This was an exploratory cross-sectional study with 317 university professors who use active methodologies in different regions of Brazil.

The survey participants showed high teaching self-efficacy for the use of active methodologies, as well as achievement/expressiveness at work (Table 1). They experienced moderate levels of professional achievement, positive and negative affects in the work context, with low levels of depersonalization and a significant rate of emotional exhaustion. It is noteworthy that no missing items were recorded in the database.

The comparative analyses of the means using MANOVA did not identify significant differences depending on the variables age group, length of professional experience, affiliation, region and income in the instrument factors. The variables education, training, number of courses and number of classes were not included due to the lack of balanced distribution. Despite the high teaching self-efficacy for the use of active methodologies in the participants of this study, many of them could be considered affected by the burnout syndrome, since they presented high levels of emotional exhaustion (43.8%) and depersonalization (67.2%), associated with low professional achievement (44.8%) in the MBI.

Along this line, the study developed by Bressa et al. (2021) identified that medical courses’ teachers who were more committed to the use of active learning methodologies and formative assessment, felt more prepared to carry out teaching activities, enhancing their beliefs about the ability to manage daily life in the classroom and mobilize students to carry out their tasks. Furthermore, teachers with a high feeling of self-efficacy showed greater openness to new ideas, as well as showing great enthusiasm for teaching and being more committed to their profession (Rocha, 2009). However, some authors believe that the use of active methodologies requires teachers with skills, training and an institutional infrastructure that is not always offered by the highly precarious Brazilian higher education set up (Abonizio, 2012; Ferreira, 2014; Souza et al., 2020).

The stressors associated with insufficient working conditions already mentioned could justify the average values found in these results, especially with regard to positive affects, negative affects and achievement/expressiveness, as in Traldi and Demo (2012) who also used the Wellbeing at work in their studies.

Given those results, it becomes important to investigate the possible impacts of frequent exposure to the different stressors that Brazilian teachers are subjected to; among the most current stressors is the challenge of inserting active methodologies into their pedagogical practice in the face of a traditionalist curricular and institutional context. In this connection we should emphasize that educational reforms and the implementation of guidelines require persistence, continuous innovation and deep cognitive involvement of the teachers with their tasks (Ferreira & Azzi, 2011; Souza et al., 2020).

The rates of emotional exhaustion and depersonalization dimensions which were considered high were actually twice as high as those found by Bernardini (2017). In contrast, the professional achievement results according to our study were much lower than those of the aforementioned author. Furthermore, all the values yielded in our investigation were much higher than those of Ferreira (2014), who identified a predominance of low professional achievement (M = 2.64) associated with significant emotional exhaustion (M = 2.45) in his study’s participants. Thus, these data reveal a worrying reality that requires greater understanding of Brazilian teachers who promote active learning and the corresponding interventions.

Pearson’s correlation analysis of the results obtained in EADOMA and the WBWS and MBI dimensions reported in Table 2, indicates that the correlations, which were mostly negative and with a weak magnitude, presented statistical significance, with the exception of the correlations between achievement and exhaustion and depersonalization and self-efficacy. The largest coefficients obtained were between depersonalization and exhaustion in a positive sense, as well as between exhaustion and negative affects. Self-efficacy correlated positively with positive affects, achievement/expressiveness and professional achievement and negatively with negative affects and emotional exhaustion.

These results, reported in Table 2, corroborate the literature in the sense that high teaching self-efficacy beliefs produce higher levels of satisfaction and well-being at work (Bandura, 1997; Bernardini, 2017; Bressa et al., 2021; Lorente et al., 2014; Silva et al., 2010), especially on the affective component of well-being (Traldi & Demo, 2012), which, in turn, is inversely proportional to the development of the burnout syndrome (Bernardini, 2017; Ferreira, 2014; Ferreira & Azzi, 2010, 2011; Paschoal & Tamayo, 2008; Silva Jr. et al., 2018).

In this connection, Ferreira (2014) also showed negative correlations of moderate magnitude between lack of professional achievement and total teaching self-efficacy, as well as between all dimensions corresponding to teaching self-efficacy and the total burnout score. Achievement is one of the key elements for promoting efficient learning conditions, considering that the less the teacher perceives himself to be professionally accomplished, the greater the chances of him/her failing to use teaching methodologies that motivate students to learn (Ferreira, 2014).

It is worth noting that although teachers had high teaching self-efficacy, they were not free from experiencing a high incidence of the burnout syndrome, which also occurred in the study by Bernardini (2017) and Ferreira (2014). Considering that there are several variables that can interfere with the constructs assessed, another objective of this investigation was to seek explanatory variables related to teaching self-efficacy, well-being at work and burnout. To this end, regression analyses were carried out for the construct (Table 3).

In this connection, when reviewing teaching self-efficacy for the use of active methodologies as a dependent variable in linear regression analyses, the data indicated that only the variables education and use of active methodology predict TSE. In general, education negatively predicts TSE, implying that the higher level of education, the lower the TSE levels tend to be. The use of active methodologies predicts positively, as the greater the frequency of their use, the higher the levels of perceived teaching self-efficacy, explaining 12% of the variance.

To verify the path analysis regarding the dimensions of the three instruments used (EADOMA, WBWS and MBI), a direct extension of multiple regression models was carried out, adopting path analysis to investigate the results in an integrated manner with a better visual representation of the data (Figure 1).

As illustrated in Figure 1, teaching self-efficacy for the use of active methodologies was explained by Personal Achievement/Expressiveness and Professional Achievement, which in turn was explained by positive affects. Negative affects explained emotional exhaustion and depersonalization, corroborating the literature related to self-efficacy from a social cognitive perspective.

The graphical representation in the path analysis allowed us to identify that personal and professional achievement are predictors of teaching self-efficacy for the use of active methodologies. Positive affects are predictors of professional achievement and negative affects are predictors of emotional exhaustion and depersonalization. Expanding the discussions about predictive variables for well-being at work, Traldi and Demo (2012) identified that organizational commitment from an affective basis explained the well-being of the participants, since the more the teacher had affective commitment the higher the experience of positive affect in the work environment, as well as the lower the affective commitment to work, the greater the prevalence of negative affect.

For interventions on teacher health to be efficient, a better characterization of the variables influencing the construct is necessary. In this connection, recent studies reaffirm that teachers who worked in public institutions with a high workload (Carlotto & Câmara, 2017) had greater psychological exhaustion and risk of burnout (Baptista, Soares, et al., 2019; Carlotto & Câmara, 2017). The regression analyses conducted by Baptista, Soares, et al. (2019) demonstrated that the total burnout score was explained by the variables depression (β = 0.27; t = 2.96) and stressful events at work (β = 0.41; t = 4.01), predicting 32.6% of the tested model. Furthermore, stressors at work mediated the relationship between lower perception of work support and higher prevalence of burnout (Baptista, Soares, et al., 2019; Baptista & Cardoso, 2021).

Although the path analysis presented in figure 1 corroborates the literature with regard to the correlations between the constructs investigated here, it is worth highlighting the prevalence of identified burnout syndrome, even with high teaching self-efficacy. Such data seem to differ from what is proposed by social cognitive theory regarding self-efficacy beliefs (Bandura, 1986, 1997; Bernardini, 2017; Ferreira & Azzi, 2010, 2011; Rocha, 2009; Silva Jr. et al., 2018).

Given the scenario described above, it is suggested that new studies focus on explanatory variables associated with the high presence of these constructs pre and post Coronavirus Disease 2019 (COVID-19) pandemic in teachers in the specific framework of higher education in which teaching action is mediated by active learning methodologies (Bacich & Moran, 2018), considering that the literature has suggested a divergent relationship between teaching self-efficacy and burnout.

The limitations of our study involve the survey design adopted to search for validity based on external variables for the instrument. It would be interesting for new studies to further investigate the relationship between teacher self-efficacy and well-being at work and burnout in order to understand the low-magnitude correlations and investigate potential seasonal interferences on the constructs.

Regarding the contributions of this study, we have an unprecedented self-report scale that assessed teaching self-efficacy for the use of active methodologies, with a single-factor structure, associated with estimates of precision and evidence based on the relationship with favorable external variables. However, the need for studies with new samples is emphasized in order to possibly generate greater evidence for the EADOMA structure besides enabling the evaluation of the effect of intervention programs that promote teacher self-efficacy in this specific teaching and learning framework mediated by methodologies. Furthermore, future research with robust samples could also review the sources of self-efficacy of teachers who promote learning through active methodologies, employing the item characteristic analysis using the Item Response Theory and the response difficulty analysis.