Intelligence: Knowns and Unknowns, APA task force report, 1996). The current adult clinical test is the WAIS-5, published by Pearson in 2024 for ages 16 to 90.

Its full-scale IQ has a mean reliability of 0.97, according to Winter, Trudel and Kaufman (2024), citing the WAIS-5 manual. That reliability implies a 95% range of roughly 5 points either side of a measured score (our calculation), and scores can differ between tests and between sittings. There is more in our guide to the WAIS IQ test, the Stanford-Binet test, are IQ tests accurate? and what an IQ test measures.

IQ tests also leave things out. The APA report says: "It is widely agreed that standardized tests do not sample all forms of intelligence. Obvious examples include creativity, wisdom, practical sense and social sensitivity; there are surely others."

What emotional intelligence (EQ) is: ability vs self-report models

The academic idea starts with Salovey and Mayer (1990). They described abilities to "monitor one’s own and others’ feelings and emotions, to discriminate among them and to use this information to guide one’s thinking and actions" (as quoted by Hughes and Evans, 2018).

Mayer and Salovey's four-branch model (1997) splits this into perceiving emotions, using emotions to help thinking, understanding emotions, and managing them (Herpertz and colleagues, 2016). The current ability test, the MSCEIT 2, reports a total plus Perceiving, Connecting, Understanding and Managing Emotions, per MHS (read 13 October 2026).

Goleman's 1995 book, subtitled "Why it can matter more than IQ", popularized a broader version that added "motivation, empathy, social skills, happiness, and achievement-orientation" (Hughes and Evans, 2018). Mayer, Salovey and Caruso (2008) describe a split between EI as mental abilities and EI as "an eclectic mix of positive traits such as happiness, self-esteem, and optimism."

O'Boyle and colleagues (2011) sort EI measures into three streams:

• Ability EI: tests with right and wrong answers, such as the MSCEIT.
• Self-report EI: questionnaires built on the four-branch model, where you rate yourself. A related view, trait EI, treats emotional intelligence as "emotional self-perceptions" within personality (Petrides, 2010, as quoted by Hughes and Evans) and is measured with questionnaires such as the TEIQue.
• Mixed EI: broad models of emotional competencies, such as Goleman-style measures.

Where does the Q come from? Reuven Bar-On, who built the EQ-i, says he coined EQ, for emotional quotient, in 1988 and scaled EQ-i scores to resemble IQ scores (Bar-On, 2006). We found no independent source confirming it.

Is EQ measured like IQ?

Only partly. O'Connor and colleagues (2019) describe ability EI measures as "tests that utilize questions/items comparable to those found in IQ tests". The MSCEIT 2 is one: a "Performance-based assessment of EI" for adults that takes about 30 minutes (MHS, read 13 October 2026).

The big difference is how a right answer is decided. IQ tests ask people to give the meanings of words or say which word does not belong with the others, and their performance "can then be scored to yield several subscores as well as an overall score" (APA task force report, 1996). For the original MSCEIT, its authors recommended "a scoring system that employed a consensus among experts" (Mayer and colleagues, 2025). The MSCEIT 2 uses a "New veridical scoring system, designed by a multidisciplinary and multicultural panel of emotion research experts" (MHS).

The other kind of EQ test is a questionnaire. The EQ-i 2.0 is "an online 133-item self-assessment instrument that uses a 5-point Likert scale to measure EI across five composite scales and 15 content subscales" (Shahid and Adams, 2020).

• Self-ratings do not track tested ability. In the MSCEIT 2 normative study, a self-estimated EI scale correlated −0.07 with the test (Mayer and colleagues, 2025).
• They can be faked. Walker and MacCann (2024) tested the TEIQue short form, a trait EI questionnaire. Scores were higher under a simulated job-selection instruction than under an instruction to answer honestly: "We conclude that people can fake emotional intelligence".

Sharing the 100 and 15 scale does not make EQ and IQ scores comparable, or even two EQ tests. In Mayer and colleagues' 2025 comparison study, the same people averaged 114 on the MSCEIT 2 and 104 on the original MSCEIT.

Precision differs too. The MSCEIT 2 makers' worked example gives a score of 130 a 95 percent range of "130 +/− 14.7" (Mayer and colleagues, 2025). At that score, the range is about three times wider than the roughly 5 points around a WAIS-5 FSIQ (our calculation). The test is more precise lower down: the same paper gives about 8.8 points either side at a score of 85.

How much do EQ and IQ overlap?

It depends which EQ you mean.

Tested EQ overlaps modestly with IQ. A meta-analysis of 87 studies found correlations of 0.29 between emotional intelligence and fluid intelligence, and 0.35 with comprehension knowledge (Bryan and Mayer, 2021). An earlier meta-analysis put the overall figure at 0.25 (Joseph and Newman, 2010). Squared, correlations of 0.25 to 0.35 mean roughly 6 to 12 percent shared variance (our calculation). Hughes and Evans (2018) conclude that ability EI "can be considered a sub-factor of general cognitive ability". For the reasoning side, see fluid intelligence.

Self-reported EQ barely overlaps with IQ, but overlaps with personality and self-ratings. Bar-On (2006) reports "a 4% overlap with IQ tests and a 15% overlap with personality tests" for the EQ-i. A 2015 meta-analysis found that mixed EI scores are largely explained by things already measured: ability EI, self-efficacy, self-rated performance, three Big Five traits and general mental ability together (multiple R = .79; Joseph, Jin, Newman and O'Boyle).

So a high IQ with an average or low EQ score, or the reverse, is common.

Which matters more at school and at work?

At school, IQ is the stronger predictor. Roth and colleagues (2015) pooled 240 samples with 105,185 participants in a meta-analysis and found a corrected correlation of .54 between intelligence tests and grades.

Emotional intelligence adds a little. MacCann and colleagues (2020) found an overall correlation of .20 between EI and academic performance, across 42,529 people, stronger for ability EI. After controlling for intelligence and Big Five personality, ability, self-rated and mixed EI explained "an additional 1.7%, 0.7%, and 2.3% of the variance, respectively." The authors note "the lack of evidence for a causal direction".

At work, the IQ estimate has come down. Schmidt and Hunter's 1998 estimate for general mental ability was 0.51, as Richardson and Norgate (2015) report. Sackett, Zhang, Berry and Lievens (2022) argued that earlier corrections caused "substantial overcorrection" and put the revised estimate at .31 operational validity.

For emotional intelligence at work, the kind of measure matters:

• Ability EI helps in some jobs and not others. Joseph and Newman (2010) found it "positively predicts performance for high emotional labor jobs and negatively predicts performance for low emotional labor jobs".
• Mixed EI looked strong, then shrank. Earlier reviews put its link to job performance at .47 (Joseph and Newman, 2010). A 2015 update put its link to supervisor-rated performance at .29, which "becomes nil" once ability EI, personality, self-efficacy, self-rated performance and general mental ability are controlled (Joseph and colleagues, 2015).

O'Boyle and colleagues' 2011 meta-analysis put all three kinds of EI at corrected correlations of 0.24 to 0.30 with job performance, with self-report and mixed EI adding the most beyond cognitive ability and the Big Five. Those EI figures predate the 2022 revision of correction methods, so they cannot be compared like for like with the .31 estimate for cognitive ability.

Can you raise your IQ or your EQ?

IQ is fairly stable in adulthood. A meta-analysis pooled 205 longitudinal studies with 87,408 participants, from preschool children to late adulthood, and found stability "consistently high" from late adolescence on (Breit and colleagues, 2024). For a 20-year-old retested five years later, its model put rank-order stability at .76 on average across cognitive abilities, and .80 for general ability. Schooling can move scores: Ritchie and Tucker-Drob (2018) found "approximately 1 to 5 IQ points for an additional year of education". Those studies measured years of schooling, so they do not show that a course taken as an adult raises IQ.

EQ training studies report moderate gains, with caveats. Mattingly and Kraiger (2019) reviewed 58 studies and found "a moderate positive effect for training, regardless of design". A 2024 meta-analysis of workplace training in emotional intelligence, empathy or emotion regulation found effects that "persisted more than three months after the training end" (Mehler and colleagues). That set is broader than EI alone, and some of its authors offer such trainings. A 2024 healthcare review warns that "the true effect of EI intervention across these methodologies is much smaller than reported" (Powell and colleagues). Mayer, answering a reader's question in Scientific American Mind (2014), is cautious: psychologists "still are not sure whether adults can enhance their emotional intelligence", though "people can almost surely increase their emotional competence".

Where the popular "EQ beats IQ" numbers come from

• Goleman. In 1998 he wrote that "IQ and technical skills are important, but emotional intelligence is the sine qua non of leadership". Hsu, Newman and Badura (2022) call his explanation "conceptual (and non-empirical)".
• TalentSmartEQ's percentages. TalentSmartEQ, a company that sells EQ training, says emotional intelligence explains most of job success and that most top performers score high on it (company article, 2022). The page names no study, sample or method. It also says "There is no known connection between IQ and emotional intelligence", which the meta-analyses above contradict.
• The CareerBuilder survey. In a 2011 poll of more than 2,600 hiring managers and human resource professionals, "Seventy-one percent said they value emotional intelligence in an employee more than IQ" (a CareerBuilder press release). It records what employers say they prefer, not what predicts performance.
• "Five times more variance." Bar-On (2006) wrote that "EQ accounts for about five times more variance than IQ", comparing his own EQ-i studies with someone else's meta-analysis. Later independent meta-analyses do not support it.

In the same 2014 reply, Mayer, who helped define emotional intelligence, wrote that some pundits say it "explains more than 85 percent of outstanding performance in top leaders", and added: "My colleagues and I have never found such claims to be true."

Which test fits your question

If you want to know about your emotional skills, an ability test such as the MSCEIT 2 measures performance, while a questionnaire records how you see yourself. If your question is about a diagnosis, school support or a legal matter, talk to a licensed psychologist.

If you want a quick reading of the IQ side, WhatIQ's test uses 24 timed picture puzzles in the matrix format Raven published, takes about 15 minutes, and gives a score on the standard IQ scale with the range it most likely sits in (whatiq.com, read 13 October 2026). It measures reasoning only: it does not measure emotional intelligence and is not a clinical assessment. To read the number afterward, see what the average IQ is.

Frequently asked questions

Which is better to have, a high IQ or a high EQ?

Neither replaces the other. For school grades, intelligence is the stronger predictor, and emotional intelligence ranked third after intelligence and conscientiousness in a 2020 meta-analysis (MacCann and colleagues). At work, a 2022 revision put the operational validity of cognitive ability at .31 (Sackett and colleagues), while tested EQ helps mainly in emotionally demanding jobs (Joseph and Newman, 2010).

Is there a standard EQ score like IQ?

By convention, yes, but the numbers are not interchangeable. The MSCEIT 2 ability test and the EQ-i 2.0 questionnaire both scale scores to an average of 100 and a standard deviation of 15 (Mayer and colleagues, 2025; Shahid and Adams, 2020). Yet in Mayer and colleagues' 2025 study, the same people averaged 114 on the MSCEIT 2 and 104 on the original MSCEIT, so even two EQ tests disagree.

Can you have a high IQ but low EQ?

Yes, and it is common. Tested emotional intelligence correlates only about 0.25 to 0.35 with intelligence across meta-analyses, from 0.25 (Joseph and Newman, 2010) to 0.29 and 0.35 in a 2021 meta-analysis (Bryan and Mayer). Squared, that is roughly 6 to 12 percent shared variance (our calculation). Self-reported scores overlap with IQ even less: Bar-On (2006) reports a 4% overlap for his EQ-i.

Can EQ be learned, and can IQ change in adulthood?

Training studies report moderate gains on EI measures (Mattingly and Kraiger, 2019; Hodzic and colleagues, 2018), though a 2024 review of healthcare training warns the true effect is likely smaller (Powell and colleagues). In adulthood, IQ rank is fairly stable over five years (Breit and colleagues, 2024). Each extra year of schooling adds about 1 to 5 points (Ritchie and Tucker-Drob, 2018), but that evidence comes from schooling, not adult courses.

Is there an IQ and EQ test?

No single test does both well, because they need different kinds of items. IQ tests such as the WAIS-5 use verbal and nonverbal tasks, scored to give an overall score (APA task force report, 1996). EQ is measured with an ability test such as the MSCEIT 2 or a questionnaire such as the EQ-i 2.0 (MHS). Emotional intelligence needs its own kind of test.

Why do people say EQ matters more than IQ?

Goleman's 1995 book, subtitled "Why it can matter more than IQ", is the main origin. Researchers call Goleman's argument "conceptual (and non-empirical)" (Hsu and colleagues, 2022), and Mayer (2014) says he and his colleagues have never found such claims to be true.