Stress, produced by work factors, refers to the contents of work itself and the tasks carried out in the work settings (Nguyen & Ermasova, 2018). It displays, both in quantitative and qualitative terms, especially in work overload situations associated with systems to increase quality, a higher demand for training, and the increasing use of information and communication technology applied to work tasks. The literature examined for this study links work overload to physical and psychological health problems. Evidence of the relation between work overload and burnout has also been found in samples of different professions such as members of health care staff (Deodhar & Goswami, 2017), nursing staff (Grigorescu, Cazan, Grigorescu, & Rogozea 2018), pharmaceutical industry employees (Andrews & Kacmar, 2014), or occupational therapists (Paulsen et al., 2014), to name a few.

In the field of teaching and education, the meta-analysis carried out by Montgomery and Rupp (2005) with 65 studies on stress in teachers, revealed a strong association between the teachers' stressors, work overload, and burnout. Another more recent meta-analysis (García-Arroyo, Osca, & Peiró, 2019) also found a positive association between work overload and burnout in teachers from 36 countries. According to Marrau (2004), job insecurity, especially in temporary or part-time teachers, leads them to an overload of work, trying to become irreplaceable, which generates a high level of chronic stress that usually causes symptoms of burnout. There is much evidence of the positive relationship between overload and burnout (e.g., Abbas & Roger, 2013; De Beer, Pienaar, & Rothmann, 2016). However, some studies have not found this relation. For example, Chennoufi, Ellouze, Cherif, Mersni, and M'rad (2012), in a study with teachers of public secondary schools, found that 75.2% of the sample reported overload as a factor of occupational stress, but there was no significant association between work overload and the burnout syndrome.

Similarly, Gomes and Dos (2011) found that teachers with high work overload presented symptoms of depression, but they did not report significant associations between work overload and burnout, concluding that the symptoms of burnout could be due to the influence of other variables. Despite this indicated lack of consistency in the results, it seems that the most intuitive opinion positively relates work overload to emotional exhaustion. Therefore, we propose the following hypothesis:

Hypothesis 1: Work overload will be positively related to emotional exhaustion.

According to Lazarus and Folkman (1991), coping strategies are behavioral and cognitive efforts carried out to deal with, reduce, or tolerate the internal and external demands generated by stressful events. These authors point out two basic styles of coping strategies: those aimed at changing the situation, trying to modify the stressor, which are called active or direct; and those that are aimed at regulating the emotional response generated by the stressor, which are referred to as evasive, indirect, or palliative. Examples of the former are: trying to control the situation or the problem, seeking and assessing alternatives to solve it, taking cost-benefit into account, modifying the pressures, procedures or resources, or reducing the participation of the self. Examples of the latter would be avoidance, distancing, or positive comparisons (García-Arroyo & Osca, 2017).

Research shows that more direct coping strategies or strategies aimed at problem solving are positively related to psychological health (e.g., Austria et al., 2012; Mark & Smith, 2012). However, there is less consensus about the more indirect strategies or strategies aimed at managing emotions, which, in some cases, are linked to lower psychological well-being (Mark & Smith, 2012), but not in all situations (Chen, Ma, & Zhu, 2014; Lin, Probst, & Hsu, 2010). For example, a study with pediatric critical care staff notes that a high percentage of variation in burnout is predicted by frequent use of emotion-focused coping style and infrequent use of problem-center coping style (Rodríguez-Rey et al., 2019). Nevertheless, Dewe and Trenberth (2004) highlight that the lack of agreement is related to the complexity of the concept and the multiple ways of conceptualizing and measuring it. Besides, it is connected to the need for matching the coping strategies to the features of the stressor or to the situation in which it occurs.

In the university educational setting, Chan (2011) observed that active coping increases satisfaction and the feeling of personal accomplishment and reduces emotional exhaustion. On the other hand, Guerrero (2003) found that teachers who resort to passive or evasive coping strategies show higher levels of emotional exhaustion. Similarly, Mazon, Carlotto, and Câmara (2008) reported that greater use of coping strategies focusing on emotions led to higher levels of emotional exhaustion. Comparable results were reported by David and Quintao (2012), who observed that teachers who used strategies focused primarily on emotion experienced higher levels of burnout. Taking into account all these findings, we propose the following two hypotheses:

Hypothesis 2: Active coping will be negatively related to emotional exhaustion.

Hypothesis 3: Evasive coping will be positively related to emotional exhaustion.

In the Latin American background, we found some studies with samples of university teachers from Colombia, Mexico, and Peru. These studies mainly analyzed the prevalence of burnout (e.g., Ibañez, Bicenty, Thomas, & Martínez, 2011 ; Rojas & Grisales, 2011) and the coping styles most commonly used (Garay, Farfán, & Moysen, 2010; Valadez, Bravo, & Vaquero, 2011), which are usually the active-focused coping styles such as confrontation, planning or seeking social support.

Some studies have analyzed the moderating effect of coping, although the results are not conclusive. For example, some authors (Peters et al., 2016) in a longitudinal study with nurses found no significant moderating effects of active coping in the relationship between work demands and emotional exhaustion. However, other authors (Peltokorpi, 2018), also in longitudinal studies, did find that the evasive coping moderated the relationship between abusive supervision and subordinates’ emotional exhaustion.

Moreover, some studies with double interactions show that it is appropriate to combine both the active and the evasive strategies (García-Arroyo & Osca, 2017). In the same way, we analyzed the moderating role of coping strategies, expecting that coping would act especially in situations of high overload. Lastly, we addressed the study of the joint effect of the two coping strategies, as it is presumed that coping will be more efficacious when combining both strategies, active and evasive, simultaneously. Based on this, we propose the two following hypotheses:

Hypothesis 4: Coping will play a moderating role in the influence of overload on emotional exhaustion, and this will be particularly evident in situations of high overload.

Hypothesis 5: Coping will play a moderating role in the influence of overload on emotional exhaustion so that, in the face of high overload situations, the conjoint use of both coping strategies will be more efficacious.

The correlation analysis shows that work overload is positively related to emotional exhaustion, supporting hypothesis 1. Active coping is negatively related to emotional exhaustion supporting hypothesis 2. However, no significant relationship between evasive coping and emotional exhaustion was found, leading us to reject hypothesis 3. There is a positive and significant relationship between work overload and evasive coping, and also between the two coping styles with each other. The results further indicate that when teaching experience increases, work overload and emotional exhaustion decrease. Table 1 presents the means, standard deviations, and correlations between the variables.

Table 1 Table 1 Means, standard deviations and correlations

* p < 0.05, **p< 0.01, *** p< 0.001

The regression results to predict emotional exhaustion are presented in table 2.

The control variable teaching experience, F (1,190) = 13.54, p < 0.001, was significant explaining 6% of the variance of emotional exhaustion. Work overload (b = 0.17, p < 0.05), increased the explained variance to 9%. Both active coping (b = - 0.31, p < 0.001) and evasive coping (b = 0.15, p < 0.05) were significant, increasing the explained variance to 18%, F (4,187) = 10.81, p < 0.001. These results confirm that these three variables contribute in the prediction of emotional exhaustion. In step IV, the interaction between work overload and evasive coping was significant (b = -0.67, p < 0.10), F (6,185) = 7.70, p < 0.001, increasing the explained variance to 20%. These results support hypothesis 4. However, the triple interaction between work overload, active coping and evasive coping was non-significant, leading us to reject hypothesis 5.

Table 2 Table 2 Regression analysis to predict emotional exhaustion

N = 200; † < 0.10, *p < 0.05, **p < 0.01, ***p < 0.001

Figure 1 shows the moderating effect of evasive coping on work overload. The use of evasive coping reduces emotional exhaustion when work overload increases. Thus, in situations with low work overload, there are no significant differences in the level of exhaustion between people who use high evasive coping and those who use low evasive coping, but when the work overload is high, the teachers who use low evasive coping are more tired than teachers who use a high evasive coping.

Figure 1 Figure 1 Modulated relationship between work overload and evasive coping to predict burnout.

The aim of this study was twofold. First, we analyzed the relationship between the stress produced by work overload and emotional exhaustion, and second, we examined the role of coping styles in this relationship. To fulfill these general objectives, five hypotheses were tested about the relationship between work overload, active and evasive coping, and emotional exhaustion. As expected, work overload is positively and significantly related to emotional exhaustion, supporting the first hypothesis. These results are consistent with many of the studies reviewed in which university teachers with higher work overload also had higher levels of burnout (e.g., Abbas & Roger, 2013; Tripken, 2012).

Hypotheses 2 and 3 relate coping styles with emotional exhaustion. A significant negative relationship was found between active coping and emotional exhaustion; however, evasive coping was not significant. These results coincide with those by Nowakowska-Domagala, Jablkowska-Gorecka, Kostrzanowska-Jarmakowska, Mortoń and Stecz (2015) who found that task-oriented coping was negatively correlated with exhaustion and found no correlation between the symptoms of exhaustion and avoidance-oriented coping. They also agree with previous studies in relation to the beneficial effect of active coping (Chan, 2011; Palacio, Caballero, González, Gravini, & Contreras, 2012) so that facing the problem directly helps to reduce exhaustion. These results also indicate that the direct relationship between evasive coping and exhaustion is not clear, as previous study results had shown, although they do support the moderating role this coping style may have.

The results found in Ecuadorian teachers are similar to those found in teachers who are not Latin American. This allows us to state that in the sample studied, the effect of coping strategies on health, and specifically on burnout, coincides with the results obtained to date in most European or American contexts and countries. Nevertheless, these results differ from presented by Ramírez et al. (2018) with samples of health professionals in Ecuador. These authors found that the use of more passive coping was associated with a higher probability of presenting exhaustion. The occupation can explain this difference. According to some authors (García-Arroyo & Osca 2018), health professionals may experience very demanding situations that produce high exhaustion. However, they have more highly structured and defined procedures to perform their tasks than teachers and therefore, the effect of evasive coping on exhaustion is different in each profession.

Hypothesis 4 analyzed the moderating role of coping in the relationship between work overload and emotional exhaustion and proposed that this effect would be particularly evident in situations of high overload. Our results show that when work overload is high, teachers who use more evasive coping styles are less tired than those who use evasive coping infrequently. No evidence of moderating effect on the active coping style was found. From this, we can understand that active coping acts directly reduce exhaustion. For example, a teacher will reduce emotional exhaustion when using active strategies directly related to work overload, such as organizing schedules or distributing tasks to be performed. However, evasive styles, such as seeking distractions or social support, will not reduce exhaustion directly but will moderate the stressor's effect.

In this way, although evasive coping has been related to the decline of psychological health, it can be beneficial in stressful situations thanks to its moderating effect. Thus, for coping styles to be effective, they must adapt to the situation. To be more precise, while active coping seeks to change the stressful situation, evasive coping is effective when it is difficult to change the situation or the stressors. For example, Oren (2012) notes that active coping is useful for employees with high levels of autonomy and control, but for those with limited capacity of control, active coping may be counterproductive for health due to the extra effort involved.

Lastly, we found no evidence of the triple interaction between work overload, active coping, and evasive coping. Probably since, in our sample, overload and emotional exhaustion report low rates ​​as well as very high active coping scores, which are very effective in a direct mode and at the same time reduce the possibility of significant interaction effects. Thus, unfortunately, hypothesis 5 was not confirmed.