Mediating and Moderating Variables in the Relationship Between Emotional Demands and Teachers’ Emotional Exhaustion

Introduction

Burnout typically applies to employees who experience chronic stress in their work environment and are unable to carry out adequate coping mechanisms (Grau Martín, 2007). The World Health Organization (WHO, 2019) added burnout as a category of problems associated with employment or unemployment to its International Classification of Diseases (ICD, version 11). This syndrome leads to several adverse outcomes, including an adverse impact on health, professional development, and individual performance. Moreover, burnout negatively affects the organizations of employees suffering from it (Mababu, 2016).

As an increasingly prevalent phenomenon in many societies, burnout is defined as being composed of three distinct dimensions: emotional exhaustion, depersonalization, and reduced personal accomplishment (Maslach & Jackson, 1981a). Of these three dimensions, emotional exhaustion is regarded by many authors as the central and defining element of burnout (Pines & Aronson, 1988; Schaufeli & Salanova, 2007; Skaalvik & Skaalvik, 2017). Then, our study focuses on this factor.

Emotional exhaustion occurs in the context of what Maslach and Pines (1977) defined as “professional care” – work that involves continuous interaction with others. Teaching, the focus of this study, has been identified as a job with a high likelihood of burnout risk (Arce et al., 2012; Moreno et al., 2000; Otero-López et al., 2008) due to the frequency of a strong personal and emotional engagement (Schaufeli et al., 2009).

Studying the variables that serve as risk and protective factors for exhaustion may help identify strategies to reduce its negative effect on teachers. Thus, the main objective of this study is to analyze the relationships of different variables (emotional demands, emotional exhaustion, and self-efficacy to cope with stress) and with emotional exhaustion in teaching as a job.

Relationship between Emotional Demands and Emotional Exhaustion

Emotional demands refer to the psycho-affective loads resulting from job-specific tasks, processes, or specific conditions (González et al., 2015) entailed by task achievement (de Puelles Casenave, 2009). Maslach and Jackson (1981b) argued that emotional demands are the first step in the continuum of worker burnout. Several studies have shown that emotional demands have a negative impact on workers’ mental health. For instance, it has been demonstrated that both the emotional demands derived from the management of emotions centered on oneself and other-oriented demands cause an increase in the perception of job stress and anxiety and decrease job satisfaction (Pugliesi, 1999). Additionally, research has shown that high emotional demands in the field of education lead to increased work stress among teachers (Rodríguez et al., 2017). Similar findings have been observed not only in the teaching field (Tuxford & Bradley, 2015), but also in other jobs, including healthcare (Mazzetti et al., 2020).

The cross-sectional and longitudinal analyses carried out by Vammen et al. (2019) revealed that an increase in emotional demands can result in moderate but significant increases in emotional exhaustion. It is unclear whether this modest impact could be a result of other factors influencing or moderating the association between emotional demands and emotional exhaustion. Several studies have shown that the relationship between emotional demands and burnout is indeed mediated by other variables, such as coping mechanisms (Peng et al., 2010), stress (Maslach & Pines, 1977; Rodríguez et al., 2017), or emotional dissonance (Lewig & Dollard, 2003).

Teaching is a job that has traditionally been associated with significant emotional demands (Baeriswyl et al., 2021; Cejudo & López-Delgado, 2017; Muñoz et al., 2009; Rodríguez et al., 2017). This is mainly because teachers, tasked with boosting interpersonal relationships in a school environment, must efficiently regulate their own emotions while managing and/or being mindful of the emotions of others (Cejudo & López-Delgado, 2017). Moreover, teachers are often exposed to high emotional situations at work, including confrontational interactions between and from students. Additionally, teachers belong to a professional group that experiences high levels of emotional dissonance due to the nature of their work, which involves frequent face-to-face interactions with students, colleagues, and parents, and needs to hide or suppress their true emotional state. As such, the examination of the possible mediating role played by emotional dissonance in the connection between emotional demands and exhaustion is particularly relevant.

Emotional Dissonance as a Mediator

Hochschild (1983) explained that emotional dissonance occurs when there is a discrepancy between emotions felt and expressed due to the need to adapt to a situation. On the job, this manifests as a conflict between genuine and expected emotions (Rafaeli & Sutton, 1987). It is reasonable to anticipate a correlation between increasing rates of emotional dissonance and increasing occurrences of emotional exhaustion. The added tension in work environment stemming from emotional dissonance results in additional demands on individuals, leading to heightened levels of exhaustion.

There is a clear relationship between emotional dissonance and exhaustion, as indicated by several studies conducted with police officers, health professionals, teachers, and others (Andela & Truchot, 2017; Andela et al., 2016; Bakker & Heuven, 2006; Heuven & Bakker, 2003; Karatepe & Dare Aleshinloye, 2009; Kwak et al., 2018; Molino et al., 2016). Likewise, research suggests that individuals experiencing emotional dissonance may have reduced resources to cope with the resulting tension, leading to higher levels of emotional exhaustion. This phenomenon is particularly prevalent in teaching (Simbula et al., 2019). Similarly, Tuxford and Bradley (2015) found that emotional dissonance contributes to increased emotional exhaustion in the context of teaching.

It is important to note that the literature has linked emotional demands to emotional dissonance. In the workplace, especially in face-to-face jobs, where workers need to maintain a positive demeanor and are expected to conceal negative emotions, an increase in emotional demands (that implies a more negative emotional state) should be likely to result in heightened emotional dissonance levels. Workers must suppress or hide their emotional tensions and negative feelings due to the high emotional demands that stem from their work context. In this sense, van Gelderen et al. (2007) argued that emotional demands at work are a significant starting point for experiencing emotional dissonance.

Several studies examining face-to-face workers have evidenced the relationship between emotional demands and emotional dissonance. In the context of call-center workers, Molino et al. (2016) found that as workloads increase, emotional dissonance levels increase as well. In a similar vein, Yulianti and Nur Madina (2018) found that police detectives who present higher levels of emotional effort and emotional demands are more likely to experience emotional dissonance. Previous studies have also shown that emotional dissonance can mediate the link between emotional demands and burnout, first with call center workers (Lewig & Dollard, 2003) and later with nurses and police officers (Bakker & Heuven, 2006). Thus, the effect of emotional demands on emotional exhaustion (Lewig & Dollard, 2003) and burnout (Bakker & Heuven, 2006) through emotional dissonance has been studied in different workplace contexts (Lewig & Dollard, 2003).

By applying previous findings to the context of teaching, it is expected that high emotional demands experienced by teachers will result in higher emotional dissonance. Teachers may attempt to hide the negative emotions derived from the high emotional demands of their job, ultimately leading to increased emotional exhaustion levels. This suggests that emotional demands within teaching can lead to emotional exhaustion due to the experience of emotional dissonance. Due to the nature of their work, teachers often find themselves in social and interactive situations that trigger emotional responses. Additionally, they are required to hide their emotions that appear in response to emotionally taxing situations. As a result of these job characteristics, we suggest that emotional demands associated with teaching can cause emotional dissonance, which ultimately leads to heightened emotional exhaustion among teachers.

Then, the following mediational hypothesis (H1) is posed: emotional dissonance plays as a mediator in the relationship between emotional demands and emotional exhaustion in teachers, such that higher levels of emotional demands in teachers will trigger higher levels of emotional dissonance, which will in turn produce higher levels of emotional exhaustion.

Self-efficacy as Moderator

Self-efficacy refers to an individual’s perception of their ability to effectively carry out a specific task (Bandura, 1977). Self-efficacy has traditionally been examined as a moderator variable in the scientific literature, particularly in association with burnout (Chacón Corzo, 2006; Fernández Arata, 2008; Pajares, 1996). Grau et al. (2001) concluded that the relationship between work stress and burnout was modulated by employees’ self-efficacy beliefs. Specifically, workers with high self-efficacy did not report negative consequences associated with stress, whereas individuals with high stress levels but low levels of self-efficacy experienced greater emotional exhaustion. Makara-Studzińska et al. (2019) present evidence demonstrating the moderating role of self-efficacy in the relationship between stress and burnout among firefighters, concluding that self-efficacy is a resource deemed necessary in high-risk professions to mitigate the impact of perceived stress on burnout.

Focusing on the education field, Dicke et al. (2014) demonstrated that disturbances caused by students in classrooms can elevate teachers’ stress levels, but only when teachers present low self-efficacy for managing the classroom. Thus, self-efficacy was found to moderate the relationship between classroom disruptions and stress. For this reason, in the current study we speculate that because disturbances can potentially increase emotional dissonance on teachers (by increasing negative emotion that must be kept hidden), self-efficacy to cope with stress will likely have a moderating role in the relationship between emotional dissonance and exhaustion.

Put it another way, self-efficacy to cope with stress may act as a protective factor against the negative effects exerted by emotional dissonance on said exhaustion. Teachers who feel capable of effectively managing job-related stress will likely handle emotional dissonance resulting from social interactions in the classroom better, suffering in consequence lower levels of emotional exhaustion than teachers who experience high emotional dissonance but lack the ability to manage job-related stress. Along these lines, Tuxford and Bradley (2015) found that teaching self-efficacy buffered the adverse impact of emotional dissonance on emotional well-being. Thus, our second research hypothesis (H2) posits self-efficacy for coping with stress acts as a moderator in the relationship between emotional dissonance and emotional exhaustion, in such a way that it reduces the negative effects of emotional dissonance on emotional exhaustion.

Building on previous literature that has established how emotional demands predict emotional dissonance (Guglielmi & Tatrow, 1998; Rodríguez et al., 2017), that emotional dissonance predicts burnout (Andela & Truchot, 2017; Andela et al., 2016; Skaalvik & Skaalvik, 2017), and that self-efficacy often acts as moderator and protective factor of burnout (Chacón Corzo, 2006; Dicke et al., 2014; Fernández Arata, 2008; Grau et al., 2001; Makara-Studzińska et al., 2019; Pajares, 1996), our final moderated mediation hypothesis (H3) states that the mediated relationship between emotional demands, emotional dissonance, and emotional exhaustion is moderated by self-efficacy in coping with stress.

All the study hypotheses are schematized in Figure 1.

Method

Participants

Our study sample consisted of 108 teachers (57.3% women), with a mean age of 45.30 years (SD = 8.68, range = 28-68) who completed an online questionnaire at three time points. More details about the sample can be found in Appendix.

Procedure

The participants completed an online survey via the Unipark academic platform on three separate occasions during the academic year: at the beginning of school, halfway through the year, and at the end. Anonymity was ensured by providing codes. Informed consent was granted prior to questionnaire access. This study received approval from the Ethics Committee of [masked] under code [masked].

Instruments

Levels of emotional exhaustion were assessed using the five items of the emotional exhaustion factor from the Spanish validation of the Maslach Burnout Inventory-General Scale (Gil-Monte, 2002). Participants’ perception of emotional demands related to completing a task was measured using the six items of the Emotional Demands Scale (Heuven & Bakker, 2003), including questions such as “Do you have to deal with clients who do not treat you with the appropriate respect and politeness?”. The extent to which individuals perceived emotional demands in their work setting was evaluated using the Emotional Dissonance Scale (Moncada et al., 2014), which consists of five items, such as “During your work, how often do you have to express positive feelings towards your clients while you actually feel indifferent?”. The term “clients,” which was used in several items of the Emotional Demands Scale and the Emotional Dissonance Scales, was substituted with “students” in our instruments to align with the sample of teachers. The Self-efficacy Scale for Coping with Stressful Situations (Tabernero et al., 2009) was used to assess professionals’ capability to manage stress, and consisted of eight items. Participants responded to all instruments using a five-point Likert scale. The internal consistency of the variables in each of the evaluation periods was high (see Table 1).

Statistical Analysis

Different preliminary analyses were conducted. First, we performed repeated measures analyses to examine the stability of the study variables over time. Then, a Pearson correlation analysis was carried out to investigate the expected direction of the variables over time. Finally, a stepwise linear regression analysis was carried out to determine the percentage of variance explained by the predictive variables. In the initial model we included the variable proposed as the independent variable (emotional exhaustion). In the second model, we added the variable hypothesized as mediator (emotional dissonance). In the third model, we also included the variable hypothesized as moderator (self-efficacy for coping with stress).

To explore the mediation hypothesis, we performed mediation analyses using the 4^th model of the PROCESS for SPSS macro, with a confidence interval of 95% and 10,000 bootstraps. The “emotional demand” variable at the first evaluation time was introduced as the independent variable. The “emotional dissonance” variable at the second evaluation time was included as the mediator. The “emotional exhaustion” variable at the third evaluation period was introduced as the dependent variable.

Furthermore, we assessed our mediation and moderation hypotheses by performing a moderated mediation analysis using the 14th model of the PROCESS for SPSS macro. The analysis was conducted with a 95% confidence interval and 10,000 bootstraps. The same independent, mediating, and dependent variables as in the previous model were introduced in this analysis. In addition, we introduced the “self-efficacy to cope with stress” variable at the second evaluation time as the moderator. Z-scores of the variables were used for the mediation and moderating analyses. Additionally, covariates such as relationship status, interaction, age, sex, and years of experience in teaching were included as covariates to control for their effects in the mediation and moderated mediation models.

Results

Preliminary Analysis

No significant differences between study variable levels over time were revealed by the repeated measures analyses. Pearson correlation analyses confirmed the expected relationships between variables (Table 1). Similarly, the linear regression analysis (Table 2) indicated that the first model accounted for 11% of the variance and identified emotional demands as a significant predictor of emotional exhaustion. In the second model, the two variables accounted for a collective 19% of the variance, with both emotional demands and emotional dissonance significantly contributing to the regression equation as direct predictors of emotional exhaustion. Finally, when adding the potential moderating variable, self-efficacy to cope with stress, the third model explained 27% of the variance, with only emotional dissonance and self-efficacy contributing significantly to the equation, while emotional demands showed no significant effect on emotional exhaustion.

Mediation and Moderated Mediation Analyses

The mediation analysis, R² = .224, F (5, 101) = 5.816, p < .001, demonstrated the mediating role of emotional dissonance expected in H1 (standardized indirect effect of emotional demands on emotional exhaustion through emotional dissonance = .115, [.033, .239]). Moreover, as depicted in Figure 2, the moderated mediation analysis, R² = .346, F (7, 99) = 7.467, p < .001, supported the moderated mediation hypothesis raised in H3.

Thus, the results endorsed both (a) H1, indicating that emotional dissonance mediates the relationship between emotional demands and emotional exhaustion, and (b) H2, showing that self-efficacy, Unconditional Interaction Test (UIT): ΔR² = .Ù 29, F(1, 99) = 4.391, p = 0.039, act as moderator in the relationship between emotional dissonance and emotional exhaustion. The results confirm the moderated mediation of H3, moderated mediation index (MMI) = -.064 [-.165, -.004]. The moderating effect of self-efficacy is evident in Table 3 and Figure 3, showing that self-efficacy buffered the negative effect of emotional dissonance on emotional exhaustion.

Covariates (sex, age, years teaching) had no significant effect on emotional dissonance or emotional exhaustion in either the mediational analysis or the moderated mediation analysis.

Discussion

Many studies have linked burnout and emotional exhaustion with compromised levels of job performance among workers in different contexts (Grau et al., 2001; Maslach & Pines, 1977; Pugliesi, 1999; Rudow, 1999). Given the importance of education in societal resilience, it is crucial to prevent emotional exhaustion in teachers and reduce its negative impact – for educators’ quality of both life and teaching. Therefore, this research focused on some of the variables involved in the emotional exhaustion of teachers and investigated the interactions established between them when explaining exhaustion. At the theoretical level, this study contributes to the existing literature by proposing that third variables play mediating and moderating roles in the link between emotional demands and emotional exhaustion. The practical implications of this study are relevant for interventions designed to mitigate teachers’ emotional exhaustion.

Our linear regression analysis revealed a significant influence of all the study variables on emotional exhaustion. However, despite the direct predictive strength of emotional demands being supported in the first and second models, this variable lost its direct predictive effect when included with emotional dissonance (the hypothesized mediating variable) and self-efficacy to cope with stress (the hypothesized moderating variable) in Model 3.

Overall, the results suggest that the direct effect of emotional demands on emotional exhaustion is blurred when additional variables are included. These findings align with the hypothesis that other variables may serve as mediators and/or moderators of the relationship between emotional demands and emotional exhaustion, as further analyses have revealed. Moreover, it also aligns with the significant but weak effect of an increase of emotional demands on the increase of emotional exhaustion in the transversal and longitudinal data found by Vammen et al. (2019), which could also be attributed to the moderation and/or mediation of additional variables affecting the outcome.

It is worth noting that the significant relationship found in the different regression models conducted for the different variables was in the expected direction. That is, higher levels of emotional demands and emotional dissonance were associated with higher levels of emotional exhaustion, as demonstrated in previous studies (Andela & Truchot, 2017; Andela et al., 2016; Bakker & Heuven, 2006; Muñoz et al., 2009; Peng et al., 2010; Rivera-Torres et al., 2013). Conversely, higher levels of self-efficacy to cope with stress were related to lower levels of emotional exhaustion, as also previously shown (Cuadrado et al., 2021; Dicke et al., 2014; Fernández Arata, 2008; Grau et al., 2001).

Together, our mediation and moderated mediation analyses supported our first hypothesis, corroborating that emotional dissonance acted as a mediator in the relationship established between emotional demands and emotional exhaustion. The relationship established between emotional demands and emotional exhaustion suffered by teachers was found to be indirect, through its impact on emotional dissonance, which subsequently directly affected emotional exhaustion. Therefore, when teachers face greater emotional demands, their levels of emotional dissonance and subsequent emotional exhaustion increase.

This finding is supported by the research carried out by Lewig and Dollard (2003) in other work settings. While the relation between emotional demands and emotional dissonance to emotional exhaustion has been studied in the teaching context (Andela & Truchot, 2017; Baeriswyl et al., 2021; Guglielmi & Tatrow, 1998; Simbula et al., 2019; Skaalvik & Skaalvik, 2017; Tuxford & Bradley, 2015), the mediating role of emotional dissonance between the two has not been studied in a professional context, as far as we know.

From a theoretical standpoint, these discoveries are novel: among teachers, emotional demands cause an increase in emotional exhaustion, specifically through their impact on emotional dissonance. Teachers frequently experience high emotional demands (Casassús, 2007; Cejudo & López-Delgado, 2017). These demands, inherent to teaching, lead to the emergence of negative emotions that teachers are expected to mask. Consequently, teachers experience higher emotional dissonance, which in turn increases their levels of emotional exhaustion. Therefore, in practical settings, socio-educational interventions aimed at reducing emotional exhaustion levels among teaching professionals should account for emotional demands and emotional dissonance as variables of interest.

Our second hypothesis (H2) was also confirmed: self-efficacy to cope with stress did moderate the relationship between emotional dissonance and emotional exhaustion. This finding indicates that elevated levels of self-efficacy to cope with stress mitigated the adverse impact of emotional dissonance on emotional exhaustion. These findings suggest that, with low levels of emotional dissonance, levels of emotional exhaustion are also low, regardless of the individual’s level of self-efficacy. However, as emotional dissonance levels rose, so did levels of emotional exhaustion, but this time high levels of efficacy acted as a protective factor. Thus, teachers who jointly exhibit moderate to high levels of emotional dissonance and high levels of self-efficacy to cope with stress report lower levels of emotional exhaustion than their counterparts who display the same levels of emotional dissonance but possess moderate to low levels of self-efficacy.

Moreover, high levels of self-efficacy led to low levels of emotional exhaustion irrespective of the levels of emotional dissonance. Although no previous literature explored this phenomenon, our finding aligns with prior studies showing how self-efficacy often moderates the relationships between different variables and emotional exhaustion (Dicke et al., 2014; Grau et al., 2001). Specifically, and similar to our study, Tuxford and Bradley (2015) demonstrate that teaching self-efficacy mitigates the detrimental impact of deep acting on emotional exhaustion.

At a practical level, interventions aimed at preventing or reducing emotional exhaustion in teachers should focus on building self-efficacy to cope with stress. Our study has shown that this variable is highly relevant to reducing emotional exhaustion of teachers, even in those with high levels of emotional dissonance resulting from the emotional demands of their work. In fact, the current study reveals that high levels of self-efficacy to cope with stress are associated with low emotional exhaustion in teachers, regardless of their emotional dissonance levels. It is therefore important to prioritize interventions targeting self-efficacy as a means of mitigating emotional exhaustion among this population. Interventions designed to enhance the self-efficacy of teachers in managing stress seem to be especially important in work settings with moderate to high levels of emotional dissonance. For these teachers, high self-efficacy levels are indispensable for mitigating emotional exhaustion levels.

Finally, our third hypothesis (H3) indicated that self-efficacy in coping with stress moderated the mediation relationship between emotional demands, emotional dissonance, and exhaustion. This finding is of theoretical importance due to its novelty. Although previous research has explored the relationship between emotional demands, emotional dissonance, and emotional exhaustion in the context of teaching work, the mediating role of emotional dissonance and the interaction effect between emotional demands, emotional dissonance, and self-efficacy have yet to be studied. Our contribution offers a potential rationale for the significant yet moderate impact of heightened emotional demands on emotional exhaustion reported by Vammen et al. (2019): emotional dissonance acting as a mediator between emotional demands and emotional exhaustion could potentially account for the limited strength of this association.

Another variable that influences emotional exhaustion and may be able to blur the relationship established between emotional demands and emotional exhaustion is self-efficacy to cope with stress, which moderates the mediated relationship between emotional demands, emotional dissonance, and emotional exhaustion.

Our findings have several practical applications. Specifically, the results suggest that emotional demands and emotional dissonance pose a risk to teachers’ well-being, while self-efficacy to cope with stress can mitigate burnout by minimizing the adverse effects of emotional dissonance on emotional exhaustion. In the context of teaching jobs, emotional demands, emotional dissonance, and self-efficacy to cope with stress are crucial factors to consider for psychosocial and socio-educational interventions aimed at mitigating burnout among teachers. In particular, our findings indicate the special relevance of self-efficacy to cope with stress for interventions targeted to preventing emotional exhaustion. In this sense, the results suggested that self-efficacy can likely help reduce the negative effects exerted on emotional exhaustion by the emotional dissonance resulting from the emotional demands usually experienced by teachers at school.

In work settings where it is difficult to reduce the emotional demands and dissonance that teachers experience, enhancing their ability to manage stress may be a more viable solution. Once self-efficacy is increased through intervention, even teachers with high levels of emotional dissonance may find themselves protected from its negative effects. As suggested by Bandura (1977) and Wood and Bandura (1989), self-efficacy can be increased through mastering experiences, observing others’ experiences, being verbally persuaded, and physiological arousal. Therefore, interventions aimed at increasing self-efficacy to deal with stress and prevent or reduce teachers’ emotional exhaustion should consider these four sources of self-efficacy. First, to enhance intervention programs for teachers, one potential action could be to gather concrete data on educators’ ability to effectively cope with workplace stress. This data collection must be used to provide them with actual information about the effective they are in coping with workplace stress, enabling them to increase doing so, thereby increasing their conscientiousness about their mastery experiences. Second, teachers should be provided with tips on managing emotions in their teaching profession, which could positively affect perception of their ability to handle stressful situations. This in turn could protect teachers from emotional exhaustion. Third, teachers need time to practice managing the emotional demands and stress inherent to their jobs. This would enable them to experience mastery and improve their ability to cope with stress. And fourth, teachers should be exposed to experienced peers to which they can identify and who themselves effectively manage job-related stress. Mentorships can serve as a significant vicarious experience, and thus can help strengthen self-efficacy to cope with stress and decrease emotional exhaustion.

Limitations and Future Research

We identified several limitations. For instance, our conclusions are based on a relatively small sample size, especially considering the study’s longitudinal design with three time-point evaluations and a high experimental death rate. Additionally, although the longitudinal data is a strength, the limited number of evaluations precluded testing for long-term interaction effects. The small number of participants who participated in the three-point data collection prevented us from continuing data collection over time, therefore hindering our ability to test the model’s stability. Future research should replicate this study with more than three evaluation points to determine the model’s stability over time. Finally, while the moderating role of self-efficacy has been extensively discussed in the literature, there is still need for further investigation.

Conclusions

The present study examined the interaction between different variables of interest that contribute to emotional exhaustion in teachers during their professional activities. The study indicates that increased emotional demands on teachers have an indirect effect on higher levels of emotional exhaustion through the impact of emotional demands on emotional dissonance, which acts as a mediator in the relationship between emotional demands and emotional exhaustion. Additionally, the study endorses the moderating role of self-efficacy in coping with stress between the variables of emotional dissonance and emotional exhaustion. As emotional dissonance levels increased among teachers, emotional exhaustion also increased.

At the same time, high levels of self-efficacy to cope with stress reduced the impact of emotional dissonance on emotional exhaustion. This has led us to infer that self-efficacy to cope with stress acts as a protective factor against the negative effects of emotional dissonance on emotional exhaustion. Emotional demands, emotional dissonance, and self-efficacy to cope with stress appear to be relevant variables to consider in interventions aimed at reducing levels of emotional exhaustion in teachers, which are known to be generally high. Therefore, interventions that focus on both increasing teachers’ self-efficacy to cope with stress and decreasing emotional dissonance should be promoted. Specifically, self-efficacy to cope with stress is shown as a significant variable to consider in interventions aimed at preventing and reducing emotional exhaustion among teachers, particularly in those whose emotional dissonance levels are moderate or high.