Emotional Intelligence

Emotional Intelligence – Nearly everything you need to know.

Emotional Intelligence is the ability to perceive, understand and influence your own and others’ emotions. By increasing your emotional intelligence you will be better equipped to guide your thinking and actions and be better prepared to achieve your goals. This article takes a critical look at the different models used for measuring emotional intelligence, the current state of research in the field, and suggests areas for improvement.    

  1. A few definitions
  2. What are the main emotional intelligence models?
  3. How has the emotional intelligence construct evolved over time?
  4. What are the main controversies, challenges and questions in the research field?
  5. What is the current thinking on emotional intelligence frameworks?
  6. How is emotional intelligence best measured?
  7. Summary

1 · A few definitions

As we build up a definition of emotional intelligence it might be helpful to look first at the concept of intelligence. The Oxford dictionary suggests that intelligence is “the ability to acquire and apply knowledge and skills”. This is an important root that is getting lost in the evolution of emotional intelligence, as we will explore later.

Intelligence has been strongly associated with educational institutions and universities and this has led to those who are judged to have a high IQ being described as academic and scholarly. Intelligence is seen by many as a general mental capability to perceive or understand – Charles Spearman applying the term ‘g-factor’ to the construct in 1925. There have been efforts in the 1970’s to break intelligence down further into fluid and crystalized intelligence by Raymond Cattell. Fluid intelligence (Gf) is about our ability to reason and think flexibly. Crystallized intelligence (Gc) refers to the accumulation of knowledge, facts, and skills that are acquired throughout life. There is a general acceptance that the tests for general intelligence, known as Intelligence Quotient (IQ) tests, can be used as a benchmark for children and adults for selection, recruitment, assessing potential, and so on. On the flip side, there is some comfort for the non-academics, who may have struggled through formal education process, when the dictionary suggests that one definition of ‘academic’ is ‘being of no practical use’. There is a great deal of interest in recognising individual strengths across a range of human qualities, rather than these be restricted to a pure academic framing. Howard Gardner, and others, have argued since the 1980’s that humans have multiple-intelligences, including visual-spatial, verbal-linguistic, logical-mathematical, and those gifted with musical (rhythmic and harmonic), naturalistic, bodily-kinesthetic, interpersonal and interpersonal intelligence. These last two categories are the core components of emotional intelligence.

A generic, widely accepted equivalent in emotional intelligence (EI) does not seem to exist. A working definition for emotional intelligence (EI) that we feel captures the array of definitions in the field is offered here:

Emotional Intelligence is the ability to perceive, understand and influence our own and others’ emotions, across a range of contexts, to guide our current thinking and actions, to help us to achieve our goals.

Lansley, 2017

2 · What are the main emotional intelligence models?

There are diverse models of EI; some are ability models, some trait models and some a mixture of both. This has to be resolved if we are to ever have what we term an ‘e-factor’ – a widely accepted framework and assessment method that allows EI to have a place alongside IQ.

Firstly, the difference between these approaches needs to be outlined:

  • Traits in their pure form are distinguishing qualities or characteristics, typically belonging to a person.
    Sometimes referred to as ‘personality’.
    Personality or traits have been defined as ‘habitual patterns of behaviour, thought, and emotion’ (Kassin, 2003). It’s how you might describe a person… introverted, optimistic, positive, hostile and so on. They are largely perception based and are usually measured by self-report or by reports and perceptions from those focusing on the person being profiled.

  • Ability is about being able to utilise knowledge, understanding and skills to perform successfully or effectively.
    Sometimes referred to as ‘competence’.
    This is often measured in professional settings using applied knowledge tests and simulations, as you would see with trainee air pilots, leaders, lawyers and brain surgeons.

  • A Mixed model often draws from both these frameworks.
    This approach can make it difficult if we try to measure it using a single assessment instrument.

This grid is a good early summary of the key factors around trait and ability models: (Furnham, 2009:143)

3 · How has the emotional intelligence construct evolved over time?

There have been many research developments over the last century, and this has accelerated in the last twenty years. Here is a walkthrough the timeline with the key developments highlighted and referenced for you.

  1. Thorndike
    EI is rooted in the century-old concept of social intelligence which Edward Thorndike positioned alongside abstract intelligence and mechanical intelligence. Gardner was keen to challenge the widespread belief by psychologists that “intelligence is a single faculty and one is either ‘smart’ or ‘stupid’ across the board” (Gardner, 2000:34).
  2. Payne
    The first use of the term emotional intelligence is often attributed to Wayne Payne’s doctoral thesis (1985).
  3. Bar-On
    Reuven Bar-On claims to have coined the EQ term in the late 1980’s to describe his approach to assessing emotional and social competence in his unpublished doctoral thesis. His model is best referred to as a mixed model (he prefers to call it an array) as it consists of fifteen factors (Bar-On, 2006:23) including skills (e.g. assertiveness and problem solving), traits (e.g. optimism), and other qualities (stress tolerance, social responsibility and self-actualisation).
  4. Mayer-Salovey-Caruso
    A landmark article was produced by Peter Salovey and John Mayer (1990:190) under three factors (focused on self and other) of utilization of emotion, regulation of emotion, and appraisal and expression of emotion. This was the adapted to what is now widely known as the four-branch model (Mayer and Salovey, 1997:11). Their model stayed true to the concept of ability and an assessment tool created named MSCEIT (Mayer-Salovey-Caruso Emotional Intelligence Test). This model led the way with ability EI, though not without its critics. Most criticisms are around the assessment methodology, the comprehensiveness of the framework and the relevance of the eight ‘tasks’ that assess them.
  5. Petrides and Furnham
    The distinction between trait EI and ability EI was highlighted by Petrides and Furnham (2000:314). Trait models gathered support due to their higher correlation with personality instruments. This logic is fine on the surface, though it is flawed if you accept the argument that EI is an intelligence. It is problematic to argue that intelligence should be based on an evolved or experience-created preference.
  6. Goleman
    In 1995 the term Emotional Intelligence became more widely recognized following the publication of a book by Daniel Goleman (1995). Goleman presented twenty-five ‘competencies’ which are a mix of skills, abilities, traits and attitudes, earning it the same mixed model label as Bar-On’s (Goleman, 1998), though Goleman centered his work on leadership in the workplace. Goleman’s 1995 book’s best-selling status, along with his follow up text (1998), may be the reason that the EI term gained such popularity; Goleman deserves, and gets, credit for this.

4 · What are the main controversies, challenges and questions in the research field?

Across the main EI models we summarise seven key statements, questions or arguments that arise:

  1. Traits and abilities have been accepted as interrelating/opposing constructs and therefore should not be mixed as primary factors in an EI model. This is reinforced by Petrides, the developer of Trait Emotional Intelligence Questionnaires, who argues that, if trait EI is a personality construct, “one should not expect it strongly to correlate with measures of psychometric (IQ) intelligence” (Petrides and Furnham, 2001:437).
  2. There are claims that the strength of a model depends on a positive correlation between EI and IQ since they measure the same overall construct of intelligence. This might be flawed. IQ and EI may not correlate positively when properly constructed and assessed.
  3. EI needs to factor in appropriateness of behavior, within the micro and macro context towards our goals. That contextual influence includes recognition of the time and place of any interaction, plus the wider cultural, ethnicity, personality, and other individual differences.
  4. A generic e model would need to be context independent; free from restrictive, contaminating frames such as well-being, leadership, management, team building, worker, and organizational awareness to enable independent application across contexts.
  5. There may be assumptions that a high EI means we are a nice person, and vice versa. This is a reasonable assumption, though there is a dark-side to EI, as there is in leadership – those with high level abilities can use or abuse those skills by applying them for personal gain. There is a fine line between constructive influence and self-serving manipulation. It could be argued that openness, wellbeing, win-win intent, and other prosocial qualities, may be socially desirable, but should not feature in an EI framework.
  6. Assessment methodology needs to be fit for purpose. Self-report, multi-rater, consensus and expert scoring all have their strengths and flaws. Also, we need to ensure ecological (real-world) validity for the varied environments we face in life and work; attention is needed towards the idea that measures need to challenge learners and test-takers when things are emotionally charged and when stakes are high.
  7. There are diverse opinions regarding which factors should feature at the core of an EI framework – traits, abilities, or a mixture of these.

This last issue is crucial in deciding a way forward.

5 · What is the current thinking on emotional intelligence frameworks?

Most accept that EI can make a difference, but the models and assessment methodology used to quantify EI are simply not good enough to form the basis of a generic model. The development of a generic model will be a vital step forward in seeking to provide a platform for future studies that can stand up to objective scrutiny. The current controversies and the challenges of EI centre on this key question:

What factors should be primary in an Emotional Intelligence taxonomy – traits, abilities or a mix of these?

If we accept that the EI model needs developing so as to be classed as an intelligence construct it would need to be generic. And since intelligence is defined across the dictionaries as “the ability to acquire and apply knowledge and skills” it follows that ability/competency would need to be at the core of the model.

Context will need to be integrated throughout; competent application of EI ability needs to be appropriate to the macro and micro context of that application. It needs to set aside the constructive or destructive motives of those who may use emotional skills and competencies. This means that directive factors such as “moderating negative emotions and enhancing pleasant emotions” (from the MSCEIT model), would need to be excluded. EI models should not force the values of the framework developer onto the ones who seek to use them, however worthy.

This problem is minor, relative to the challenges of the trait-based and mixed models. The trait models include factors such as optimism, commitment, impulsiveness, and self-esteem, and they mix in skills such as emotional self-awareness, self-control and empathy. We could argue, then, that these models should all be labelled as mixed models, as they are all contaminated with skills or competencies. This is a challenge to those who wish to measure these different qualities reliably with one measurement tool and explains one of the reasons why most trait model developers resort to self-perception tests.

If we wish to classify emotional skills and competencies as an intelligence, then we have to use ability criteria at the core.

We are researching and developing a model at the Emotional Intelligence Academy (www.eiagroup.com) that focuses purely on those components which meet the following criteria:

  1. Fit our definition of Emotional Intelligence.
  2. Generic and capable of applying across contexts so need to be uncontaminated by a focus on specific occupations or societal roles (leadership/teams/work).
  3. Neutral of value judgements about what a person’s goals might be.
  4. Drawn from models based on good research.
  5. Allow users to recognise and incorporate the contribution of other constructs such as traits, personality, mindsets, attitudes, values and other qualities that are important to themselves or their social/organizational entity.

This model is undergoing peer-review early in 2019.

6 · How is emotional intelligence best measured?

Some models and tools, as we have seen, combine several types of EI qualities. Here are some examples:

• Knowledge and Understanding (Emotion triggers).
• Skills (Empathy).
• Competencies (Ability to read facial expressions).
• Traits (Optimistic).
• Attitudes (Service orientation).
• Other Qualities (Inspirational).

It is very difficult to measure, if not impossible, to measure all these aspects of EI with one instrument.
The ‘types’ of tests used can be summarized as follows:

  • Self Report – A questionnaire where you score your perceptions of your own EI against the criteria in the tool.
  • 360 Degree – A questionnaire that is completed by those who experience your performance – usually at work. Often including line manager, peers and direct reports. They score their perceptions (usually anonymously) of their perceptions of your EI against the criteria in the tool.
  • Consensus Scoring – Based upon the agreement of a large number of people. For example, if 70 percent of people felt that a photo was of a very happy person, then the best answer for the photo would be “happiness”.
  • Expert Scoring: Raw – In the expert method, subject matter experts (SME’s) determine which test answers are correct (right/wrong). Some weight the answers (better/worse). Raw scores are used so if you get maximum correct responses you score 100%.
  • Expert Scoring: Percentile – As above though all those who have taken the test are scored as percentiles, with the highest scorer at 100% and the lowest at 0% (even though the raw range may be from 17% to 93%).

Then there is the actual design of the instrument, how the data is collected from the one being measured.
These vary again and usually fall into one or more of the following groups:

  • Knowledge and Understanding Questionnaires

    Which can be pen and paper exercises, online or even via a personal interview.
    They can be:

    1. Right/wrong answers (tick-box style).
    2. Weighted answers (more points for better answers).
  • Ability Tests

    By facing the user with tasks to see how they apply their knowledge and understanding.
    They include:

    1. Simulations (as good as the simulation replicates the context where the person is likely to apply skills in real life).
    2. Assignments (tasks and exercises to test the application of knowledge and understanding – can be hard to replicate more challenging contexts we face in interpersonal interactions).
    3. Situational judgement tests (based on the notion that a competent person can recognise competence being performed when they see/hear it).
    4. Observation of you performing (intrusive and contaminates the context, plus a great deal of EI happens inside the mind and is not outwardly observable).
  • Psychometric Profiling

    Where preferences or choices along a scale or between several options are captured and then correlated against broader themes, qualities and traits into a synthesizing report. These can be presented as raw scores or as percentiles against the population who have already completed the test.

Some of these are self/automatically scored at the end of the test.
Others are controlled by the instrument administrators who qualify to interpret results and give feedback to users.

6.1 · Ability tools

If you wish to test competence and skills then it makes sense to explore the ‘ability’ instruments where you can test yourself against the (expert scoring) right answers, determined by subject matter experts and excellent EI practitioners, or slightly more controversially, against the database of respondents (consensus scoring). You also need to determine if the tasks you perform in the assessment test the abilities that will prepare you for the context where you plan to apply your EI.

The main argument against ability models, usually against the MSCEIT tool, is that even though someone knows how he/she should behave in an emotionally laden situation, it doesn’t necessarily follow that the person could actually carry out the reported behaviour. To be fair to the MSCEIT tool, the developers do try to address this by using a range of tasks, but they accept the assessment is not free of technical issues and they continue to improve it.

6.2 · Trait tools

Most of the trait-based tools are self-report. Some believe that self-perception is not reliable enough unless the tool is just for personal reflection. If your EI is not excellent then it might be very hard to judge how you perform in reading your own emotions and managing them; this kind of thinking gets drowned out during emotional episodes. It’s equally challenging to judge ourselves on how we read and interact with others from an EI standpoint.
Those working in the more challenging areas of human behaviour have concerns with the “reliability of self-report about mental states (such as emotions). Although much of the research on emotion has presumed that research subjects and our patients during psychotherapy can readily report on their subjective experience through questionnaires and interviews, findings to date show that most people report only the most recent or most intense of their emotional experiences (e.g., Kahneman, Fredrickson, Schreiber, & Redelmeier, 1993; Rosenberg & Ekman, 1994) and are subject to bias.”
We are all loaded with biases, preferences and passion that can filter how we see ourselves when dealing with others. The prominent view in the scientific literature is the trait EI view, re-interprets EI as a collection of personality traits. Claiming they are merely preferences and are not context specific.
What, unfortunately, happens in practice is many tool designers tend to default to the simple, cheap, quick self-report (some addition muti-rater feedback/ratings from others) and suffer the criticisms around these approaches.

A selection of EI measurement tools can be found on the CREIO website (www.eiconsortium.org) which includes the ones below (linked directly through to the site):

• BarOn Emotional Quotient Inventory
• Emotional & Social Competence Inventory
• Emotional & Social Competence Inventory – University Edition
• Genos Emotional Intelligence Inventory
• Group Emotional Competency Inventory
• Mayer-Salovey-Caruso EI Test (MSCEIT)
• Schutte Self Report EI Test
• Trait Emotional Intelligence Questionnaire (TEIQue)
• Work Group Emotional Intelligence Profile
• Wong’s Emotional Intelligence Scale

Inclusion here is not an endorsement by The Emotional Intelligence Academy. In fact, CREIO themselves make it clear on their site that “Many tests that promise to measure emotional intelligence have appeared in recent years. Some of these tests seem promising, but many have not been empirically evaluated. As a service to our visitors, we have reviewed many of these tests and selected those for which there is a substantial body of research (at least five published journal articles or book chapters that provide empirical data based on the test). However, the inclusion of a test on this web site does not constitute an endorsement of that test by CREIO”.

6.3 · The way forward?

We are not there yet. But there is a great deal of interest in getting to this e-factor from end users, recruiters, academics and other professionals.

Ability has to be at the core for EI to be considered as an intelligence, though traits and personality must feature within an EI model – and indeed is central to the majority of the popular EI models in existence. Our personality might be defined by our idiosyncratic biases, scripts, preferences, styles, experiences, values, beliefs, motivations and there may be other clinical, cultural and biological differences that need to be considered so we can better understand ourselves and manage any contamination such differences might bring to our thinking and behaviour. We need to work out and consider such differences in ourselves and others in any critical interactions with them for two reasons: firstly, to help us better understand what others feel and value and, secondly, to help us adopt a selfless, empathic approach in flexing to others’ preferences to help us engage, explore and understand others to enable productive discourse towards our (and/or their) goals.

Taxonomies and psychometric models or instruments may have some value in labelling and differentiating human traits and personality types though there is a risk in using these labels as shortcuts. We are cognitive misers and such heuristics can be used to over-generalise and categorise individuals for convenience. They are merely simple, often efficient rules which are used to form judgments and make decisions. They are mental shortcuts that can steer us to focus on one aspect of self or others and ignore the rest.

Trait or personality labels can be welcomed and used by some individuals to deal with activities or behaviour that they are keen to sustain or avoid, whilst offering an option to absolve themselves of responsibility (“I am an introvert so please ask someone else to do the business networking”; “I am shy so please don’t ask me to do the public presentation”; “I am not an ENTJ so I will never make a CEO”).

So, if trait labels can be misused and are often argued as being long term, or even fixed (i.e. cannot be ‘developed’) then integration of traits into an ability model needs to be done with care. If we can adopt traits without permanently labelling ourselves, and/or those we deal with, then we suggest knowledge of traits is crucial. The value of using such heuristics can help us to define and manage our own biases and preferences when dealing with others so we can connect, engage and build rapport in the early stages of an interaction by flexing our preferences to what we recognise (during an interaction) as the preferences of others.

7 · Summary

There are some exciting research projects going on at the moment at The Emotional Intelligence Academy, and elsewhere, some yet unpublished. This article hopes to capture some of the key issues and remind us all to stay focused on the core of Emotional Intelligence and towards the possibility of an e-factor.

We welcome any feedback and input as we do our small part in moving towards an IQ equivalent for EI with adults this year, with a solid model and a reliable instrument that measures what it says it measures.

Our plan is then to adapt this research for children so we can help add emotional skills to the core curriculum of reading, writing and mathematics, to help children survive, thrive and make a positive difference for their own wellbeing and in their current and future communities.

Further Information


Bar-On, R. (2006) ‘The Bar-On model of emotional-social intelligence (ESI) 1.’ Psicothema, 18(Suplemento) pp. 13–25.

Furnham, A. (2009) ‘The importance and training of emotional intelligence at work.’ In Assessing Emotional Intelligence. Springer, pp. 137–155.

Gardner, H. E. (2000) Intelligence Reframed: Multiple Intelligences for the 21st Century. Hachette UK.

Goleman, D. P. (1995) ‘Emotional intelligence: Why it can matter more than IQ for character, health and lifelong achievement.’

Goleman, D. P. (1998) ‘Working with Emotional Intelligence and Emotional Intelligence.’

Kassin, S. M. (2003) Essentials of psychology. Prentice Hall.

Lansley, C. (2017) ‘Emotional Intelligence. Diploma in Behavior Analysis and Investigative Interviewing Module.’

Mayer, J. D. and Salovey, P. (1997) ‘What is emotional intelligence.’ Emotional development and emotional intelligence: Educational implications, 3 p. 31.

Payne, W. L. (1985) A Study of Emotion: Developing Emotional Intelligence; Self-Integration; Relating to Fear, Pain and Desire. PhD Thesis. Dissertation.

Petrides, K. . and Furnham, A. (2001) ‘Trait emotional intelligence: Psychometric investigation with reference to established trait taxonomies.’ European journal of personality, 15(6) pp. 425–448.

Petrides, K. V. and Furnham, A. (2000) ‘On the dimensional structure of emotional intelligence.’ Personality and Individual Differences, 29(2) pp. 313–320.

Salovey, P. and Mayer, J. D. (1990) ‘Emotional Intelligence.’ Imagination, Cognition and Personality, 9(3) pp. 185–211.

Cliff Lansley
Article by Cliff Lansley

Expert in emotional intelligence, behavioural analysis and high stake deception detection contexts. Cliff holds; B Ed (Hons), MIOD, MABPsych, Cert Ed.

Training & development in the social sciences

Further your knowledge and skills with our range of training programs teaching foundation skills up to degree level qualifications.

View Courses

Get in Touch

If you would like to find out more please call us on +44 (0) 161 413 8930 or fill in the form below:

© Copyright 2009-2022 • Emotional intelligence Academy Limited • All Rights Reserved