The ability model of emotional intelligence
Before discussing the model of El proposed by the original authors of the test as described in Chapter 1, it is necessary to trace the logic of creating a single concept of “emotional intelligence,” as set forth in their work.
Understanding the concept of emotional intelligence requires consideration of the two components of this term: intelligence and emotions. Since the eighteenth century, in psychology, it has become a generally accepted practice to divide the psyche into three relatively independent areas: thinking, emotions, and will. Cognition enables the body to learn from the environment and to solve problems in new situations. It often serves to satisfy motives or create positive emotions. Cognition includes perception, memory, and the ability to solve problems. It is continuous and carries out flexible, intentional processing of information based on learning. The concept of intelligence is typically used by psychologists to characterize how well the cognitive sphere functions: abilities to isolate concepts, to combine them, to act upon and with them, to build judgments, and to arrive at abstract conclusions. Emotions belong to the second, so-called emotional sphere of mental functioning, which includes emotions, moods, evaluations, and other sensory states proper, but it also includes fatigue or vigor. Emotions develop in humans and in mammals. Their purpose is to respond to changes in the relationship between the individual and the environment (including the place that the person believes that she occupies in the environment). For example, anger arises in response to a threat or injustice; fear arises in response to danger. Emotions do not obey a clear line of behavior, but instead respond to external changes in relations (or changes in the subjective perception of relations). In addition, each emotion generates several basic behavioral responses to relations; for example, fear induces the desire to fight or run. Emotions, therefore, are more flexible than motivations, although not as flexible as cognition. The will is connected with the conscious purposefulness of a person, with the premeditation of her actions, that is, with motivation. It implies a person’s independence, not only in making decisions, but also in initiating actions, in their implementation and control. Often, the will is spoken of as purposeful control, implying its indissoluble connection with the mind, cognition, and opposing the emotional and spontaneous.
Much of the research looks at how motivation interacts with emotions and how emotions interact with cognition. For example, motives affect emotions when unmet needs lead to increased anger and aggression. And emotions affect cognition, when a good mood encourages a person to think positively. One would expect that the interaction of emotions and cognition gives rise to emotional intelligence. But since these three areas are combined in the more complex working of personality, it is better not to consider the emotional, motivational, or cognitive elements separately. Rather, the spotlight shifts to more general personality characteristics or social processes that take into account the entire triad. For example, self-esteem is a combination of perceptions of self at all three levels.
In the history of psychology, emotions and cognition were sometimes viewed as antagonists. According to modern ideas, emotions convey information about relationships and it is assumed that emotions and intelligence can function together. Emotions reflect the relationship between a person and another individual, their family, situation, society. For example, happiness might indicate a friend’s success; sadness - loss and disappointment.
Definitions of El should in some way connect emotions with intelligence while preserving the specific meanings of these two terms. Over the past 30 years many studies have been devoted to the interaction of emotions and cognition. Emotions, as is known, change thinking in many ways. For example, the mood of a person can change the course of her thought process. People who are in a good mood think that they are quite healthy compared to others, that economic indicators are rising, and that Moscow is not a bad city compared to Arzamas. People in a bad mood can decide that their health is much worse than that of others, that the economy is in crisis, and that Moscow is a typical modern city and is not very suitable for living a comfortable, good life. The above example demonstrates that emotions and thinking are closely interrelated. The sphere of cognitive and affective abilities also includes the study of emotional self-control, for example when a person hides her anger. The ability to feel emotions, to have access to their regulation and awareness contributes to emotional and intellectual growth, because it allows one to manage not only the emotional process, but also the process of thinking, which combines intelligence and emotions. Even in the middle of the twentieth century Wechsler asked himself: are non-intellectual, that is, emotional and volitional (motivational) abilities allowed as factors in general intelligence? He answered this question in the affirmative, but clarified that they predict intellectual behavior, but are not part of the intellect (Wechsler, 1958). Thus, most researchers of intelligence agree that personality traits do not belong to the intellect, but predetermine achievements. There is an old theoretical tradition that distinguishes mental abilities (that is, cognition) from motivations and emotions. Empirical evidence illustrates that mental abilities are not linearly related to personality traits (there are no direct correlations, only some weak and rather complex connections were found) (see, e.g., Mayer, Caruso, Zigler, & Dreyden, 1989; Sternberg & Ruzgis, 1994). In this regard, Scarr argues that “to call them intelligence does not do justice either to theories of intelligence or to the personality traits and special talents that lie beyond the consensual definition of intelligence” (Scarr, 1989, p. 78). But the designation of these qualities as “non-intellectual features” of the intellect potentially obscures their meaning (see Salovey & Mayer, 1994; Sternberg, 1997).
The broad definition of intelligence allows a distinction of its different types: verbal, spatial, visual, social, and others. Therefore, intelligence is represented by a wide range of abilities. The idea of different types of intelligence existed in this field of research from the very beginning. One of the types proposed was social intelligence, originally defined as “the ability to understand and influence people” (Thorndike & Stein, 1937, p. 275). These social/intellectual skills can also be directed toward one’s own self, so social intelligence in a broader context can mean the ability to understand and control oneself. Currently, there is a surge in interest in social intelligence and its measurement (Lyusin & Ushakov, 2004, 2009). Sternberg and colleagues (Sternberg, Conway, Ketron, & Bernstein, 1981) asked non-professionals to describe an intellectual person. Many of the identified characteristics were socially significant features. For example, they included the ability to accept others as they are, the ability to allow for mistakes, and the ability to show interest in the world. The construct of emotional intelligence is close to the concept of social intelligence. Mayer and Salovey believe that El is a section of social intelligence that includes the ability to control one’s own feelings and emotions and the also the feelings and emotions of others, to distinguish them, and to use this information to control thinking and actions (Mayer, Caruso, & Salovey, 1999). The understanding of El by Mayer and Salovey also partially overlaps with Gardner’s (1993) ideas about social intelligence, which he calls personal intelligence. This approach to intelligence, in fact, was an attempt to combine mental abilities and personality variables. Like social intelligence, personal intelligence (divided into intrapersonal and interpersonal intelligences) includes knowledge about oneself and others. One of the aspects of this intelligence, concerning the senses, is close to El. But unlike Gardner’s personal intelligence, El does not include a general assessment of one’s own feelings and the feelings of others, but, rather, focuses on knowing and using one’s own emotional states and those of others in order to solve problems and control behavior.
El is understood as the ability to recognize the meanings of emotional patterns, to reason, and to solve problems based on them (Mayer & Salovey, 1997; Salovey & Mayer, 1990). El describes a multitude of discrete emotional abilities. These emotional abilities can be divided into four classes or branches as described in Chapter 1. The most basic skills and abilities use perception and evaluation of emotions. For example, at first a baby will learn about emotions transmitted through facial expressions. From birth, she imitates emotional expressions of happiness, sadness, fear, surprise. An infant compares her sensations with the observations of pain or happiness reflected on the faces of parents and other adults. As s/he develops, the child learns to more accurately distinguish feelings and evaluate their sincerity. It is also important that people summarize emotional experience into more general categories, which makes it possible to recognize their specific expression, referring to one category or another. This ability allows one to understand the basic emotional manifestations of different people, of different races and ages, making this emotional “language” a means of non-verbal social communication.
The second set of abilities involves a mental process of comparing emotions with each other and with other existing concepts and ideas. For example, the image of an emotional experience is preserved, allowing one to compare it with a similar feeling caused by sound, color, or taste. The third level includes an understanding of emotions, their causes, and consequences. Anger comes from injustice, fear often gives way to relief, despondency can distance us from others. For example, a woman looks extremely angry, but an hour later, she is already ashamed of her anger. It is likely that changes in emotional states were influenced by certain events such as if she expressed her anger inappropriately or found that she falsely believed that a friend had betrayed her. El, therefore, includes the ability to recognize emotions, predict their dynamics, and adequately judge their causes. The fourth and highest level of El includes the management and regulation of our own emotions and those of other people. For example, knowing how to calm down after anger or the ability to alleviate another person’s anxiety.
The model in question predicts that emotionally intelligent people probably grew up in favorable home conditions, where attention was paid to emotions in parenting, where they experienced emotions, choosing socially acceptable patterns for communication and discussion of feelings, and developed experience in the emotional field through ethics and aesthetics, moral development, or spiritual feelings (Mayer & Salovey, 1995).
El as a mental ability in the model of Mayer and Salovey is theoretically distant from the traditional representations of intelligence and its verbal and behavioral manifestations. For example, compared to social intelligence, El includes internal (intrapersonal) emotions that are important for personal growth. As well as that, El, more so than social intelligence, focuses on the emotional aspects of problems as opposed to the social or political aspects. This distinguishes it from many aspects of social competence. In many tests of verbal intelligence, questions are addressed specifically to social competence (e.g., “What is celebrated on May 97”), although according to the majority of researchers, social intelligence is very different from general intelligence (e.g., Sternberg & Smith, 1985). Theoretical foundations and an emphasis on El mean that this ability model can be a good alternative to traditional measurements.
Mayer and Salovey believe that El is closely connected with cognitive intelligence, since they postulate the unity of affect and intelligence, which corresponds to the Russian theoretical traditions of the school of Vygotsky and Rubinstein. A similar solution is proposed by Ushakov, who believes that “social intelligence is on par with other types of intelligence, forming with them the ability for a higher kind of cognitive activity - generalized and mediated” (Ushakov, 2004, p. 18).
The model of mental abilities can be considered to be more of an El, since it relies to the greatest extent on the integration of emotional and intellectual abilities. El emphasizes the boundaries between cognitive and non-cognitive abilities and points to the role of both cognitive and non-cognitive abilities in the process of socialization. The ability model is fundamentally different from the mixed type models. But both can be useful in studying human efficiency and success in life.
By 1997, Mayer and Salovey refined and expanded their model of El (Mayer & Salovey, 1997). In the redesigned model, a new emphasis was placed on the cognitive component of El associated with the processing of information about emotions. Also in this model there appeared a component associated with personal and emotional growth. In the light of these changes, the concept of El has received a new definition, as the ability to process information contained in emotions: to determine the meaning of emotions and their connections with each other, to use emotional information as a basis for thinking and decision-making. Further analysis of abilities related to the processing of emotional information allowed Mayer and Salovey to identify four components of emotional intelligence, which were called “branches” and are described in Chapter 1.
Based on this hierarchical model, the authors created the first experimental version of the methodology for the study of emotional intelligence - MEIS (Multi-factoral Emotional Intelligence Scales). It consisted of 12 subtests (from two to four subtests for each “branch”) and included 402 questions. For each question there were offered several answers. Scoring was conducted on the basis of a consensus of expert assessments or a given standard. MEIS was effective in measuring levels of El in the model of the four branches of El, but had certain limitations. Some 402 questions made the test too time- consuming to be used for practical or research purposes. Also, after the test was created, it became apparent that there were several possibilities for improvement, in particular, in terms of clarification of the content of the available scales and the addition of several new ones. By 1999, Caruso joined Mayer and Salovey (Mayer et al., 1999) and together they developed the MSCEIT Research Version 1.1 (RV 1.1) test, and by 2002 they proposed a modified version - MSCEIT V2.0 (the Mayer-Salovey-Caruso Emotional Intelligence Test; Mayer, Salovey, & Caruso, 2002). This test consists of 141 questions combined into eight sections - two for each “branch” - representing the components of emotional intelligence (see Table 2.1). This method turned
26 Elena A. Sergienko et al
Table 2.1 MSCEIT Version 2.0 results, structure, and levels
Source: Authors’ own compilation.
out to be more balanced and consistent than the previous ones, demonstrated good psychometric results, and is now the most prevalent.
Since 2006, the laboratory of developmental psychology at the Institute of Psychology of the Russian Academy of Sciences has been working on the adaptation of the MSCEIT V2.0 methodology for the Russian-speaking respondents. The final stage of this adaptation is standardization. Having successfully completed the Russian translation and confirmed the quite good psychometric indicators of the adapted version of MSCEIT V2.0 (Sergienko & Vetrova, 2010), we proceeded to standardize the methodology with the detection of the cultural specificity of the test. The purpose of this final stage of work was to gather a regulatory set of data for the comparison with similar data from the authors of the test. We also assumed that the cultural specificity of El would manifest itself primarily in the field of sex and age differences. At present, the results of 3827 respondents (1578 men and 2249 women) aged from 16 to 72 years are available for analysis.
The regulatory set of data for the English-speaking version of MSCEIT is based on data collected primarily from sites in the United States, but a number of other countries also participated in data collection, including the UK, Canada, Malta, South Africa, Australia, Switzerland, Scotland, Philippines, India, Slovenia, and Sri Lanka. The regulatory set of data for MSCEIT is a compilation of data from three samples. In total, these samples create a regulatory set of data of 5000 respondents (see Table 2.2; Mayer et al„ 2002).
The Russian-speaking regulatory sample is at this point smaller than the English-speaking sample (3827 people) and is represented mainly by people from Russia, Belarus, and Ukraine. Nevertheless, the geographical coverage in Russia is quite wide - Moscow, St. Petersburg, Perm, Tomsk, and other major cities. However, the majority of respondents filled out a paper version
Table 2.2 The distribution of the English-speaking regulatory sample by age and sex (N=5000)
Note: For 15.80% of the sample there is no information on sex or age. Source: Authors' own compilation.
Table 2.3 The distribution of the Russian-speaking regulatory sample by age and sex (N=3827)
Source: Authors' own compilation.
of the test. Only 253 people (64 men and 189 women) completed the test electronically. In addition, the collection of data for the regulatory sample still continues. The already available amount of data allows comparisons with similar indicators provided by the authors of the test. As in the English- speaking group, in our sample there is a certain “bias” toward the female respondents (Table 2.3). Also, the ages of the largest number of respondents range from 20 to 29 years in both groups (35.54 percent in the English- speaking sample and 44.94 percent in the Russian-speaking sample). However, in the English-speaking sample the representatives of the older generation (aged over 30 years) account for only 14.86 percent of the entire standardization sample (7.64 percent of men and 7.22 percent of women). While in the Russian-speaking sample, the indicated age range accounts for 38.91 percent of the entire sample (15.63 percent of men and 23.28 percent of women). Perhaps this age-related “imbalance” was formed due to the fact that, in the English-speaking sample, for 15.80 percent of respondents there is no information on sex or age. In our sample, information on sex and age is presented for all members of the standardization group.
28 Elena A. Sergienko et al
In the area of the achieved level of education, there are no noticeable differences between the English-speaking and Russian-speaking regulatory samples - the largest part of the respondents (72.90 percent in the English- speaking sample and 78.26 percent in the Russian-speaking sample) are students or graduates of higher educational institutions. In addition, the specifics of studying El as a predictor of a certain professional success, implies the study of respondents who currently receive or already have a higher education.
Tables 2.4 and 2.5 present data on average values, standard deviation, standard errors, and average score asymmetry for all indicators of the MSCEIT methodology in the English-speaking and Russian-speaking versions. Comparison of average values by the Student’s T-test showed that in the Russian-speaking version of the test all the indicators are statistically significantly lower than those of the English-speaking version (p<0.0001). However, the ratio between the indicators for the individual scales of the Russian-speaking sampling methodology corresponds to the English- speaking sampling. For example, the highest average score is for the branch
Table 2.4 MSCEIT results for the English-speaking sample (N=5000)
Source: Authors’ own compilation.
Table 2.5 MSCEIT results for the Russian-speaking sample (N=3827)
Source: Authors' own compilation.
of emotional intelligence “Understanding emotions,” and the lowest is for the branch “Managing emotions.” However, in sum, they form the Strategic Domain, which in terms of the average is equal to the Experiential Domain and the general level of El.
It should be noted that, despite the size of normative samples (5000 people in the English-speaking sample and 3500 people in the Russian-speaking sample), the distribution of data remains abnormal across all scales of the MSCEIT V2.0 methodology. As a result, when working with the obtained data, we were forced to use non-parametric methods of statistical processing. In addition, like our American colleagues, we had to resort to the nonlinear standardization of raw points through the calculation of percentiles to get IQ points.