Beyond Emotional Intelligence: Validation of a Model of Emotional Competence Applied to Teachers

Abstract

Is there an approach that goes beyond the debate between models that argue that emotional intelligence is a skill and those that maintain this construct is a personality trait? This study aims to validate a theoretical model of competent ‘emocreativity’ and to study the psychometric properties of the instrument that measures the emotional competencies included in this model. The participants were 177 teaching professionals and 370 students enrolled in university teacher training programs, who completed two instruments: the Self-Assessment Questionnaire of Emotional Competencies (CACE) and the TMMS24 using a Google form. The results indicate that the CACE has robust psychometric properties, given that the exploratory factor analysis (EFA) is confirmed by the confirmatory factor analysis, and the model’s components show internal consistency. In addition, its convergent validity has been compared with another Emotional Intelligence questionnaire (TMMS24). The conclusion is that the CACE responds to the factorial structure of the initial theoretical model, organized into four factors: emotional awareness, emotional change, emotional bonding, and emotional creativity, thus offering empirical validation of an instrument that can complement traditional approaches to the study of emotional intelligence.

1. Introduction

Emotional competencies are those that enable us to perceive, assess, and express our own emotions and understand those of others (Mendoza & Rueda, 2022). These skills include intrapersonal aspects, such as managing one’s own emotions, and interpersonal aspects, such as commitment to others, empathy, and teamwork (Busso et al., 2017; Rodríguez, 2018; Kostiv, 2022). They are essential for learning, professional performance, mental health, and individual and collective well-being (Figueroa & Rodríguez, 2023). Therefore, they appear to be crucial for both academic and social success (Zych et al., 2022).

1.1. Approaches to Emotional Intelligence

One of the references for the construct of emotional intelligence (EI) is Mayer and Salovey’s (1997) four-branch model. The authors conceive of EI as a skill, that is, a cognitive-affective ability to manage emotions through thought. For these authors, it involves four fundamental skills: ‘accurately perceiving, assessing and expressing emotions; accessing and/or generating feelings that facilitate thinking; understanding emotions; and regulating emotions, promoting emotional and intellectual growth’ (Mayer & Salovey, 1997, p. 5). The idea behind this approach to EI is that individuals perceive and feel their own emotions (emotional awareness), understand them and observe their evolution (emotional understanding), and manage them in themselves and with others (Machado, 2022). Thus, emotions are not only confined to the private sphere, but are incorporated into the social sphere (Dahlgren, 2018).

Another approach to EI is to conceive of it as a set of personality traits linked to a sense of emotional self-efficacy (Petrides & Furnham, 2001). This differentiation in the conception of emotional intelligence (as a skill and as a personality trait) has been supported by various meta-analyses (Joseph & Newman, 2010; Martins et al., 2010; Matthews et al., 2012; Petrides, 2011; Van Rooy et al., 2005), as pointed out by Kostiv (2022).

A third perspective on the study of EI is the competency approach. This assumes that one can be effective in intrapersonal and interpersonal socio-emotional interactions (Collie, 2020), that this way of dealing with emotions can be learnt, and, therefore, that it is teachable (Rodríguez, 2018). As Kostiv (2022) points out, while emotional intelligence as a general construct refers to the potential to deal with one’s feelings using reason and other motivational and personality dispositions, competent emotionality (Rodríguez, 2018) refers to a person’s ability to apply that potential to effectively resolve different situational demands, which are susceptible to change given their greater connection to context.

1.2. Emotional Competencies

The theoretical proposal presented in this paper for validation, which we have called ‘competent emocreativity’ (Rodríguez, 2018), is based on the two skills that have already been agreed upon by the models cited, specifically the ability to be aware of one’s emotions (emotional awareness) and to adjust them (emotional regulation). In addition, others have been incorporated, such as emotional recycling or the ability to modify maladaptive emotions; emotional bonding such as the ability to establish healthy and stable bonds and relationships with others; and emotional creativity referring to the ability to open up to new experiences and undertake life projects.

Competent emocreativity is based on emotional awareness, which involves perceiving emotions, recognising them as one’s own, and understanding their evolution. Starting from this prerequisite, one can begin to manage emotions, i.e., emotional change, which involves taking responsibility for one’s own emotions, establishing self-control when necessary, regulating their experience and modifying them when appropriate by displacing or replacing maladaptive emotions. Human emotions are necessarily social, which is why the model establishes emotional bonding as a necessary competence, which involves empathising with other people, communicating with them emotionally and committing emotionally, showing them appreciation and affection. Finally, a particularly novel aspect of the model refers to the competence of emotional creativity, which involves the ability to be self-confident; showing oneself to be an original person who is different from others (creative self-confidence); being open to change and innovation in one’s own life and experience; and, finally, having the initiative to develop projects aimed at community well-being (Rodríguez, 2018, p. 44) (Figure 1).

There is evidence of the relationship between emotional and creative aspects (Averill, 1999a, 1999b). In fact, it has been proven that deeper levels of emotional awareness, i.e., those that involve recognising and understanding emotions, have a greater impact on emotional creativity than more superficial levels such as the mere perception of emotions (Hernández-Jorge et al., 2020).

1.3. Emotional Competence and the Teaching Profession

As noted at the outset, it is important for us to manage our emotions appropriately, because they contribute to social well-being and professional or occupational well-being. Specifically, emotional management is essential in professions that require direct contact with people, such as education. Perhaps for this reason, the educational community has been the subject of various studies aimed at improving emotional intelligence, working with different groups ranging from teachers and schoolchildren to families and management teams (Lorenzo, 2017).

In this regard, it has been observed that when teachers possess emotional competencies, they promote a positive classroom environment and improve students’ academic, social, and emotional outcomes (Brackett & Salovey, 2007; Jennings et al., 2017). In this way, teachers can act as models for the emotional and interpersonal behaviours of their students (Vergaray et al., 2021). This is clearly marked by teacher–student interaction, which must have an affective and close component, where emotions play a fundamental role (Di Fabio, 2015). In fact, to teach and learn, the relationship between teachers and students must be filled with favourable feelings (Sehnem et al., 2021). A relatively recent review points out that a positive teacher–student relationship plays an important role in learning and performance and is associated with students’ school adjustment and social skills (L. García et al., 2019). In short, there is ample evidence that teachers’ awareness of their social and emotional functioning plays a central role in their ability to teach effectively, build appropriate interpersonal relationships with students, and foster a positive classroom climate (Granziera et al., 2023; Madigan & Kim, 2021; Mennes et al., 2024; Wang et al., 2023; Klusmann et al., 2022).

In addition to the influence that teachers’ emotional competencies have on different aspects of their students, we also consider it essential that teachers possess these competencies not only in their professional lives but also in their personal lives. In fact, as Collie et al. (2025) point out, in addition to addressing teachers’ self-efficacy regarding teaching, recent research has been addressing aspects of teachers’ socio-emotional self-efficacy, both in terms of their social interaction with other professionals and their perception of emotional self-efficacy, such as understanding or regulating their own emotions.

This line of research has been developed with both practicing teachers and trainee teachers at primary and secondary education levels, using primarily self-reported questionnaires. Some results reveal that perceiving that they manage their emotions well and their commitment to teaching them to students is related to the effectiveness that teachers perceive in managing their classroom and interacting with their students (Goegan et al., 2017). Collie (2025) related teachers’ emotional competencies to the five major personality factors. In her study, the author observed that openness, extraversion, kindness, and conscientiousness are positively related to perceived emotional competencies, while emotional instability is negatively related to this variable. At the same time, she indicates that emotional competencies are associated with higher levels of well-being and fewer intentions to change jobs, noting that emotional competencies explain these two variables more than others such as gender or professional experience.

1.4. The Vulnerability of the Teaching Profession

The importance of emotional skills as perceived by teachers has become more evident in situations of vulnerability. One such situation was experienced during the COVID-19 pandemic. A report by the Spanish Mental Health Confederation (2021) indicates that 90% of teachers suffered from sleep disturbances, nervousness, irritability, or tension (89.5%), and feelings of unhappiness with their work (42.7%). Moreover, after lockdown, teachers took on new responsibilities related to monitoring their students’ health, in addition to their educational duties, which may have caused psychological difficulties leading to stress and burnout (Cuesta et al., 2023).

Kostiv (2022) indicates that one of the lessons we must learn after suffering the global pandemic crisis is the urgency of addressing the motivational and emotional consequences that adversity has on education, and specifically on members of the educational community. In fact, after the pandemic, teachers say that managing emotions and motivation for learning are some of their main concerns, while other aspects such as content acquisition take a back seat (Trujillo Sáez et al., 2020).

Currently, this vulnerability is not just a circumstantial issue, but rather the teaching profession itself has a high component of emotional exhaustion. Specifically, a relationship has been detected between the emotional commitment that teachers establish with their students and the tendency to develop affective disorders. It therefore seems that the bond that teachers establish with their students can lead them to become emotionally ‘drained’ and increase their vulnerability, precipitating them towards affective imbalance. This happens when teachers lack the skills to understand and regulate their emotions (Chinea, 2020). In short, if teachers form emotional bonds with their students without the skills that help them cope with the emotional strain this causes, it could lead to a deterioration in their mental health (Rodríguez & Batista, 2024).

It is therefore important to equip these professionals with emotional skills that help them cope, both in their initial and ongoing training, as suggested by various authors such as B. García (2018) and Kostiv (2022) and institutions such as UNESCO (2020). One line of work has focused on training teachers in emotional skills, which has been found to have a positive impact on their well-being, resilience, and, in general, emotional skills, thus providing them with strategies to help them cope with adversity (Pozo-Rico et al., 2023). This applies to both active teachers and those in training (Zych et al., 2022).

1.5. Assessment of Emotional Competencies

Three types of instruments have been used to assess emotional competencies in educational contexts: questionnaires or self-reports, performance measures, and assessments carried out by external observers (Extremera & Fernandez-Berrocal, 2015). These latter hetero-reported measures are based on the perceptions of parents, teachers, and classmates regarding the emotional abilities of students (Denham et al., 2012; Kliewer et al., 2016). Specifically, a recent study provides a psychometrically validated hetero-evaluative instrument that assesses the emotional and creative competencies of schoolchildren from teachers’ perspectives, whose factorial structure is organized into three dimensions: emotional awareness, emotional regulation, and creativity (Hernández-Jorge et al., 2021).

As already mentioned, self-assessment measures based on the person’s own assessments of their emotional behaviour have also been used. Examples of these instruments are the TMMS-24 in its Spanish adaptation (Fernandez-Berrocal et al., 2004), the Bar-On EQ-i: YV (Bar-On & Parker, 2000), Collie’s (2025) Positive and Social Emotional Competencies (PSEC) Scale, and performance measures such as the MSCEIT (Mayer & Salovey, 1997) and Fernandez-Berrocal’s (2015) TIEFBA (Botin Foundation Emotional Intelligence Test).

Different studies have used self-assessment measures of teachers’ emotional competencies, obtaining different evidence. In general, it seems that teachers feel more pleasant than unpleasant emotions depending on the context in which they find themselves (Becker et al., 2014), and that they perceive and regulate their emotions appropriately with the aim of generating emotional and intellectual growth (Garrido & Gaeta, 2016). There is also evidence that future teachers who perceive themselves as emotionally competent also tend to perceive themselves as self-efficacious (Romero et al., 2022). Another important aspect is that there are differences between future teachers and practicing teachers, as trainee teachers tend to show less development of emotional skills than practicing ones, as demonstrated, among others, by the research carried out by Pertegal-Felices et al. (2011), Martínez-Saura et al. (2022), Goegan et al. (2017), Zambrano-Mendoza (2021), and Khunaivi et al. (2023). This suggests that professionals probably develop these competencies as they progress in their profession.

In this study, we have assessed the emotional competencies of teachers using a self-assessment tool that captures the dimensions proposed by Rodríguez (2018) in his concept of ‘competent emocreativity.’ We believe that this model broadens the traditional construct of emotional intelligence (EI) for several reasons. First, its approach is competency-based (Mikolajczak, 2009), that is, it isolates a contextual ‘know-how’ with emotional experiences, which we believe transcends the debate between trait EI and skill EI. It also addresses its measurement from a self-assessment perspective. Second, it broadens its vision of this construct by adding new competencies for managing human emotions, such as emotional recycling and the ability to modify emotions using feelings (Greenberg, 2023), and which, in our case, we have called transformative feelings, such as love, compassion, or forgiveness. In addition, it includes another divergent competence, namely emotional creativity (Averill, 1999a, 1999b; Ivcevic et al., 2007; Rodríguez, 2022), which refers to the ability to feel emotionally unique, affirming oneself in that uniqueness, opening oneself up to discovering new experiences, and undertaking life projects. Third, this model transcends the exclusively intrapersonal approach to address the inclusion of emotional competencies that help establish emotional bonds with others, which enable healthy and mutually enriching interpersonal relationships, using skills such as empathy, emotional communication, and commitment to others.

Therefore, the objective of this research is to empirically test the model of ‘competent emocreativity’ (Rodríguez, 2018) through the validation of an instrument that uses the appropriate psychometric properties to be able to isolate, with guarantees, the dimensions included in it. Additionally, this evaluative tool is applied to study the emotional competencies of teachers, both in initial training and in active service.

2. Materials and Methods

2.1. Participants

A total of 547 primary and secondary school teachers participated in this study, of whom 370 were trainee teachers and 177 were active teachers. The active teachers belong to public, private, and charter schools in the Canary Islands (Spain). This group consisted of 137 women and 40 men, with an average age of 42.5 years old. Regarding the participants in training, 268 are women and 101 are men, with an average age of 21.3 years old. All of this is summarized in

Table 1. Characteristics of participants.

Samples	Participants	Age		Sex
		$\overline{\mathbf{x}}$	Sx	Male	Female	Other
Active teachers	177	42.5	12.30	40	137
Trainee teachers	370	21.3	3.25	101	268	1
Total	547	28.2	12.4	141	405	1

.

2.2. Instruments

We have used two instruments. The instrument to be validated is the Self-Assessment Questionnaire on Emotional Competencies (CACE, acronym in Spanish) aimed at primary and secondary school teachers. It was developed based on the curricular competencies and assessment criteria included in the curriculum for the area of ‘Emotional and Creativity Education’ in primary education, prepared by the Department of Education of the Autonomous Government of the Canary Islands (2022). These competencies and criteria are emotional awareness (the perception of one’s own emotions, their recognition and understanding); emotional change (emotional responsibility, self-control, adjustment, and emotional recycling); emotional connection (empathy, affective communication, and emotional commitment); and emotional creativity (creative self-confidence, openness to change and innovation, and a vital entrepreneurial attitude).

The questionnaire consists of 13 statements, including a definition of emotional competence for each of the above curriculum assessment criteria, which are further enriched with contextual examples. Participants rate on a Likert scale from 0 to 5 (0 = never, 5 = always) their perception of the use of each of the competencies included in the questionnaire, which are based on the adaptation of Rodríguez’s (2018, p. 44) model to the curriculum developed by the Autonomous Government of the Canary Islands, in the update made to it with the implementation of the LOMLOE law in 2022, and which are itemised below.

Emotional awareness:

• Emotional perception: Becoming aware of the bodily sensations associated with emotional experiences.
• Emotional recognition: Identifying one’s own emotions and those of others and accepting them without dissociating oneself from them.
• Emotional understanding: Understanding and analysing the emotions one experiences, their relationship to what has happened before, and their consequences.

Emotional change:

• Emotional responsibility: Taking responsibility for one’s own behaviour, the consequences of one’s feelings, and the harm that can be done to oneself and others through behaviour derived from one’s emotions.
• Self-control: Regulating emotional impulsivity and its expression through reflexivity, tolerance for frustration, and overcoming difficulties.
• Emotional adjustment: Regulating the experience of one’s own emotions, and adjusting their intensity and duration, particularly those that manifest themselves in a maladjusted way and cause discomfort.
• Emotional recycling: Endogenously promoting adaptive emotions and states of emotional security, through which maladaptive emotions that are experienced can be displaced or replaced.

Emotional bonding:

• Empathize: Identify, understand, and interpret other people’s emotions and experience emotions based on the feelings of others.
• Emotional communication: Expressing one’s own emotions, needs, and experiences to others without discomfort and demonstrating authenticity and closeness, as well as a willingness to listen actively and respectfully to the feelings shared by others.
• Emotional commitment: Connecting emotionally with others, showing appreciation, affection, and value toward them, and developing altruistic and supportive behaviours.

Emotional creativity:

• Creative self-confidence: Self-assurance when expressing oneself as an original person who is different from others.
• Openness to change and innovation: Interest in unusual experiences and ideas and receptivity to what is different, new, or alternative, as well as a willingness to explore reality through the senses.
• Life entrepreneurship: Initiative to propose, develop, and implement entrepreneurial projects aimed at community well-being.

We also used the TMMS-24, which assesses emotional intelligence. This questionnaire is based on the TMMS (Trait Meta-Mood Scale) developed by the Salovey and Mayer (1990) research group. The original scale is a trait scale that assesses metaknowledge of emotional states using 48 items, which has been adapted to the Spanish population (Fernandez-Berrocal et al., 2004). This version consists of three dimensions with eight items each: emotional awareness (the ability to be aware of and pay attention to one’s own emotions); emotional clarity (the ability to understand and be clear about one’s own emotions); and emotional repair (the ability to regulate and manage emotions effectively).

2.3. Procedure

The project was submitted to the Research Ethics and Animal Welfare Committee of the University of La Laguna (CEIBA), which considered it to comply with the ethical requirements for this type of research and issued a positive report: CEIBA2024-3438.

Once CEIBA’s approval was obtained, the questionnaires were created in a Google form and sent to primary and secondary school teachers participating in a broader project to monitor the implementation of Emotional Education and Creativity (EMOCREA). The sample was incidental, as teachers who wanted to participate did so. The trainee teachers were studying for a teaching degree at the University of La Laguna in early childhood and primary education, and/or a master’s degree in teacher training at the same university. In the Google links, all participants were asked to complete the questionnaires, which set out the objective of the research and the informed consent to participate. Once the Excel template with the data was obtained, the data were analysed.

2.4. Data Analysis

Statistical analysis was performed using R software version 4.4.0 (R Core Team, 2024) through the RStudio interface version 2024.12.0.467 (Posit Team, 2024). To examine the latent structure of the emotional competencies questionnaire, an exploratory factor analysis (EFA) was performed sequentially, followed by a confirmatory factor analysis (CFA). The total sample (N = 547) was randomly divided into two independent subsamples: an exploratory subsample (N = 177) for the EFA and a validation subsample (n = 370) for the CFA. The selection of the optimal number of factors in the EFA was determined using the nfactors package (Raiche & Magis, 2022), applying varimax rotation under the assumption of factor independence. The parallel analysis and Kaiser criteria (retain factors with eigenvalues > 1) guided this decision. The internal consistency of the scales was assessed using Cronbach’s alpha (α) and McDonald’s omega (ω) coefficients, ensuring reliability in the measurements. Subsequently, the CFA was performed on the validation subsample using the lavaan package (Rosseel, 2012). Given the ordinal nature of the items, the weighted least squares maximum likelihood (WLSMV) estimator was used, applying scaling corrections to χ² to mitigate biases due to non-normality. The goodness of fit of the model was evaluated using multiple indices:

• χ²/gL: Values ≤ 3 indicated a satisfactory fit.
• CFI (Comparative Fit Index) and TLI (Tucker–Lewis Index): ≥0.90 (acceptable) and ≥0.95 (excellent).
• RMSEA (Root Mean Square Error of Approximation): ≤0.08 (adequate) and ≤0.05 (excellent).
• SRMR (Standardized Root Mean Square Residual): ≤0.08 as an acceptance threshold.

Additionally, three multiple linear regression models were built with the stats package R Core Team (2024) to assess the predictive power of emotional competencies on the TMMS-24 factors (attention, clarity and repair). Predictor variables were centred on the sample mean to reduce multicollinearity and to interpret effects in relation to population averages. Finally, the assumptions of normality, homoscedasticity, linearity, and independence were verified, ensuring the inferential robustness of the models.

To study the emotional competences between practising and trainee teachers, a comparison of independent sample means was carried out for each factor of the CACE questionnaire. A non-parametric analysis using the Mann–Whitney U statistic was used as the assumption of normality of the study variables was not met. A statistical significance level of p < 0.05 was established.

3. Results

3.1. Factor Analysis of the Self-Assessment of Emotional Competences Questionnaire

Figure 2 shows the results of the correlation study between the items of the self-assessment questionnaire on emotional competencies (CACE) that was conducted prior to the exploratory factor analysis.

The determination of the number of factors was supported by the results of the parallel analysis and the Kaiser criterion, which converged in the selection of a four-factor structure. This decision was visually supported by the correlation matrix and by the sedimentation plot, where a clear inflection point with four factors is observed (Figure 3).

Table 2. Rotated component matrix.

	Factors
	1	2	3	4
1. Emotional perception	0.63
2. Emotional recognition	0.93
3. Emotional understanding	0.73
4. Emotional responsibility		0.48	0.55
5. Self-control		0.70
6. Emotional adjustment		0.79
7. Emotional recycling		0.61
8. Empathising			0.86
9. Emotional communication			0.51
10. Emotional commitment			0.66
11. Creative self-confidence				0.62
12. Openness to change and innovation				0.66
13. Life entrepreneurship				0.63

shows the rotated coefficients, which range from 0.48 to 0.93. The rotated components are associated with the factors: emotional awareness (factor 1), emotional change (factor 2), emotional bonding (factor 3), and emotional creativity (factor 4). The item ‘emotional responsibility’ shows values above 0.4 in both factor 2 and factor 3. In line with the theoretical conceptualisation of the questionnaire, it was determined that this item should be included in the emotional change factor. The factorial solution explained 64% of the accumulated variance, divided into 18% emotional awareness, 15% emotional change, 19% emotional bonding, and 12% emotional creativity, demonstrating an adequate explanatory capacity.

To study the internal consistency of the CACE factors, Cronbach’s alpha (α) and McDonald’s omega (ω) coefficients were calculated for each of them. The results show that the CACE presents adequate internal consistency in each of the components (

Table 3. Internal consistency of the CACE factors.

Components	Elements	Cronbach’s α	McDonald’s ω
Emotional awareness	3	0.77	0.79
Emotional change	4	0.71	0.73
Emotional bonding	3	0.77	0.78
Emotional creativity	3	0.66	0.67

).

Confirmatory factor analysis indicated a good fit of the four-factor structure of our data (Figure 4). The results confirm that the proposed model adequately fits the empirical data: χ²/gL = 3, CFI = 0.968, TLI = 0.957; RMSEA = 0.075 (90% CI: 0.063–0.088); SRMR = 0.067.

3.2. Relationship Between the Self-Assessment Questionnaire of Emotional Competencies (CACE) and the TMMS-24 Emotional Intelligence Test

To evaluate the predictive power of the emotional competencies of the CACE on the TMMS-24 factors, three multiple linear regression models were performed, one for each factor (attention, clarity and repair).

Model 1, with the TMMS-24 attention factor as the criterion variable, explained 17.95% of the variance of said variable (R² = 0.179, adjusted R² = 0.169), with a residual standard error of 4.93. The global F statistic confirmed the robustness of the model (F(4, 338) = 18.44, p < 0.001), indicating that, as a whole, the evaluated predictors provide significant explanatory capacity for the attention variable. The analysis of the coefficients showed that both emotional awareness and emotional bonding had positive and highly significant effects on the attention factor of the TMMS-24. For each unit increase in these variables, an average increase of 0.67 and 0.63 points was associated with an increase in the attention variable, respectively. In contrast, emotional change and emotional creativity did not show a significant effect (

Table 4. Results of the analysis of the coefficients of model 1.

		EE	SE	t	p
Model 1: Attention	Emotional awareness	0.675	0.266	4.243	<0.001 **
	Emotional change	−0.046	0.122	−0.379	0.704
	Emotional bonding	0.632	0.179	3.530	<0.001 **
	Emotional creativity	0.154	0.121	1.269	0.205

Note: ** p < 0.001.

).

The model residuals showed a range between −14.50 and 11.27, with a median close to zero (median = 0.59) and balanced quartiles (Q1 = −3.13, Q3 = 3.39). The distribution was symmetrical, with no evidence of heteroscedasticity or violation of the assumptions of normality, as well as those of linearity and independence of the residuals (Figure 5).

Model 2, with the TMMS-24 clarity factor as the criterion variable, explained 40% of the variance of this variable (R² = 0.4, adjusted R² = 0.392), with a residual standard error of 5.23. The overall F statistic confirmed the robustness of the model (F(4, 338) = 56.32, p < 0.001), indicating that, taken together, the evaluated predictors provide a significant explanatory capacity for the attention variable.

The analysis of the coefficients showed that emotional awareness, emotional change, and emotional creativity presented positive and highly significant effects on the TMMS-24 clarity factor. For each unit increase in these variables, an average increase of 1.02, 0.81, and 0.31 points in the clarity variable was associated, respectively. In contrast, emotional attachment did not show a significant effect (

Table 5. Results of the analysis of the coefficients of model 2.

		EE	SE	T	p
Model 2: Clarity	Emotional awareness	1.028	0.282	6.080	<0.001 **
	Emotional change	0.817	0.130	6.275	<0.001 **
	Emotional bonding	0.173	0.190	0.909	0.363
	Emotional creativity	0.318	0.129	2.464	0.014 *

Note: ** p < 0.001; * p < 0.05.

).

The model residuals showed a range between −15.89 and 13.44, with a median close to zero (median = 0.59) and balanced quartiles (Q1 = −3.35, Q3 = 3.36). The distribution presents a slight asymmetry towards negative values, although without evidence of serious heteroscedasticity or violation of the assumptions of normality, as well as those of linearity and independence of the residuals (Figure 6).

Model 3, with the TMMS-24 repair factor as the criterion variable, explained 35.9% of the variance of this variable (R² = 0.359, adjusted R² = 0.351), with a residual standard error of 3.32. The overall F statistic confirmed the robustness of the model (F(4, 338) = 47.37, p < 0.001), indicating that, taken together, the evaluated predictors provide a significant explanatory capacity for the attention variable. The analysis of the coefficients showed that emotional change and emotional creativity presented positive and highly significant effects on the TMMS-24 clarity factor. For each unit increase in these variables, an average increase of 0.82 and 0.26 points was associated in the repair variable, respectively. In contrast, emotional awareness and emotional bonding did not show a significant effect (

Table 6. Results of the analysis of the coefficients of model 3.

		EE	SE	t	p
Model 3: Repair	Emotional awareness	−0.177	0.107	−1.656	0.098
	Emotional change	0.823	0.082	9.960	<0.001 **
	Emotional bonding	−0.013	0.120	−0.112	0.911
	Emotional creativity	0.260	0.082	3.169	0.001 **

Note: ** p < 0.001.

).

The model residuals showed a range between −11.98 and 9.30, with a median close to zero (Median = 0.20) and balanced quartiles (Q1 = −1.96, Q3 = 2.16). The distribution presents a slight asymmetry towards negative values, although without evidence of serious heteroscedasticity or violation of the assumptions of normality, as well as those of linearity and independence of the residuals (Figure 7).

3.3. Comparison of Emotional Competencies Between Practicing Teachers and Trainee Teachers

Mann–Whitney U tests were performed to compare the distributions of scores of practicing and trainee teachers on the four CACE factors. Due to significant violations of the normality assumption (Shapiro–Wilk: p < 0.001 for all factors), this non-parametric approach was chosen. The results are shown in

Table 7. Results of non-parametric comparisons of CACE factors between practicing teachers and trainee teachers.

	Teachers	Median	Median	SD	U	p	Rank Correlation
Factor 1	Practicing teachers	12	11.98	2.61	26,872.5	0.001	−0.179
	Trainee teachers	12	11.58	1.87
Factor 2	Practicing teachers	11	11.25	2.04	30,999.50	0.308	−0.053
	Trainee teachers	11	11.03	2.15
Factor 3	Practicing teachers	12	11.35	2.40	29,859.50	0.092	0.088
	Trainee teachers	12	11.73	2.04
Factor 4	Practicing teachers	11	11.13	2.50	29,031.50	0.031	−0.113
	Trainee teachers	11	10.55	2.75

. Statistically significant differences are observed in the factors of emotional awareness and emotional creativity. In both cases, in-service teachers present higher means than pre-service teachers, although the small effect size and medians coincide in the two groups. The factors emotional change and emotional bonding do not present statistically significant differences.

4. Discussion and Conclusions

The aim of this study was to empirically test the model of ‘competent emocreativity’ (Rodríguez, 2018) through the validation of an instrument that uses the appropriate psychometric properties to be able to isolate, with guarantees, the dimensions included in it.

A first conclusion suggests that the Self-Assessment Questionnaire of Emotional Competencies (CACE) responds to a factorial structure that is consistent with the pre-established theoretical model (Rodríguez, 2018), organized into four factors in the sample used to establish the principal component analysis. These factors are emotional awareness, emotional change, emotional bonding, and emotional creativity.

It should be noted that although the original model differentiates, from a theoretical and conceptual perspective, between emotional regulation and recycling, both competencies respond to the same principle of action on emotions. The first (regulation) involves adjusting the intensity, duration, and unpleasantness of the emotional experience, and the second (recycling) addresses it with the intention of compensating for, replacing, or transforming maladjusted feelings into more adaptive emotional responses. Therefore, it is consistent that the structural model has integrated these two competencies into one (emotional change), since they respond to the same logic of modifying and improving the way people experience their feelings to avoid succumbing to emotional distress.

Another item worth interpreting concerns ‘emotional responsibility,’ which appears to be shared by two factors: change and emotional connection. From a theoretical perspective, this ability would have a transversal nature, allowing it to be integrated into any of the isolated general competencies, as it connects with the content of each of them. We chose to include it in emotional change because it is more consistent with the reference theoretical model (Rodríguez, 2018) and its curricular implementation.

Regarding the psychometric merits of the instrument used, and specifically its reliability, measured by internal consistency, it was more than adequate in all the analyses and indices used. These results, together with the fact that the factorial structure was confirmed, using the sample of trainee teachers as a contrast group, lead us to conclude that the competent emocreativity model confirms its construct validity.

Once empirical confirmation of the model had been obtained, predictive validation with the emotional intelligence (EI) construct remained to be addressed. Through regression analysis, the models showed favourable data regarding the relationship between the CACE and the TMMS-24. Indeed, the factors of the former not only predict a significant proportion of variance in the variables of the latter, but also that the predictive relationship expresses a convergence between the content in both instruments. Specifically, the emotional awareness of the CACE (which includes emotional perception and recognition) and emotional connection (which includes empathy) predict the emotional attention factor of the TMMS-24 (which measures the ability to identify one’s own emotions and those of others). The CACE factors of emotional awareness (which also includes emotional understanding), change (which includes self-control, in terms of reflexivity), and emotional creativity have been confirmed in other studies to be closely related to understanding (Hernández-Jorge et al., 2020), and predict the clarity of feelings factor of the TMMS-24, which also measures the ability to understand emotions. Finally, the change (which includes regulation) and emotional creativity (which includes creative self-confidence) factors of the CACE predict the emotional repair factor, which is also fundamentally regulation.

This clear correspondence between both instruments indicates that they are evaluating the same construct (emotional intelligence) from different perspectives. The CACE uses a competency-based perspective (Mikolajczak, 2009), given the way its items are worded; and the TMMS-24, from the concept of EI ability, as its authors defend (Mayer & Salovey, 1997). This leads us to conclude that this predictive and content-based relationship serves as evidence for the convergent validation of the CACE instrument.

This interpretation is supported in the scientific literature by authors such as Mikolajczak (2009), who proposes the competency-based approach as an intermediate path between trait EI and EI ability models, consisting of the integration of cognitive and personality components into the same construct, adding the contextual dimension of performance adjusted to the task demand and environmental circumstances (Kostiv, 2022).

Finally, regarding the application of the model to the sample of practicing teachers versus trainee teachers, the results indicate that the former perceive themselves as more competent in their ability to perceive, recognize, and understand their emotions (emotional awareness) and in feeling confident in expressing unique creativity, as well as being open to new ideas and formulating projects to improve their environment (emotional creativity). There are no differences in the other two CACE factors (emotional change and connection).

The results are broadly consistent with previous studies that have compared the emotional competencies of these groups of teachers. Pertegal-Felices et al. (2011) found that trainee teachers displayed consistently lower levels than practicing teachers. These findings were subsequently corroborated by Martínez-Saura et al. (2022), who found that practicing teachers displayed significantly higher levels of emotional clarity and repair compared to trainees. Likewise, the study by Zambrano-Mendoza (2021) found that only 21.1% of trainee teachers consistently developed emotional competencies, while 44.4% implemented them occasionally, suggesting a significant gap in early professional development. Khunaivi et al. (2023), in their qualitative study with trainee teachers, found that emotional competencies were essential for preventing teacher burnout, improving well-being, and creating a positive learning environment. This convergence of results with those obtained in our research could be interpreted by considering that practical experience can be a crucial factor in the development of these skills, especially in the case of emotional awareness, given that it is a skill that appears in any emotional management process as a starting point. Similarly for emotional creativity, the greater the experience, the greater the self-confidence to feel capable of coping with the challenges that teaching demands.

Even after achieving empirical validation of our model, a more detailed study of the convergence of the factors isolated in the CACE with other assessment tests of emotional management skills from a competency perspective remains pending. We must also be aware of the limitations that may be imposed by the self-report bias of the instrument, as well as the limitations of the cross-sectional design of our research.

It would also be relevant to address the generalizability issues inherent in the sample by applying the instrument and the underlying model to other professionals. This would confirm that the ‘competent emocreativity’ model is applicable in diverse contexts and with different samples of participants.

Another critical aspect to highlight is the redundancy of the ‘emotional responsibility’ item in several factors. Although we have addressed this issue using a criterion of consistency with the theoretical model, it is necessary to investigate empirically the potential transversal quality of this competency.

To sum up, we believe that the results of this study not only provide a reliable and valid assessment tool, but also a useful empirical model from a competency-based perspective, one that has learning potential from which to design both initial and ongoing training aimed at promoting teacher acquisition of the same emotional competencies that should be developed in students through the ‘Emotional and Creativity Education’ curriculum area.

We believe that our research provides implications from which to formulate an empirically proven training proposal aimed at teachers, enabling them not only to help their students learn how to manage their complex and increasingly damaged emotional world, but also to take on the urgent challenge of protecting themselves by looking after their professional and personal well-being.

The above has not only practical implications, but also consequences for the design of training policies. As we have shown, the importance of introducing emotional intelligence into initial teacher training has been present in the literature for twenty years (Bisquerra Alzina, 2005; Fernández-Berrocal & Ruiz, 2008; Teruel, 2000); however, the majority of teachers have not had training opportunities to develop their emotional skills. Although the range of training options has increased in recent years (Palomera et al., 2008), there is no guarantee that these proposals are based on research results, nor that they are experiential (Kostiv, 2022) and skills-based, as we propose in the model validated in this work.

In conclusion, this study has provided empirical evidence for approaching teacher training from an emotional intelligence model, which, without claiming to be superior to traditional models (because this was not what we sought to prove with our research), offers an alternative focused on emotional skills that can be acquired through systematic teaching–learning processes by both students and teachers.