An empirical, quantitative, ex post-facto, cross-sectional survey design was carried out (Calderón Saldaña and Alzamora de los Godos, 2018).

In this study, 9,058 teachers residing in different Argentinian provinces participated. They worked in public or private education institutions, or both, in one of the four educational levels, in special or permanent education of youngsters and adults. The selection of participants was conducted through a non-randomized procedure of volunteer recruitment. Demographic and working characteristics of the sample are exposed in Table 1. The registered teachers completed all the required tests. The test basis was free of missing data.

In order to collect information about the demographic and working characteristics of the participants, closed questions were used, which revealed information regarding gender, age, educational level, work seniority and institutional management, among other aspects.

In order to analyze the situations that teachers perceive as stressful, a Scale of Teachers’ Stressors in Times of Pandemic (Oros et al., 2020) was used, which consists of 21 items, with five Likert-type answer options (Not stressful = 1, A little stressful = 2, Somewhat stressful = 3, Quite stressful = 4, Very stressful = 5). The items are factorially grouped into five factors: work environment and work overload (e.g., Having little time to do all the tasks involved in remote work), use of new technologies (e.g., Learning how to use and master new technologies), uncertainty about the duration and consequences of the pandemic for the teacher and the students (e.g., Feeling uncertain toward the future, not knowing when the pandemic will end), the organizational aspect of the educational institution (e.g., Feeling that superiors do not understand the difficulties of working under these conditions), and relationship with the students’ environment, conflict and role ambiguity (e.g., Receiving multiple and simultaneous inquiries from students and/or parents). The internal consistency values reported by the authors ranged between ω = 0.78 and 0.85 for the factors, being of ω = 0.93 for the full scale. In this study, the values obtained were: alpha = 0.85 for Work environment and work overload factor, alpha = 0.80 for Use of new technologies factor, alpha = 0.83 for Uncertainty about the duration and the consequences of the pandemic for the teacher and the students factor, alpha = 0.81 for Organizational aspect of the educational institution factor, and alpha = 0.77 for Relationship with the students’ environment, conflict and role ambiguity factor.

In order to know the symptoms of stress, the Scale of physical psychoemotional indicators of stress (Oros and Neifert, 2006) was administered. This scale consists of 22 items grouped in three dimensions: Cognitive deficit (e.g., “I find it difficult to focus,” “I forget things easily”) (alpha = 0.79), Nervousness (e.g., “I get nervous easily,” “I feel I worry excessively about everything” (alpha = 0.74) and physical symptoms (“I have insomnia or difficulty in falling asleep,” “I have neck and back pain” (alpha = 0.62). The items were answered in a Likert scale with five answer options (Never, Hardly ever, Sometimes, Often, Always). The internal consistency values for this study sample were: alpha = 0.84 for Cognitive deficit, alpha = 0.86 for Nervousness, and alpha = 0.74 for Physical symptoms.

Finally, in order to evaluate the burnout, a Spanish adaptation by Seisdedos (1997) of the Maslach Burnout Inventory (MBI), by Maslach and Jackson (1986) was used. This instrument consists of 22 items assessed with a Likert-type scale. Individuals indicate, through a range of seven adjectives, ranging from “Never” (0) to “Every day” (6), how often they experience each of the situations described. The factorization of the 22 items shows three orthogonal factors, which are called: (a) Emotional exhaustion, (b) Depersonalization, and (c) Personal fulfillment. The study of the MBI internal consistency in its original version in English showed, through the Cronbach alpha index, good values for each of the three subscales (0.90 for Emotional exhaustion, 0.79 for Depersonalization, and 0.71 for Personal fulfillment). In our country and region, this self-administered inventory has been used in various studies, showing pyschometric properties between moderately acceptable and satisfactory (e.g., Marucco et al., 2009; Malander, 2019). In this sample, Cronbach alpha values were very good for Emotional exhaustion (0.89) and Personal fulfillment (0.85), and fairly acceptable for Depersonalization (0.64).

Due to the special conditions of social isolation in which this research was developed, the data collection was made through an online form. Volunteer recruitment was carried out through social media, electronic mail and digital messaging services, in some cases with the support of institutional and jurisdictional education authorities. The activity demanded approximately 15 min. The data collection started 46 days after the preventive social isolation was decreed in our country, in a period between April 27 and September 15, 2020. Teachers were provided with an email address for inquiries and an optional section at the end of the form to express opinions and comments.

The teachers’ answers were quantified and statistically processed. No missing data were recorded either at an item level or at a scale level.

To identify the situations valued as threatening, and the level of perceived stress, basic statistical analyses were conducted (calculus of means and standard deviations). In order to make descriptive comparisons between the different dimensions of the Stressor Scale, these means were considered by the number of items in each dimension.

The percentages of adherence to each answer section were calculated so as to estimate the general level of stress teachers perceived, taking into consideration the options of the Likert scale that were used in the evaluation (Not stressful = 1, A little stressful = 2, Somewhat stressful = 3, Quite stressful = 4, Very stressful = 5). In that way, scores weighted between 1 and 2 were considered indicative of Low level of stress, scores between 2.01 and 3 as indicative of moderately low level of stress, scores between 3.01 and 4 as moderately high, and scores over 4.01 as a high level of stress perception.

The assumptions of homoscedasticity and normality of the variables (univariate and multivariate) were also tested to determine the most adequate statistical test to study the hypotheses. For this purpose, the Kolmogorov-Smirnov, Box, Bartlett and Levene tests were used, from which the non-compliance with the mentioned assumptions was confirmed.

Therefore, with the purpose of analyzing possible differences among the psychophysical symptoms according to the level of stress, a non-parametric Kruskal Wallis H analysis was performed. Previously, the scores for each stressor were categorized in four groups (low stress, moderately low stress, moderately high stress and high stress) using the same criteria as with the general value. The Kruskal Wallis H test was also used to investigate the differences in the three manifestations of burnout based on perceived stress on the one hand, and psychophysical symptoms on the other. The psychophysical symptoms variable was categorized for this analysis using two cut points (< 33.33 = low symptomatology, 33.33–66.66 = moderate symptomatology, > 66.66 = high symptomatology).

The effect size for the H tests was estimated with the squared epsilon statistical (E²_R), considering for its interpretation the Cohen’s criteria for the analogous test of partial eta²: 0.01 (small), 0.06 (moderate) and 0.14 (large). In all cases, post hoc contrasts were conducted through the non-parametric Mann-Whitney U-test.

The actions carried out in this work complied with the international ethical recommendations for research with human beings (i.e., Asamblea Médica Mundial, 2013; American Psychological Association, 2017). The teachers participated anonymously and voluntarily, stating their consent before answering the instructions on the form. No incentives of any kind were provided in exchange for participation. The information collected was treated confidentially, and was not accessed by persons outside the investigation.

According to the analysis of weighted means, of the five major stressors that were assessed, the preeminence of Uncertainty about the duration and the consequences of the pandemic is highlighted, followed in order of importance by Work environment and work overload, Relationship with the students’ environment, conflict and role ambiguity, Organizational aspect of the educational institution, and Use of new technologies factors (see Table 2).

On the other hand, statistical analyses revealed that 62.1% of teachers in the sample display general levels of stress between moderately high and high.

The analysis indicated significant differences in the psychophysical symptomatology according to the levels of perceived stress. The size of these differences was between moderate and large, the largest being the cognitive deficit. Post hoc contrasts showed significant differences among all the comparison groups, for all the variables included in the analysis (p < 0.001). The frequency with which physical symptoms, nervousness, and cognitive deficit are experienced increases significantly as the perception of stress increases (see Table 3).

The results revealed statistically significant differences in the three dimensions of Burnout depending on the level of stress. The size of the effect was larger for the Emotional exhaustion dimension (see Table 4). The post hoc contrasts showed significant differences among all the comparison groups for the dimensions of Emotional exhaustion and Depersonalization, whereas for Personal fulfillment a few pairs of non-significant average ranges were observed (see Table 5). In general, it is observed that as the level of stress increases, Exhaustion and Depersonalization increase, and Personal fulfillment decreases. An exception to this tendency was seen in Uncertainty about the consequences of the pandemic factor; in this case, both high and low levels of stress were associated with lower Personal fulfillment.

Statistically significant differences were also observed in the three dimensions of Burnout according to the frequency of the psychophysical symptoms experienced. Again, the effect size was larger for Emotional exhaustion (see Table 6). The post hoc contrasts showed significant differences among all the comparison groups, for all the variables included in the analysis (p < 0.001). The higher the score in Cognitive deficit, Physical symptoms and Nervousness, the higher the score in Emotional Exhaustion and Depersonalization, but the lower the score in Personal fulfillment.

Of the five stressors assessed, uncertainty about the consequences of the pandemic for the teacher and the student stood out for its intensity (e.g., not knowing how the socioeconomic situation will continue, knowing there are students who do not have the technological resources to work remotely, being uncertain about the future, not knowing when the pandemic will end, and not knowing if students are understanding the contents). Preeminence of stressors associated with the working environment and work overload was observed (e.g., overlapping of work with household tasks, work schedule that is disorderly, unpredictable or different from the usual schedule, and lack of time to perform the tasks involved in remote work), as well as organizational aspects of the educational institution (e.g., receiving a greater number of requirements and demands from superiors), and the relationship with the students’ environment (e.g., receiving multiple and simultaneous inquiries from students and/or parents). As regards stress, it was noticed that 62.1% of educators presented high and moderately high levels of stress.

Similarly, studies conducted with Latin American teachers reported that they presented high levels of stress, anguish and anxiety during the suspension of face-to-face classes due to the global COVID-19 pandemic (Becerra Hernández, 2020; Casimiro Urcos et al., 2020). In Europe, under the same circumstances, Klapproth et al. (2020) found that German teachers experienced moderately high levels of stress on average. Likewise, they found that most educators pointed to the lack of technological equipment, internet connectivity, excessive workload and students’ demotivation as internal and external obstacles that made distance educational work difficult.

MacIntyre et al. (2020) examined work stressors that teachers who taught foreign languages in different educational levels and countries were exposed during school closings. These authors found that teachers identified work overload and concern for their family’s health as the most stressful. This was followed by loss of control over work, overlapping of work with household tasks, loss of control over personal decisions, irregular schedules and concern about finances. Coinciding with this study, they also highlighted that some educators mentioned not knowing when the pandemic would end as a source of stress.

In Argentina, between the months of April and June of 2020, different syndicates conducted studies in order to learn about aspects associated with the health and working conditions of educators of all educational levels and modalities, in the context of isolation and mandatory social distancing. These organizations reported that most teachers pointed out their work-related activities had increased and demanded more time. Consequently, they had to dedicate more hours to their daily work. A high percentage of educators reported not having adequate technological resources nor a comfortable isolated place to work at home. Moreover, combining work-related activities, household task and caring for their children, elderly or other family members was largely problematic for teachers. They also reported receiving communications outside working hours and contradictory instructions from educational authorities. Likewise, they revealed difficulties in supporting students’ schooling, mainly because many students did not have the necessary equipment and connectivity to work from home. Lastly, a significant number of educators expressed feeling concerned, anguished and suffering from stress (see Asociación del magisterio de Santa Fe, 2020; Confederación de Trabajadores de la Educación de la República Argentina, 2020; Sindicato Argentino de Docentes Particulares, 2020; Sindicato Unificado de Trabajadores de la Educación de Buenos Aires, 2020).

Results indicated that as the perception of stress increases, so do cognitive deficit, nervousness and various physical symptoms. These findings confirm the initial hypothesis and are consistent with the preliminary report by Arias et al. (2020). The authors observed that after 60 days of the onset of the quarantine in Argentina, teachers reported a significant increase in their workload and that this was associated to physical symptoms, especially lower back pain and discomfort in their arms and wrists. Similarly, Nieva’s (2020) study, conducted in Argentina during the context of social isolation due to COVID-19, registered a positive and significant correlation among various occupational stressors, somatization and other psychological symptoms in teachers. Likewise, 62% of Argentine teachers consulted by Casali and Torres (2021) expressed having worsened some psychophysical signs linked to stress in the context of the pandemic.

Similar results have also been documented before the onset of this sanitary contingency. For instance, Oramas Viera et al. (2007) assessed a sample of Venezuelan teachers, observing that the perception of stressors was positively and statistically significantly related to different functional symptoms (cognitive, affective, conative and psychosomatic). Comparably, Harmsen et al. (2019) found that the perception of stressors in teachers in the Netherlands kept a positive correlation with various stress responses, such as loss of pleasure, poorer sleep quality, fatigue, etc.

The general stress model proposed by Lazarus and Folkman (1986) and later taken up by Peiró and Salvador (1993) to explain work stress provides a solid frame for the interpretation of these results. According to their postulates, the stimuli that are perceived as threatening, in this case coming from the occupational context of teaching, produce a series of changes in the individual at a social, emotional and physiological level. These changes, with immediate and long-term effects, may produce suffering and deterioration of health and well-being (Ramos et al., 1997). Lazarus (2000) states that stress processes affect health in two ways. In the first place, they alter the neurochemistry of the body, that is, chemical and hormonal defenses that produce changes in sympathetic activity, secretion of catecholamines, etc., are activated as a reaction to stress (Krantz et al., 1985). Secondly, stress may favor the adoption of dysfunctional behaviors and coping styles. People exposed to situations perceived as stressful may attempt to obtain relief through the implementation of inadequate strategies, such as the consumption of alcohol, tobacco and other harmful substances, and excess or lack of food, which would derive in several psychophysical ailments.

On the other hand, some specific symptoms, such as those associated with the lower back, may be clearly explained by the large number of hours that teachers had to work sitting down in their own homes, possibly in an incorrect posture and/or using ergonomically inadequate furniture. Similarly, the excess of activities, the lack of time and the impossibility of leaving their residencies (due to the isolation dispositions) may have significantly increased sedentary lifestyles, also generating the appearance or exacerbation of these symptoms and others associated to the quality of sleep, to eating habits or to mental state (Bane et al., 2021).

Beyond the particular impact the stated situations may have had, it was probably the combination of all that explained the manifestation of various psychophysical symptoms presented when facing the stress involved in teaching in such particular contextual conditions.

The results showed that as stress increases, so do Emotional exhaustion and Depersonalization, with Emotional exhaustion having a greater size effect. Regarding personal fulfillment, it was observed that, in general, it decreases as the perception of stress increases. It was also noted that the more teachers experience Cognitive deficit, Physical symptoms and Nervousness, the lower their Personal fulfillment, and the higher their Emotional exhaustion and Depersonalization. These results confirm the second hypothesis presented in this study.

Studies conducted with teachers before the COVID-19 pandemic was declared have reported results similar to those obtained in this study. For instance, a study with Venezuelan teachers found that Burnout syndrome was significantly correlated with the perception of stress and its symptoms. In this study, Emotional exhaustion also showed a higher correlation with perception of stress and psychophysical symptoms (Oramas Viera et al., 2007). In a similar way, a research conducted with Portuguese teachers found that the perceived stress was inversely proportional to Personal fulfillment and directly proportional to Emotional exhaustion and Depersonalization (Teles et al., 2020).

As stated before, burnout syndrome is a prolonged response to chronic stressors at work. It appears when the coping strategies used to handle work stressors are not functional. The individual, unable to mitigate or eliminate the source of stress, generates a series of physiological, cognitive, affective, attitudinal and behavioral responses that work as mediating variables between perceived stress and its consequences (Gil-Monte, 2005; Maslach, 2009). In this context, it is possible to think that teachers who experienced intense levels of work stress under the non-face-to-face teaching modality may have used ineffective coping strategies. In response, they possibly saw their emotional resources depleted, developed negative attitudes toward the students and experienced feelings of low competence and success at work.

It was surprising that, in reference to Uncertainty about the consequences of the pandemic for the teacher and the student dimension, both high and low levels of stress were associated with lower Personal fulfillment. This finding is partially dissenting with the starting hypothesis, and not easy to interpret, especially considering the lack of empirical background on the topic. It would be appropriate to deepen on this topic in future research.