Materials and method

Participants

The study involved 166 teachers from 13 public schools within a regional educational superintendent area in the state of Minas Gerais, Brazil. Regarding personal data, the results showed that the majority of the sample were female (72.9 %), while the minority were male (27.1 %). The average age of the participants was 41 years, with a standard deviation of 10.67 years. Regarding marital status, most of them were married (49.4 %), followed by single individuals (31.9 %). The number of children varied, with the majority having between 1 and 2 children (47 %). Half of the participants identified themselves as the family breadwinner (50 %).

In terms of occupational data, most participants worked in only one school (55.4 %), predominantly in public schools (89.8 %). The average weekly workload was 35 hours. The participants taught various grades in elementary school, with the largest proportion teaching 6th, 7th, and 8th grades (51.2 % each), followed by 9th grade (45.8 %). The average duration of employment as a teacher was 14.22 years. Most participants had a degree in Human Sciences (62.7 %), followed by Biological Sciences (18.1 %) and Exact Sciences (16.3 %). Postgraduate education was common among the sample, with 78.9 % of participants having completed postgraduate studies.

Regarding health-related information, most participants (80.8 %) had seen a doctor within the past year. Approximately one-fifth of the participants (20.5 %) required sick leave in the first semester of 2023, while the majority (78.9 %) did not. Most participants (57.2 %) were not undergoing continuous health treatment. Around 44 % of participants reported some form of health complaint at the time of data collection, with 44 % of these complaints related to work as teachers. Finally, most participants (77.1 %) expressed a desire to receive training in social skills and/or materials on socio-emotional development. Complete sample data are detailed in Table 1 T2.

Table 1: Sociodemographic and occupational characterization: personal, occupational, and health-related data 

Results

Table 2 presents the Pearson correlation coefficients describing the relationships between burnout (ISB), social skills (IHS-2), coping (IEC), and socio-occupational variables. The ISB-global negatively correlated with the global IHS-2 (r = -.273, p < .01), indicating a weak negative correlation between burnout syndrome and the general social skills of teachers. Similarly, negative correlations were observed between ISB-global and IHS in each factor: factor 1 - assertive conversation (r = -.270, p< .01), factor 3 - expression of positive feelings (r = -.239, p < .01), and factor 5 - social skills (r= -.397, p < .01), meaning that the lower the assertive conversation, expression of positive feelings, and social skills, the higher the burnout score. Regarding coping (IEC), global IEC had a non-significant positive correlation with ISB-global (r = .079, p = .32), suggesting no correlation. Among the coping factors, only IEC factor 2 - distancing (r = .190, p = .02) and factor 6 - escape-avoidance (r = .389, p< .01) showed significant positive correlations with ISB-global, indicating that higher distancing and escape-avoidance correlate with higher ISB scores. Coping factors: factor 7 - problem-solving (r= -.215, p < .01) and factor 8 - positive reappraisal (r = -.158, p < .01) showed significant negative correlations with ISB-global, suggesting that lower use of problem-solving and positive reappraisal strategies correlates with higher ISB scores. For socio-occupational variables, teachers' age negatively correlated with ISB-global (r= -.241, p < .01), indicating that younger teachers are more likely to experience higher levels of burnout syndrome. The duration of teaching also showed a significant negative correlation with ISB-global (r = -.192, p < .01), indicating that teachers with less experience had higher burnout scores.

Table 2: Pearson Correlation Coefficient between burnout syndrome (ISB), social skills (IHS), coping (IEC), and some sociodemographic and occupational variables of teachers 

^Notes:r: Pearson Correlation Coefficient. Classification: very weak .00 to .19; weak correlation .20 to .39; moderate correlation .40 to .69; strong correlation .70 to .89; very strong correlation .90 to 1.0. *p < .25

Table 3 presents the sample sizes, means, standard deviations, test statistics, p-values, post-hoc comparisons, and effect sizes related to the ISB-global score and selected socio-occupational variables. Medium effects (d ≥ 0.5) were observed between the ISB-global score and marital status, family provider role, taking health-related leave in the first semester of 2023, continuous health treatment, and health complaints related to work. Specifically, teachers who were single, family breadwinners, took health leave in the first semester of 2023, reported health complaints related to their teaching work, or were undergoing continuous medical treatment tended to have higher overall burnout scores (ISB).

Table 3: Sample sizes, means, standard deviations, test statistics, p-value, post-hoc, and effect size (classification) between ISB-Global and some socio-occupational variables 

^Notes:n: sample size; M: mean; SD: Standard Deviation; df: degrees of freedom, t(df): value of the t-test statistic for independent samples; df_between: degrees of freedom between groups; df_within: degrees of freedom within groups, F(df_between ; df_within): value of the ANOVA statistic; Sig.: p-value; SSE: Sum of squares between groups; SST: Total sum of squares; MSD: mean squares within groups; r: effect size for the t-test for independent samples = sqrt(t²/(t²+df)) (^{Field, 2021}); ω: effect size for ANOVA = sqrt((SSE - df_between*MSD)/(SST + MSD)) (^{Field, 2021}); effect size classifications: (0;0.10) = Small (S), (0.10;0.50) = Medium (M), (0.50;1) = Large (L) (^{Field, 2021}). * For Sig. < .25

Given the lack of studies allowing for theoretical consolidation, multiple linear regression was performed using the stepwise selection method to define significant predictors of teacher burnout (ISB-global). The variables presented in Table 1 were selected based on the significance of the t-test for independent samples, ANOVA, correlation tests (Table 2), and avoiding the multicollinearity effect. Tables 2 and 3 showed that variables with p < .25 related to burnout syndrome (ISB-global) were: age, marital status, number of children, family breadwinner, teaching experience, need for sick leave in the first semester of 2023, continuous health treatment, current health complaints, whether the complaint is related to teaching, and social skills (IHS factor 5). The multiple linear regression results are presented in Table 4. The assumptions for linear regression analyses were analyzed and confirmed, i.e., K-S = 0.513 (p = .95), ensuring residual normality, and the Durbin-Watson statistic was 2.276, ensuring residual independence.

Table 4 presents the multiple linear regression results explaining teacher burnout. This analysis aimed to identify which studied variables most contributed to predicting the syndrome. The results indicated that social skills (HIS factor 5) were the most significant predictor (highest standardized beta value in absolute terms) compared to other variables. Second was health complaints related to work; then, teaching experience; next, being the family breadwinner; and finally, continuous health treatment. The regression model (R = 0.60, F (5, 122) = 13.779, p < .001) was given by: ISB-global = 2.034 + 0.209 x family breadwinner -0.015 x teaching experience + 0.195 x continuous health treatment + 0.395 x work-related health complaint - 0.281 x social skills. These results indicate that higher average scores on the ISB-global are associated with lower social skills, less teaching experience, ongoing health treatment, current work-related health complaints, and being the primary family breadwinner among teachers.

Table 4: Multiple linear regression analysis to explain teachers' burnout syndrome 

^Notes: B: unstandardized regression coefficient; β: standardized regression coefficient; t: t-statistic; p: p-value; DW: Durbin-Watson; K-S: Kolmogorov-Smirnov.

Discussion

This study found a negative correlation between burnout and social skills, partially aligning with the studies by ^{Almeida et al. (2020}) and ^{Buscatto et al., (2018}), which reported significant difficulties in interpersonal relationships in the presence of the syndrome. Another investigation indicated that individuals with burnout tended to social isolation and reported difficulties in socialization (^{Lucca, 2021}). Furthermore, research by ^{Bolsoni-Silva et al. (2015}) and ^{Del Prette and Del Prette (2017}; 2022) pointed out that burnout hinders the development of social skills.

Regarding the association between burnout and social skills, a study targeting medical residents found the same result as the present study, identifying a negative correlation between the variables (^{Pletti, 2021}). Similarly, a study by ^{Achkar et al. (2016}) showed that a varied repertoire of social skills was negatively associated with burnout. In this direction, research by ^{Schoeps et al. (2019}), with 340 Spanish teachers, found that the development of social skills reduced the occurrence of the syndrome. Among the correlated social skills, social poise (Factor 5) suggested the strongest association in this study group.

As for the relationship between burnout and coping, no significant correlation was found in this research. Nonetheless, the study by ^{David and Quintão (2012}) compared coping between primary and higher education teachers, noting that university professors employed a variety of coping strategies and had lower burnout rates compared to primary education teachers. This helps understand the insignificance found in this study, as the target group was composed of elementary school teachers, who also did not show a significant relationship between burnout and coping. Similarly, studies by ^{Ullrich et al. (2012}), with 469 teachers from German schools, highlighted the importance of coping in addressing work-related exhaustion. Additionally, ^{Hallum et al. (2013}) showed that emotion-focused strategies protect against burnout. Given these different findings, future research that includes teachers from various educational segments will be crucial to broaden the discussion.

The coping strategies most adopted by teachers in this study were withdrawal and escape-avoidance, which differed from the results of ^{Cabellos (2020}), whose coping primarily involved positive reappraisal and social support. ^{Fiorilli et al. (2016}) also found, in a study with 149 Italian teachers, that the lack of social support was directly associated with burnout complications. These differences in results may be due to cultural issues, curriculum structure, and school infrastructure, which can either aid or hinder teacher performance (^{Ullrich et al., 2012}). Lastly, a study by ^{Ramón (2015}) concluded that problem-focused and positive reappraisal strategies were associated with greater professional satisfaction.

Regarding the relationship between social skills and coping, this investigation did not identify a significant association. However, there was a positive correlation between the repertoire of social skills and the strategies of social support, problem-solving, and positive reappraisal. This means that the more extensive the repertoire of social skills, the more teachers used these resources. Partially, ^{Fiorilli et al.'s (2016}) study corroborated these findings, as social support was the strategy that most helped prevent burnout in the Italian educational context. Thus, ^{Jeter's (2013}) studies with American teachers considered socioemotional development relevant for reducing work-related exhaustion and for the correct adoption of coping.

Another empirical study relating the repertoire of social skills and coping was by ^{Bezerra et al. (2022}), whose results indicated that the acquisition of social skills enhanced the adoption of coping strategies. The research by ^{Soares et al. (2019}) described that interpersonal skills aided in problem-solving, suggesting a relationship between these skills and coping strategies.

As for the results related to socio-occupational variables, it was found that the average burnout score increased when the teacher was the primary family breadwinner, had more children, was younger, had less teaching experience, was single, reported health complaints, and was undergoing continuous medical treatment related to work. Finally, it was identified that the lower the teacher's social poise, the higher the burnout index. These variables, therefore, were predictors of teacher professional exhaustion in this study, highlighting the social poise deficit, but further investigations are suggested.

Regarding being the family breadwinner, the teacher showed concern about family support, and in trying to reconcile it with professional demands, psychological wear increased, as reported in a study with 310 teachers (^{Goebel & Carlotto, 2019}). Also, concerning the family context, the number of children was a complicating factor, possibly because parental responsibilities intensified as the number of children increased, explaining the fatigue resulting from multiple roles (Goebel & Carlotto, 2019).

With respect to the shorter duration of teaching being an indicator of burnout, it was assumed that classroom experience contributed to developing skills to manage stress (^{Pereira-Neto et al., 2019}). Additionally, studies by ^{Koga et al. (2015}) emphasized that teachers with less experience showed more insecurity despite their enthusiasm for teaching. Similarly, ^{Santos et al. (2023}) reported that the more experienced the teacher, the lower the risk of work-related exhaustion.

In relation to age, ^{Koga et al. (2015}) found that younger teachers had more difficulties dealing with the challenges of student relationships. Complementing this assertion, ^{Santos et al. (2023}) found that older teachers felt more secure and fulfilled, as their unrealistic expectations about the profession were lower. The findings of this investigation align with the results of these two studies. However, a study by ^{Baptista et al. (2019}) indicated that older professionals get sick more often, as prolonged exposure to work stressors culminates in burnout. Studies are recommended to address the contradictions of these results.

Among the participants, single teachers had higher burnout scores compared to their married counterparts, contrasting with the findings of ^{Palage et al. (2020}), who emphasized marriage as a potential source of emotional support during challenging periods. Another investigation considered married individuals more vulnerable to burnout due to the added responsibilities and concerns unique to the family context (^{Abacar et al., 2020}). Similarly, married individuals showed a greater propensity for professional exhaustion compared to singles, due to marital demands (^{Pimenta et al., 2021}). In contrast, a study by ^{Abacar (2021)} found no statistically significant relationship between marital status and burnout. Again, studies are recommended to address these contradictory results.

Teachers who reported health complaints and were undergoing continuous medical treatment at the time of data collection associated these issues with their teaching responsibilities, thereby displaying higher indicators of burnout. This finding may be attributed to the symptoms of the syndrome itself necessitating treatment (^{Lima & Dolabela, 2021}). Furthermore, psychological disorders resulting from professional exhaustion, such as depression, also require intervention. ^{Bolsoni-Silva et al. (2018}) supported this assertion, reporting a 23 % prevalence of depression among elementary school teachers and identifying significant positive correlations between depression and burnout.

Regarding the social poise deficit as the most significant predictor of burnout, scholars have pointed out that developing social skills is a protective resource against work-related illness (^{Mérida-Lopéz & Extremera, 2017}; ^{Schoeps et al., 2019}). Social poise includes behaviors such as feeling comfortable in a group of unfamiliar people, approaching or disagreeing with authorities, asking when an explanation is not understood, and greeting strangers. It is assumed that these behaviors are challenging because interaction in familiar environments, such as school, facilitates communication (^{Achkar et al., 2016}). However, not everyone speaks up during meetings, as apathy is one of the symptoms of burnout.

Conclusion

While this study yielded significant and consistent results aligning with existing literature, it is important to exercise caution when generalizing the findings. The teachers involved were exclusively from a single geographic region, and data collection was limited to the public education system. Furthermore, the study's reliance solely on self-report instruments presents another limitation; involving additional informants such as school administrators and students could have provided valuable insights.

Moreover, as a cross-sectional study, this research did not establish causal relationships between variables. Future experimental studies incorporating interventions are recommended to explore causality and examine specific categories of social skills deficits that contribute to the development of burnout. Replicating this study across different teacher populations is also advised to enhance comprehension and discourse on the subject matter.

The results of this study indicated that the deficit in social skills, especially the lack of social poise, could be a significant predictor of burnout in teachers. Therefore, social skills training could be considered, as 77.1% of the teachers in this study were willing to receive it. Another relevant consideration is that, in the face of burnout, teachers who used coping strategies showed less emotional exhaustion. It was also found that teachers who are family breadwinners, single, and have many children tend to experience work-related exhaustion. This conclusion can help support teaching by giving more consideration to the family context.

Another important finding from this study is that leave requests, reports of diminished well-being, and ongoing health treatments are prominent indicators of burnout. This underscores a potential gap in teacher healthcare support within schools, with 80.7 % of respondents seeking medical consultations in the current or previous month, and 44 % attributing these visits to work-related reasons.

Conversely, a noteworthy finding related to the age and length of service of the respondents challenges prevailing literature: younger teachers with less teaching experience displayed higher scores on the burnout inventory. This highlights the potential benefits of integrating new teachers into the school environment and bolstering support networks to alleviate these challenges.

Finally, to reduce the incidence of burnout among teachers, it is important to consider offering socioemotional skills training in schools and foster the adoption of coping strategies among elementary school teachers, as professionals equipped with strategies for dealing with stressful situations will be better prepared for interactions within the school community. For instance, developing an educational booklet that incorporates practical and supportive strategies for teachers' mental well-being, informed by findings from this study and insights from social skills literature, could serve as a valuable initiative.