Part 2: The Ability Model and Measurement, The Mixed Model and Measurement, and The Trait Model and Measurement

The Ability Model

Salovey and Mayer’s conception of EI strives to define EI within the confines of the standard criteria for a new intelligence (Mayer et al., 2001; MacCann et al., 2014).

Following their continuing research, their initial definition of EI was revised, however after pursuing further research, their definition of EI evolved into “the capacity to reason about emotions, and of emotions, to enhance thinking…

…It includes the abilities to accurately perceive emotions, to access and generate emotions so as to assist thought, to understand emotions and emotional knowledge, and to reflectively regulate emotions so as to promote emotional and intellectual growth.” (Mayer and Salovey, 1997).

The ability based model views emotions as useful sources of information that help one to make sense of and navigate the social environment (Mayer and Salovey, 1997; Salovey and Grewal, 2005).

The model proposes that individuals vary in their ability to process information of an emotional nature and in their ability to relate emotional processing to a wider cognition.

This ability is seen to manifest itself in certain adaptive behaviours.

The model claims that EI includes four types of abilities:

Perceiving emotions – the ability to detect and decipher emotions in faces, pictures, voices, and cultural artifacts—including the ability to identify one’s own emotions.

Perceiving emotions represents a basic aspect of emotional intelligence, as it makes all other processing of emotional information possible.

Using emotions – the ability to harness emotions to facilitate various cognitive activities, such as thinking and problem solving.

The emotionally intelligent person can capitalise fully upon his or her changing moods in order to best fit the task at hand.

Understanding emotions – the ability to comprehend emotion language and to appreciate complicated relationships among emotions.

For example, understanding emotions encompasses the ability to be sensitive to slight variations between emotions, and the ability to recognise and describe how emotions evolve over time.

Managing emotions – the ability to regulate emotions in both ourselves and in others. Therefore, the emotionally intelligent person can harness emotions, even negative ones, and manage them to achieve intended goals.

The ability EI model has been criticised in the research for lacking face and predictive validity in the workplace (Bradberry and Su, 2003). However, in terms of construct validity, ability EI tests have great advantage over self-report scales of EI because they compare individual maximal performance to standard performance scales and do not rely on individuals’ endorsement of descriptive statements about themselves (Brackett and Mayer, 2003).


The current measure of Mayer and Salovey’s model of EI, the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) is based on a series of emotion based problem solving items (Salovey and Grewal, 2005; Mayer et al., 2003).

Consistent with the model’s claim of EI as a type of intelligence, the test is modeled on ability based IQ tests. By testing a person’s abilities on each of the four branches of emotional intelligence, it generates scores for each of the branches as well as a total score.

Central to the four branch model is the idea that EI requires attunement to social norms.

Therefore, the MSCEIT is scored in a consensus fashion, with higher scores indicating higher overlap between an individual’s answers and those provided by a worldwide sample of respondents.

The MSCEIT can also be expert scored, so that the amount of overlap is calculated between an individual’s answers and those provided by a group of 21 emotion researchers (Salovey and Grewal, 2005).

Although promoted as an ability test, the MSCEIT is unlike standard IQ tests in that its items do not have objectively correct responses.

Among other challenges, the consensus scoring criterion means that it is impossible to create items (questions) that only a minority of respondents can solve, because, by definition, responses are deemed emotionally “intelligent” only if the majority of the sample has endorsed them.

This and other similar problems have led some cognitive ability experts to question the definition of EI as a genuine intelligence – (Pfeiffer, 2001) suggested that EI should be tentatively viewed as only a possible form of intelligence, awaiting further validation and development.

In a study by Føllesdal, the MSCEIT test results of 111 business leaders were compared with how their employees described their leader (Føllesdal, 2008).

It was found that there were no correlations between a leader’s test results and how he or she was rated by the employees, with regard to empathy, ability to motivate, and leader effectiveness.

Føllesdal also criticised the Canadian company Multi-Health Systems, which administers the MSCEIT test.

The test contains 141 questions but it was found after publishing the test that 19 of these did not give the expected answers.

This has led Multi-Health Systems to remove answers to these 19 questions before scoring but without stating this officially.

Other measurements:

Various other specific measures have also been used to assess ability in emotional intelligence and they include:

  • Diagnostic Analysis of Non-verbal Accuracy (Mayer, 2008) – The Adult Facial version includes 24 photographs of equal amount of happy, sad, angry, and fearful facial expressions of both high and low intensities which are balanced by gender. The tasks of the participants is to answer which of the four emotions is present in the given stimuli (Mayer, 2008).
  • Japanese and Caucasian Brief Affect Recognition test (Mayer, 2008) – Participants try to identify 56 faces of Caucasian and Japanese individuals expressing seven emotions such happiness, contempt, disgust, sadness, anger, surprise and fear, which may also trail off for 0.2 seconds to a different emotion (Mayer, 2008).
  • Levels of Emotional Awareness Scale (Mayer, 2008) – Participants read 26 social scenes and answers their anticipated feelings and continuum of low to high emotional awareness (Mayer, 2008).

The Mixed model

The model introduced by Daniel Goleman (Goleman, 1998) focuses on EI as a wide array of competencies and skills that drive leadership performance. Goleman’s model outlines five main EI constructs (for more details see “What Makes A Leader” by Daniel Goleman, best of Harvard Business Review 1998):

  • Self-awareness – the ability to know one’s emotions, strengths, weaknesses, drives, values and goals, and recognise their impact on others while using gut feelings to guide decisions.
  • Self-regulation – involves controlling or redirecting one’s disruptive emotions and impulses and adapting to changing circumstances.
  • Social skill – managing relationships to move people in the desired direction
  • Empathy – considering other people’s feelings especially when making decisions
  • Motivation – being driven to achieve for the sake of achievement

Goleman includes a set of emotional competencies within each construct of EI.

Emotional competencies are not innate talents, but rather learned capabilities that must be worked on and can be developed to achieve outstanding performance.

Goleman posits that individuals are born with a general emotional intelligence that determines their potential for learning emotional competencies (Boyatzis, Goleman and Rhee, 2000).

Goleman’s model of EI has been criticised in the research literature as mere “pop psychology” (Mayer, Roberts and Barsade, 2008).


Two measurement tools are based on the Goleman model:

  • The Emotional Competency Inventory (ECI), which was created in 1999
  • The Emotional and Social Competency Inventory (ESCI), a newer edition of the ECI was developed in 2007. The Emotional and Social Competency – University Edition (ESCI-U) is also available

These tools developed by Goleman and Boyatzis provide a behavioural measure of the emotional and social competencies.

The Emotional Intelligence Appraisal, which was created in 2001 and which can be taken as a self-report or 360-degree assessment (Bradberry and Greaves, 2009).

The Trait Model

Konstantinos Vasilis Petrides (“K. V. Petrides”) proposed a conceptual distinction between the ability based model and a trait based model of EI and has been developing the latter over many years in numerous publications (Petrides and Furnham, 2000; Petrides et al., 2001). Trait EI is “a constellation of emotional self-perceptions located at the lower levels of personality hierarchies and integrates the affective aspects of personality” (Petrides, Pita and Kokkinaki, 2007).

In lay terms, trait EI refers to an individual’s self-perceptions of their emotional abilities.

This definition of EI encompasses behavioural dispositions and self-perceived abilities and is measured by self report, as opposed to the ability based model which refers to actual abilities, which have proven highly resistant to scientific measurement.

Trait EI should be investigated within a personality framework (Petrides and Furnham, 2001).

An alternative label for the same construct is trait emotional self-efficacy.

The trait EI model is general and subsumes the Goleman model discussed above.

The conceptualisation of EI as a personality trait leads to a construct that lies outside the taxonomy of human cognitive ability.

This is an important distinction in as much as it bears directly on the operationalisation of the construct and the theories and hypotheses that are formulated about it (Petrides and Furnham, 2000).


There are many self-report measures of EI (Peres, Petrides, Furnham, 2005), including the EQ-i, the Swinburne University Emotional Intelligence Test (SUEIT), and the Schutte EI model.

None of these assess intelligence, abilities or skills (as their authors often claim), but rather, they are limited measures of trait emotional intelligence (Petrides, Pita ad Kokkinaki, 2007).

The most widely used and widely researched measure of self-report or self-schema (as it is currently referred to) emotional intelligence is the EQ-i 2.0.

Originally known as the BarOn EQ-i, it was the first self-report measure of emotional intelligence available, the only measure predating Goleman’s best selling book.

There are over 200 studies that have used the EQ-i or EQ-i 2.0. It has the best norms, reliability, and validity of any self-report instrument and was the first one reviewed in Buros Mental Measures Book.

The EQ-i 2.0 is available in many different languages as it is used worldwide.

The TEIQue provides an operationalisation for the model of Petrides and colleagues, that conceptualises EI in terms of personality (Petrides and Furnham, 2003).

The test encompasses 15 subscales organised under four factors: well-being, self-control, emotionality and sociability.

The psychometric properties of the TEIQue were investigated in a study on a French speaking population, where it was reported that TEIQue scores were globally normally distributed and reliable (Mikolajczak and Leroy, 2007).

The researchers also found TEIQue scores were unrelated to nonverbal reasoning (Raven’s matrices), which they interpreted as support for the personality trait view of EI (as opposed to a form of intelligence).

As expected, TEIQue scores were positively related to some of the Big Five personality traits (extraversion, agreeableness, openness, conscientiousness) as well as inversely related to others (alexithymia, neuroticism).

A number of quantitative genetic studies have been carried out within the trait EI model, which have revealed significant genetic effects and heritabilities for all trait EI scores (Vernon et al., 2008).

Two recent studies (one a meta-analysis) involving direct comparisons of multiple EI tests yielded very favourable results for the TEIQue (Martins, Ramalho and Morin, 2010; Gardner and Qualter, 2010).