IIER logo 3
Issues In Educational Research, Vol 15, 2005
[ Contents Vol 15 ] [ IIER Home ]

Emotional intelligence and teaching: Further validation evidence

Chris Perry and Ian Ball
Deakin University
Further evidence is presented to demonstrate the validity of a new measure of emotional intelligence: Reactions to Teaching Situations (RTS). Using criterion-related groups of high and low scorers on the RTS, it is shown that high scorers give more responses coded as emotional intelligence in their answers to sentence completion tasks relating to ten situations found in teaching. The questions of convergent and discriminant validity is tackled by examination of correlations of emotional intelligence scores and scores on the Multiple Intelligences Checklist for Adults (MICA) and information processing preferences as measured by the Myers-Briggs Type Indicator (MBTI). The results confirm that emotional intelligence (as assessed by the RTS) bears significant relationships to both intrapersonal and interpersonal intelligences and also to linguistic intelligence, but emotional intelligence shows no significant relationships with the MBTI preferences.


Within the area of research that has its focus on individual differences, the notion of emotional intelligence and its study as a concept has raised considerable interest over the past decade (Day, 2004). As Austin, Saklofske and Egan (2005) note,
the idea that people differ in measurable ways in their emotional skills is an interesting idea in its own right, suggesting the opening up of an area of individual differences assessment not currently covered by existing measures of intelligence and personality (p.547).
Salovey and Mayer (1990) were early adopters of the term emotion intelligence which they initially categorized in five domains: self-awareness, managing emotions, motivating oneself, empathy and handling relationships. These researchers describe emotional intelligence as a form of social intelligence that
involves the ability to monitor one's own and other's emotions, to discriminate among them, and to use the information to guide one's thinking and action (Mayer & Salovey, 1993, p.433).
In later work by Mayer and Salovey (1997), a revised model of emotional intelligence gives more emphasis to the cognitive components of the concept, focussing on the potential that understanding of emotional intelligence may give us in terms of intellectual and emotional growth. This revised model details four branches of emotional intelligence: identifying emotions, using emotions, understanding emotions and managing emotions (Caruso, Mayer, & Salovey, 2002, p.307).

A more popularised approach to emotional intelligence is taken by Goleman (1997), who refers to emotional intelligence as an individual's ability to assess a particular social situation and to use information to guide his or her emotional response. Goleman views emotional intelligence as a meta-ability, "a master aptitude, a capacity that profoundly affects all other abilities, either facilitating or interfering with them" (p.80).

Schutte, Malouff, Hall, Haggerty, Cooper, Golden, and Dornheim (1998, p.167) suggest that Goleman's work, by drawing attention to the concept of emotional intelligence, has fostered the idea of "viewing the experience and expression of emotions as a domain of intelligence."

Views of emotional intelligence

While the more recent examination of the concept of emotional intelligence may have clarified and given depth to our understanding, emotional intelligence has been documented in the psychological literature on social intelligence as far back as work of Thorndike in the 1920s, for example. The more recent work of Sternberg and Gardner has encouraged an examination of a more multiple view of intelligence beyond the standard concept of intelligence as a single immutable factor (g) and as measured by a single test type (IQ). Sternberg (1985) views intelligence as a relatively singular construct composed of multiple subtheories or components, while Gardner (1983) claims
(t)o my mind, a human intellectual competence must entail a set of skills or problem solving - enabling the individual to resolve genuine problems or difficulties that he or she encounters and, when appropriate, to create an effective product - and must also entail the potential for finding or creating problems - thereby laying the groundwork for the acquisition of new knowledge (p.60).
Gardner's theory of multiple intelligences (originally seven, later an eighth and possible ninth) does not include emotional intelligence per se, however his descriptors of interpersonal intelligence (the capacity to recognise the intention of others) and intrapersonal intelligence (the ability to understand one's own feelings and motivations) appear to share some commonality with emotional intelligence. This commonality could provide a perspective on the nature of emotional intelligence and as Schutte et al. (1998) suggest, may have influenced the development of models of the concept.

Goleman (1997) asserts that while Gardner's personal intelligences are a form of emotional intelligence, they emphasise only cognitive elements of emotions rather than the full range of emotional abilities. In Goleman's view they emphasise the metacognitive nature of emotional understanding rather than focussing on the emotions themselves. However the metacognitive aspects of emotional understanding are seen to be central to the information processing view of emotional intelligence.

Interest has also come from perspectives on personality. Jung, according to Myers, McCaulley, Quenk and Hammer (1998), initially attempted to explain individual differences in personality from his observations about two types of people, introverts and extraverts. Further observation and development of his theory of psychological types introduced the pairs of mental opposites, the two perceiving functions, Sensing (S) and Intuition (N) and the two judging functions, Thinking (T) and Feeling (F). Jung's theory of eight types was subsequently expanded by Myers and Briggs by the addition of the Judging-Perceiving dichotomy building on Jung's description of an auxiliary function that supported and complemented the dominant function in every type. This development resulted in a theory about sixteen different psychological types.

The Myers-Briggs Type Indicator (MBTI) is a well established and widely used measure based on Jung's type theory and gives information about a person's preferences on four dichotomies - extraversion versus introversion (E-I); sensing versus intuitive (S-N); thinking versus feeling (T-F); and judging versus perceiving (J-P).

The aim of the MBTI is to identify, from self-report of easily recognised reactions, the basic preferences of people in regard to perception and judgement, so that the effects of each preference, singly and in combination, can be established by research and put to practical use. (Myers & McCaulley, 1985, p.1)
This aim is because type theory was seen not only as means for human understanding but also as a catalyst for the realisation of human potential.

Type theory asserts that individuals have well established habits of perception and judgement. These functions, according to Pearman and Fleenor (1997) "are used in various degrees and intensities depending on the situation in which the person finds themselves, and the habit of use which has become more comfortable for the person" (p.185). Essentially this is an information processing view and is consonant with the model of emotional intelligence being validated here. Pearman and Fleenor also note that "(t)he type model assumes growth, change and adaptation, and that while patterns of behaviour make up the types, all types can be effective in work and relationships" (p.187). This view of type and behaviour contrast strongly with the more dominant dispositional factors (traits) model.

Lawrence (1998) asserts that neither Goleman nor Mayer and Salovey take psychological type into account in their views of emotional intelligence. Lawrence sets out to enrich the models and draws attention to the two ways of making rational judgments in the Myers-Briggs Type Indicator (see Myers et al., 1998). These ways are thinking and feeling. Lawrence indicates that Jung had to distinguish between emotion and feeling, between emotional response and feeling judgment. For Jung, coming to a conclusion by means of a feeling judgement, a rational process, is very different from coming to a conclusion by means of emotional reaction. To Lawrence, experiencing an emotion involves no reasoning, but a person making a rational judgement (either through thinking or feeling criteria) can and should use emotional experience to inform and help the rational process.

In exploring the linkages between the theory related to multiple intelligences and that related to MBTI, Nardi (2001) focuses on the four temperaments of psychological type. His analysis of the distinctions between interpersonal and intrapersonal intelligences is as follows:

Interpersonal intelligence includes recognizing moods, motives and intentions: experiencing empathy; making distinctions between people and predicting behaviour; and knowing proper cultural kinship and relational roles (Nardi, 2001, p.72).
Nardi contends that there are many facets to interpersonal intelligence including, ability to establish and maintain rapport, accurately reading others' intentions, motivating or organising others, being compassionate, and actively listening and observing.
Intrapersonal intelligence includes an understanding of one's self, awareness of one's strengths and weaknesses, and realism about personal goals (Nardi, 2001, p.75).
Nardi asserts that it also includes reflecting on and regulating one's thoughts and feelings alone and in interpersonal relationships. The self-reflective aspect is seen to be unique to this intelligence.

Nardi also notes that Goleman's concept of emotional intelligence and Gardner's multiple intelligences are different but compatible theories (Nardi, 2001, p.122). He posits that emotional intelligence includes interpersonal and intrapersonal intelligence.

Pearman (2002) links the intrapersonal arena (the inner world) from Gardner's model to Introversion in the Myers-Briggs model. The key elements that define these links are identified as: self-awareness, self-regulation, emotional self-control, flexibility, motivation, achievement, resilience, and, well-being and stress management. This is balanced by the interpersonal arena (the extraverted world) which is linked to Extraversion in the Myers-Briggs model. The essential components of emotional intelligence in this linkage are: demonstrative empathy, energy, social skill, tolerance, persuasiveness, and ability to lead. Pearman notes that persons may exhibit low, average, or high awareness and expressiveness of these qualities, and that : "it is likely that some of them are not fully developed for you, while others are finely tuned" (Pearman, 2002, p.12). Pearman asserts that "there is little doubt that you already use intrapersonal and interpersonal processes that express degrees of emotional intelligence" (Pearman, 2002, p.2). His book details how people with different psychological types might set out about developing their emotional intelligence ability.

Lawrence (1998) notes that the concept of emotional intelligence has stirred a lot of interest among education theorists and practitioners. Thus it is relevant to ask what relation emotional intelligence has to the existing constructs of multiple intelligences proposed by Gardner (1983), and the personality types indicated by Myers and Briggs (see Myers et al., 1998) and whether a new measure bears any systematic relations to these widely used constructs.

Recent research related to emotional intelligence has continued to broaden our understanding of the concept, particularly as it may relate to the teaching profession, showing for example, that this intelligence is gender related with women scoring higher on measures of emotional intelligence than men (Brackett, Mayer & Warner, 2004; Kafetsios, 2004; Perry, Ball & Stacey, 2004; Wu, 2004; Van Rooy, Alonso, & Viswesvaran, 2005). Findings also suggest that emotional intelligence increases with age, that is older people display higher levels of emotional intelligence than do younger people (Kafetsios, 2004; Wu, 2004; Van Rooy, et al., 2005).

Initial focus on the concept of emotional intelligence from writers such as Mayer and Salovey and Goleman has led to the development of several comprehensive theoretical frameworks (Petrides & Furnham, 2000). As noted by Schutte et al. (1998, p.167) "(t)hese models do not contradict each other but they do take somewhat different perspectives on the nature of emotional intelligence." One theoretical perspective sees emotional intelligence as a noncognitive intelligence, that is, a part of personality or a trait (Bar-On, 1977; Goleman, 1997; Mendes 2003). Others view emotional intelligence from a perspective of information processing, that is as an ability (Mayer & Salovey, 1997; Mayer, Salovey & Caruso, 1999). Petrides & Furnham (2000, p.319) conclude that "trait EI appertains to the greater personality realm whereas information-processing EI is an attempt to chart new territory in the field of human ability." One of the implications of the ability conception of emotional intelligence is that it may be subject to modification through training and experience. The authors see that in the teaching profession there is scope for the use of strategies to enhance emotional intelligence.

Validity of measures

While differences in theoretical perspectives may not be contradictory, they remain issues to be examined when attempting to validate any measures of emotional intelligence.

Barchard and Russell (2004) and MacCaan, Matthews, Zeidner and Roberts (2004) are among a vocal group who assert that researchers need to examine the psychometric properties of emotional intelligence measures. Also Day (2004, p.251) argues that "it is necessary to gather multiple indices of validity as evidence of the validity of a measure."

In a prior study (noted below) where we give account of the development of the RTS (Perry, Ball & Stacey, 2004), we demonstrate its factorial structure, which is one index of construct validity. This exploratory factor analysis produced a four factor solution that provided partial support for the separation of the four branches in that three of the factors seemed to be aligned with each of the three of the four factors. The other, a more general factor, comprised a mixture of branches. In this study we turn to two other validity indices by comparing the responses by criterion related groups, and addressing the question of how the RTS relates to theoretically similar and different constructs.

Background to this study

In a recent study (Perry, Ball & Stacey, 2004) we reported on work that resulted in the development of a usable measure of emotional intelligence which related directly to the work of teachers in schools. For this measure, the Reactions to Teaching Situations (RTS) we drew on the four-branch model of emotional intelligence as a framework. A series of ten teaching situations were presented.

The situations written for the measure were typical of those that could be expected as part of the practice of teaching and three independent judges agreed that they were reasonably common situations likely to be experienced by teachers involving learners, authority figures, other staff and parents. The situations varied in emotional character: from those that could be threatening to a person's feelings of self worth, to those validating the professionality of the individual. The situations were presented in a random order and were described as follows.

  1. One of your students, whose learning is generally slow and erratic, has just made a breakthrough and has acquired a concept you have been teaching for some time.

  2. A parent has lodged a formal complaint about your teaching methods which you feel is totally unjustified and blown out of all proportion. Moreover you are unsure about how 'just' the Principal will be in handling this issue.

  3. Your students are actively involved in their group work, but you sense that a few are taking advantage of you, and becoming noisy and unproductive.

  4. Your level co-ordinator calls you in and says: "Your student assessments have been too generous, and you need to do them all again".

  5. A student, who has the reputation for being difficult to handle, loses it totally on an excursion where you are in charge, and puts on a temper tantrum.

  6. A student, who has recently made a special effort with a piece of work, says: "You are the best teacher I've ever had".

  7. Your initial ideas have been highly valued and adopted in practice by your teaching team.

  8. You find that you were not included in a staff group invitation to go for drinks after school.

  9. While on yard duty you hear one student making a negative comment about a student from a racial group to which you also belong.

  10. In your most recent performance review, your team leader gives you very positive feedback and states your performance has exceeded expectation.
The participants for the study were a cohort of 357 students undertaking the second year of a four year teacher education course. The cohort contained 76% females and 24% males. The age distribution was 81% 25 years or under; 11% 26-30 years; 5% 31-35 years; 3% 36 years or over (see Perry, Ball & Stacey, 2004). These participants were asked to consider how they would feel and think in each of the ten situations. Each of the six possible reactions was rated as to the likelihood of that particular reaction being made, ie, rating how they would typically deal with these particular emotional aspects of teaching. The ratings were made on a five point Likert scale labelled: Never likely, Seldom likely, Sometimes likely, Usually likely, Always likely.

The analyses of the likely responses to the ten situations, based on the four branch model of emotional intelligence formed the new measure and was considered successful. The alpha reliability estimate of 0.82 was acceptable and compared favourably with other measures, such as Bar-On's EQ-I.

Objectives of the study

In this study we focus on the open-ended responses gained as part of the RTS questionnaire responses and use these to further validate that the concept measured is emotional intelligence. What people say in this section of the RTS should reflect differences along the dimension of the concept, that is, these responses should discriminate between people in regard to their emotional intelligence ability.

In addition in this study, the results of the RTS are compared with other measures that have some connections to emotional intelligence to show whether the concept of emotional intelligence is the same as or different from the concept of multiple intelligences and personality type.


The following is a description of the process and the results of comparing open-ended responses written by high and low scorers on the Reactions to Teaching Situations (RTS) measure of emotional intelligence (Perry, Ball & Stacey, 2004). As part of the initial data collection respondents had been asked to provide completions to two stems. The stems were the same for each of the ten situations and were: "It is most likely that I would feel .......... and then I would .............." The open-ended responses were included in the initial measure in order to gather samples of actual language used by individuals, but also provided the opportunity to examine their content for evidence of emotional intelligence.

Three judges with education and teaching backgrounds and who were knowledgeable about the Mayer et al. (1999) model of emotional intelligence considered whether the open-ended responses indicated one of the four branches of emotional intelligence. These responses were judged without any knowledge of the scoring status of the respondents. The set was randomly presented and contained the ten highest total scores on the RTS and ten lowest ten scores, randomly presented. The group of highest scorers had a mean score of 161.9 with a range from 159 to 168, whereas the group of low scorers had a mean score of 108.9 and a range from 101 to 114. These classifications represented the extreme scores on the RTS measure of emotional intelligence. It was expected that their open-ended responses would show a qualitative difference in the responses between both groups, if the RTS was a valid measure of emotional intelligence.

Tables 1 and 2 below present the results of these analyses of the content of the open-ended responses from both groups. The results are the sum across ten respondents, ten situations and three judges.

Table 1: Frequency of judged responses made to the stem It is most likely that I would feel ... by the two groups

Judgement about open-ended responsesStatus of RTS group
Low scorersHigh scorers
Indicating emotional intelligence167254
Non-emotional intelligence13346
The result of Chi square for this table was 61.7 (df=1) p<.001 indicating that the results were unlikely to be due to chance factors alone.

It is clear from the results in Table 1 that both groups seem to be able to provide statements to the first stem: "It is most likely that I would feel" which indicated identification of emotions, but the group of high scorers on the RTS provide many more responses of this sort compared with the group of low scorers.

Table 2: Frequency of judged responses made to the stem and then I would ... by the two groups of RTS scorers

Judgement about open-ended responsesStatus of RTS group
Low scorersHigh scorers
Indicating emotional intelligence65121
Non-emotional intelligence235179
The result of Chi square for this table was 24.4 (df=1) p<.001 indicating that the results were unlikely to be due to chance factors alone.

Table 2 indicates a differential in respect to the responses to the second stem and then I would. It can be seen that the high scorers provided about twice as many responses that indicated emotional intelligence than did the low scorers. The greatest frequency can be noted for the low scorers not providing a response that could be judged to convey emotional intelligence.

Examples of the scoring process for the responses are given below. To qualify as an emotional intelligence response, the wording had to indicate the use of one of the four branches of the Mayer et al. model in the view of the judges. Examples are also provided where responses were not judged as displaying emotional intelligence.

Two examples from each group will be presented to clarify this process.

Consider the situation: A student, who has recently made a special effort with a piece of work, says: "You are the best teacher I've ever had".

One respondent wrote: "It is most likely that I would feel very pleased with both myself and my learners - we had worked well together and then I would reflect and see what I did to motivate/engage the student". Here the feeling of being pleased goes both ways - to the teacher and the student, and the act of reflection investigates what strategies have been successfully employed.

In the second case, another respondent wrote: "It is most likely that I would feel satisfaction and then I would praise the student on their work but not forgetting the methods used to elicit the response". Here, the feeling is clearly represented. The consequent action is a reinforcement for the student and a consideration of a strategy used by the teacher in order that this situation of positive feelings might be replicated.

Contrast these responses from high scorers on the RTS with the following responses to the same situation, from two low scorers on the RTS.

One respondent wrote: "It is most likely that I would feel that the student is feeling confident in her working ability which the student should be proud of and then I would (no response)".

Another respondent wrote; "It is most likely that I would feel pleasantly surprised and then I would (no response)".

In both these responses from low scorers there is the identification of a feeling state, but no response about maintaining or recreating or strengthening that emotional response. No further action seems to be needed by these respondents. There is no further investigation of consequences or actions that could be taken. While there may be reasons for non responses, for example lack of practical experience, or lack of willingness to speculate, it would appear justified to regard non response as non application of emotional intelligence. Those who have given a response and where those responses indicate emotional intelligence should be positively identified.

Next consider the qualitatively different responses to the following situation: A student, who has the reputation of being difficult to handle, loses it completely on an excursion where you are in charge, and puts on a temper tantrum.

Two responses from low RTS scorers are presented first.

A respondent wrote: "It is most likely that I would feel uneasy about the situation, and then I would caution the student for such behaviour". The judges scored the 'uneasy' part of the statement as identifying an emotion, but did not see that the action taken was in any way linked with the emotional response.

Another respondent wrote: "It is most likely that I would feel the need to act and then I would act to rectify the situation". In scoring this statement, the judges felt that the respondent did not identify an emotion, nor emotional intelligence as defined, but simply a cognitive response to the behaviour.

Both of the above statements were made by low scorers on the RTS, and fail to provide an adequate explanation or description of either the action to be taken or how the emotionally charged atmosphere is likely to be dealt with. There does not appear to be any evidence that the emotional state of the student or the teacher bears on the situation, rather the reactions are cognitive in nature. The statements lack evidence of emotional intelligence at work.

Now, contrast the above statements with those written by high scorers on the RTS.

One respondent wrote: "It is most likely that I would feel angry and think he was doing it to get at me and then I would try to deal with my emotions and gain control". Here the judges felt the person was displaying evidence of emotional intelligence, recognising an emotion and using or managing that emotion in his or her own subsequent behaviour.

Another respondent wrote: "It is most likely that I would feel annoyed at the student, and then I would take them aside and discuss what upset them and try to handle it. I'd keep this student with me for a while to calm down and justify their behaviour". The judges felt that such a response firstly identified an emotion, eg, feeling annoyed, and then indicated how this emotion could be managed to alleviate the problem. This was taken to indicate an awareness of managing emotionally charged situations, and indicated some understanding of an emotion and how to manage it successfully in solving the problem. In this case the intervention was focussed on managing the student's emotional state.

In considering these two statements a fairly clear linkage can be seen between the existence of emotional intelligence and what might be regarded as 'good' teaching practice and student management skills: the appropriateness of a response to further the development of the student. This linkage question is discussed at greater length later.

Relationships between the RTS and other measures

Scores on the Reactions to Teaching Situations (RTS) measure of emotional intelligence were correlated with two other measures to determine the extent to which they were discriminant or convergent. In both cases, the other measures were administered at the same time as the RTS.

Multiple intelligences

The Multiple Intelligences Checklist for Adults (MICA) (McGrath & Noble, 1995) was developed to provide a quick guide to an adult's status of development of seven of the multiple intelligences identified by Gardner. In the following discussion, the names of the factors are those used by Gardner although the authors of the instrument have somewhat different terminology. The Checklist was developed using a rational difference approach where items were validated by persons with demonstrated expertise in the particular intelligences.

As noted earlier, a number of authors such as Goleman (1997) have pointed out that two of the intelligences identified by Gardner appear to be involved in emotional intelligence. These are the interpersonal and intrapersonal intelligences.

Using the guidance from such literature, it was hypothesised that Gardner's intrapersonal and interpersonal intelligences would be positively correlated with RTS scores. Table 3 confirms that hypothesis where both intrapersonal and interpersonal scores on the MICA are positively correlated with RTS scores to an extent not attributable to chance factors (a probability less than one in a thousand).

Table 3: The correlations between RTS scores and the intelligences measured
by the Multiple Intelligences Checklist for Adults

IntelligencesRTS scores

The other statistically significant correlation that can be observed in Table 3 relates to Linguistic Intelligence. It can be seen from the earlier section on open-ended responses, that the high RTS scorers seem to write longer responses, and therefore a correlation with Linguistic Intelligence with RTS, although unexpected, might be readily explained.

Myers-Briggs Type Indicator (MBTI)

It was hypothesised that the Sensing-Intuition and Thinking-Feeling preferences of the MBTI would be associated with differential responses on the RTS because those preferences represent distinctive kinds of perception and judgement, ie, different ways of gathering information and using decision-making strategies. The Sensing-Intuition preference represents a difference between a focus on in-the-moment attention, sensory awareness, and a realistic perspective (Sensing) versus a focus on the abstract, big-picture issues and future implications (Intuition). The Thinking-Feeling preference represents a difference between looking at logical structures, making use of critique, showing scepticism (Thinking), and on the other hand, an empathetic response, decision-making from a perspective of seeking harmony with others and one's own ideals (Feeling). It was not hypothesised that relationships would be found with either Introversion-Extraversion or with Judging-Perception, and emotional intelligence.

It was considered that the perception of I would feel would be sensory information most clearly associated with the Sensing preference: that those with a Sensing preference would correlate with RTS scores. The decision about and then I would would be linked with the decision making process of what action would be most appropriate, and the nature of emotional intelligence would suggest that Feeling preferences which take the needs of others into account in their decision-making would show correlations with RTS scores.

Table 4: Correlations of RTS scores with Myers-Briggs Type Indicator Preferences

MBTI preferencesRTS scores

Table 4 above records the correlations between RTS scores and scores on the four sets of MBTI preferences. It is clear that all correlations are close to zero and do not differ significantly from chance expectations. Therefore this evidence did not support the hypothesised relationships.


Results for the sentence completion task suggest that those with a more developed emotional intelligence (high scorers on the RTS) are likely to behave differently than those with less developed emotional intelligence (low scorers on the RTS). The data reported in Tables 1 and 2 which show higher frequencies for responses that indicate emotional intelligence, imply that these beginning teachers are more sensitive than their lower scoring colleagues to their own emotions and the emotions of others. In perceiving situations they give greater consideration to the needs of students than to their own needs. This sensitivity to the emotional needs of others can be viewed from two aspects. Firstly, it recognises the uniqueness of the interaction between teacher and student and the skill of the more emotionally intelligent teacher in using this interaction to create more authentic learning opportunities which extend beyond the mere transition of knowledge. Secondly, this indicates the greater insight and self knowledge on the part of the teacher about how, in emotionally charged situations, they may manage their own emotional responses more effectively.

In devising the instrument (the RTS), the researchers had a concern that the alternative responses to each of the scenarios expressing the fourth branch of emotional intelligence, that of managing emotions, were often reflections of 'good teaching' as if managing emotions and effectively managing classrooms were interchangeable. The two aspects discussed above allow us to reconcile this concern to some extent, that is, good teaching practice does reflect the exercise of emotional intelligence.

The evidence in Table 3 clearly shows some modest overlap between the RTS and three of the intelligences as measured by the MICA. The correlations with interpersonal and intrapersonal intelligences are less than 10% of the variance accounted for and although this shows an overlap, it does not indicate that the constructs of emotional intelligence and multiple intelligences are identical. These results confirm prior predictions (Goleman, 1997). The emergence of a significant correlation with linguistic intelligence found in this study may be associated with the highly verbal nature of the task. The low correlations with the remaining intelligences can be taken to demonstrate some discriminant validity, that is emotional intelligence does not share common variance with these other constructs.

The expected correlations with distinctive ways of perceiving and judging did not occur but as has been discussed earlier, the literature about emotional intelligence and psychological type has been equivocal.

It is interesting to note that Pearman and Fleenor (1997) looking at a multivariate analysis of relationships between emotional intelligence (using selected scores on the California Psychological Inventory) and the MBTI types, also report that all types can show emotional intelligence ie, that there were zero correlations with the four preferences. Their results are seen to provide evidence about the significant differences between the sixteen types, and also to confirm that each of the sixteen types have different paths to effectiveness (pp.193-194). These results also support a developmental view which identifies those behaviours not currently expressed and ability to learn to use them effectively.

Sjoberg's psychometric analysis of a series of measures purporting to be emotional intelligence (2001) also reports low or near zero correlations between emotional intelligence and MBTI preferences, and concludes that "whatever is measured by the Myers-Briggs scales, it seems to be unrelated to emotional intelligence" (p.94).

That finding is in conflict with those reported by Higgs (2001). The research was specifically designed to examine hypothesised relationships. Higgs found that the preference for Intuition was significantly and positively related to higher levels of EI as measured by the model proposed and validated by Dulewicz and Higgs (1999).

Given the range of individual differences found in the RTS measure, it seems pertinent to regard emotional intelligence as a measure of an ability, that is for teachers to develop an "awareness of the importance of listening to the information or feedback the emotion is giving one at the workplace" (Matthews, Zeidner, & Roberts, 2002, p.542). It does appear that the RTS scores are associated with a teacher's ability or capacity to learn 'on the job', as Matthews et al. (2002, p.177) suggest, an intelligence is a capacity to learn. However as Day (2004, p.261) asserts, given the current state of knowledge "the degree to which individuals can increase their level of emotional intelligence is unknown."


Mayer and Cobb (2000) assert that we are only beginning to learn about emotional intelligence, and that we must continue to gather data to support the assumption of the existence of this concept. After a comprehensive literature review, Day (2004) also highlights the need for more work to be done in order to provide a clearer conception of emotional intelligence. Day (2004) further argues that this work must be based on the use of sound methodology and statistical techniques.

However, in spite of these cautions, the work that has been completed has identified emotional intelligence as a legitimate research area (Day, 2004).

For educational practitioners, the research on emotional intelligence has increased our awareness of the role this concept may play in the professional life of teachers in schools and classrooms. Matthews et al. (2002, p.534) suggest that, "the benefits of emotional intelligence appear to reside mainly in raising awareness of emotional intelligence issues and motivating educators to take emotional issues seriously." Now that a reliable and valid measure of emotional intelligence has been developed in the context of teaching, it will be possible to investigate the modifiability of existing levels of emotional intelligence and to further investigate any relationship between, for example, years of experience, teacher status, and subject specialisms. It is now possible to examine which strategies might be more effective than others in working with pre-service and practicing teachers to extend their sensitivities to the need of learners.

In assessing the validity of the RTS as a measure of emotional intelligence, this study has demonstrated evidence of factorial structure and relationships to respondents' perceptions of their feelings and likely actions in emotionally charged situations. Also relationships have been found with other concepts of emotional intelligence (intrapersonal and interpersonal intelligence) and discriminant validity has been demonstrated with aspects of personality (MBTI and other multiple intelligences).

The validation of the Reactions to Teaching Situations as a measure of emotional intelligence as detailed in this study (and an earlier study) gives teachers a productive approach to understanding this significant concept and to encourage the further development of practical approaches to programs based on an understanding of emotional intelligence as a capacity to learn.


Austin, E., Saklofske, D. & Egan, V. ( 2005). Personality, well being and health correlates of trait emotional intelligence. Personality and Individual Differences, 38, 547-558.

Barchard, K. & Russell, J. (2004). Psychometric issues in the measurement of emotional intelligence. In G. Geher (Ed.), Measuring Emotional Intelligence: Common Ground and Controversy. New York. Nova Science Publishers.

Bar-On, R. (1977). Emotional Quotient Inventory: Technical Manual. Toronto. MultiHealth Systems.

Brackett, M., Mayer J. & Warner, R. (2004). Emotional intelligence and its relation to everyday behaviour. Personality and Individual Differences, 36(6), 1387-1402.

Caruso, D., Mayer, J. & Salovey, P. (2002). Relation of an ability measure of emotional intelligence to personality. Journal of Personality Assessment, 79(2), 306-320.

Day, A. (2004). The measurement of emotional intelligence: The good, the bad and the ugly. In G. Geher (Ed.), Measuring Emotional Intelligence: Common Ground and Controversy. New York. Nova Science Publishers.

Dulewicz, S. & Higgs, M. (1999). Can emotional intelligence be measured and developed? The Leadership and Organisational Development Journal, 20(5), 242-252.

Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelligences. New York. Basic Books.

Goleman, D. (1997). Emotional Intelligence. New York. Bantam Books.

Higgs, M. (2001). Is there a relationship between the Myers-Briggs Type Indicator and emotional intelligence? Journal of Managerial Psychology, 16(7/8), 509-533.

Kafetsios, K. (2004). Attachment and emotional intelligence abilities across the life course. Personality and Individual Differences, 37(1), 129-145.

Lawrence, G. (1998). Emotional intelligence: Does it develop differently in thinking types and feeling types? In Proceedings of the Conference: Counter attack; Rising to the challenges to education: The role of psychological type. Orlando, March 5-8.

Matthews, G., Zeidner, M. & Roberts, R. (2002). Emotional Intelligence: Science and Myth. The MIT Press. Cambridge, Massachusetts.

Mayer, J. & Cobb, C. (2000). Educational policy on emotional intelligence: Does it make sense? Educational Psychology Review, 12(2), 163-183.

Mayer, J. & Salovey, P. (1993). The intelligence of emotional intelligence. Intelligence, 17, 433-442.

Mayer, J. & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D. Sluyter (Eds.), Emotional development and emotional intelligence: Educational implications. New York. Basic Books.

Mayer, J., Salovey, P. & Caruso, D. (1999). Test Manual for the Mayer, Salovey, Caruso Emotional Intelligence Test: Research Version 1.1. (3rd edition) Toronto Multihealth Systems.

MacCaan, C., Matthews, G., Zeidner, M. & Roberts, R. (2004). The assessment of emotional intelligence: On frameworks, fissures, and the future. In G. Geher (Ed.), Measuring Emotional Intelligence: Common Ground and Controversy. New York. Nova Science Publishers.

McGrath, H. & Noble, T. (1995). Seven ways at once: Classroom strategies based on the seven intelligences, Book 1. Melbourne: Longman.

Mendes, E. (2003). The relationship between emotional intelligence and occupational burnout in secondary school teachers. Dissertation Abstract 2003-95008-284. Dissertation Abstracts International Section B: The Sciences & Engineering, Vol 63(10-B), June 2003 4951. US: Univ Microfilms International.

Myers, I. & McCaulley, M. (1985). Manual: A guide to the development of the Myers-Briggs Type Indicator. Palo Alto: Consulting Psychologists Press.

Myers, I., McCaulley, M., Quenk, N. & Hammer, A. (1998). MBTI Manual: A guide to the development and use of the Myers-Briggs Type Indicator, 3e. Palo Alto: Consulting Psychologists Press.

Nardi, D. (2001). Multiple intelligences and personality type: Tools and strategies for developing human potential. Huntington Beach: Telos.

Pearman, R. (2002). Introduction to type and emotional intelligence: Pathways to performance. Palo Alto: Consulting Psychologists Press.

Pearman, R. & Fleenor, J. (1997). Sixteen patterns of leadership effectiveness: A multivariate analysis of observational variables and the MBTI. In Demarest, L., & Golatz, T. (Eds.), Proceedings of the Second International Research Conference: The Myers-Briggs Type Indicator and Leadership. Gainesville : Center for Applications of Psychological Type.

Perry, C., Ball, I. & Stacey, E. (2004). Emotional intelligence and teaching situations: Development of a new measure. Issues in Educational Research, 14(1), 29-43. http://www.iier.org.au/iier14/perry.html

Petrides, K. & Furnham, A. (2000). On the dimensional structure of emotional intelligence. Personality and Individual Differences, 29, 313-320.

Salovey, P. & Mayer, J. (1990). Emotional intelligence. Imagination, Cognition and Personality, 9, 185-211.

Schutte, N., Malouff, J., Hall, L., Haggerty, D., Cooper, J., Golden, C. & Dornheim, L. (1998). Development and validation of a measure of emotional intelligence. Personality and Individual Differences, 25(2), 167-177.

Sjoberg, L. (2001). Emotional intelligence: A psychometric analysis. European Psychologist, 6(2), 79-95.

Sternberg, R. (1985). Beyond IQ: A Triarchic Theory of Human Intelligence. New York: Freeman.

Van Rooy, D., Alonso, A. & Viswesvaran (2005). Group differences in emotional scores: theoretical and practical implications. Personality and Individual Differences, 38(3), 689-700.

Wu, S. (2004). Development and application of a brief measure of emotional intelligence for vocational high school students. Psychological Reports, 95(3), 1207-1218.

Authors: Dr Chris Perry (email: perryac@deakin.edu.au) is a Faculty Fellow in the Faculty of Education, Deakin University. She combines school based consultancies with research in the area of individual's growth and development especially as it relates to cognitive development.

Mr Ian Ball (email: gmagpa@bigpond.net.au) is currently a Faculty Fellow at Deakin University, and manages the Psychological Type Research Unit. Since retirement as an Associate Professor, he has been involved in conducting school reviews and contributing to a range of consultancies for education authorities.

Please cite as: Perry, C. and Ball, I. (2005). Emotional intelligence and teaching: Further validation evidence. Issues In Educational Research, 15(2), 175-192. http://www.iier.org.au/iier15/perry.html

[ Contents Vol 15 ] [ IIER Home ]
© 2005 Issues In Educational Research. This URL: http://www.iier.org.au/iier15/perry.html
Created 20 Feb 2006. Last revision: 20 May 2006.
HTML: Clare McBeath [c.mcbeath@bigpond.com] and Roger Atkinson [rjatkinson@bigpond.com]