Teachers' Response to Stress, Anxiety and Depression During COVID-19 Lockdown: What Have We Learned From the Pandemic?

BACKGROUND: As a result of national lockdown due to the COVID-19 outbreak, teachers were forced to suspend their classes and replace them with online teaching and home schooling. Additional stressors such as competing family responsibility have increased their worries and mental health problems. The aim of this study was to determine the impact of COVID-19-related worries on teachers' emotional symptoms, considering the mediating role of several protective factors.
METHODS: A total of 614 Chilean teachers (94.60% women) participated in this study using a cross-sectional design and incidental sampling method. Self-report data was collected assessing emotional symptoms, COVID-19-related worries, life satisfaction, affect balance, and resilience. Descriptive analyses, Pearson's correlations, hierarchical regressions, and mediation models were conducted.
RESULTS: The results indicated that emotional symptoms were associated with prepandemic physical and mental health problems, higher levels of worries and negative affect, as well as lower levels of life satisfaction and resilience. Results from the mediation models showed that the negative impact of COVID-19-related worries on emotional symptoms was alleviated by affect balance and resilience.
CONCLUSIONS: These findings highlight the importance of addressing the risk and protective factors for teachers' mental health during exceptional situations such as the ongoing pandemic.

The outbreak of SARS‐CoV‐2 in Wuhan by the end of 2019 and its rapid global spread eventually led to the COVID‐19 pandemic. Due to the lack of effective treatments and with the purpose of reducing the number of infections and deaths, governments around the world decreed more or less restrictive isolation measures. , Specifically in Chile, a state of alarm was decreed, imposing general lockdown from March until July 17, 2020. In this exceptional situation, schools remained closed and millions of schoolchildren were partially or completely deprived of access to education, further aggravating socioeconomic differences.

Most of the previous research have studied the impact of school closures on students in terms of mental health and learning, and have addressed the need to prepare teachers for changes in teaching methods in the future. , , , , However, few studies have addressed the psychological distress of teachers in the context of COVID‐19 lockdown despite the fact that their mental health may directly influence the emotional well‐being of their students. , , ,

Teaching is considered a highly stressful profession, especially in high schools dealing with adolescents. A recent meta‐analysis of the emotional impact of COVID‐19 on teachers concludes that 17% experience anxiety, 19% depression, and 33% stress. Teaching is considered to be a highly stressful profession because teachers have to cope with work overload, emotion management, and conflict mediation in the classroom on a daily basis. , In addition, teachers have experienced a great amount of stress during the pandemic, adapting to remote teaching and learning, reconciling their personal and work responsibilities, the fear about their future and the pandemic itself, thus, their emotional distress may be particularly severe. , ,

The psychological impact of social isolation can be intense, substantial, and long‐lasting, and usually includes a wide range of psychological symptoms such as stress, anxiety, and depression. These emotional difficulties have been exacerbated in individuals with previous physical and mental health problems. In this sense, the appearance and aggravation of emotional difficulties has been greater in women. , Furthermore, the recent literature has indicated that age moderates the psychological impact of the pandemic, showing that young people are more likely to suffer during this health crisis. ,

However, some studies have focused on protective factors reducing the effects of social isolation and lockdown. , , Some studies have reported positive outcomes of COVID‐19 lockdown, with some people starting to value certain aspects of their lives that they did not previously appreciate. These resilient individuals adapt to change more easily, overcome adversity and emerge emotionally strengthened, and therefore show higher life satisfaction and emotional balance. , , , Similarly, experiencing positive emotions after adversity may help people to better manage their emotions and may protect them against future mental health problems. Hence, maintaining adequate levels of positive and negative affect in stressful situations may promote mental health and well‐being.

Considering the above, the main purpose of the present study was to determine the risk and protective factors of teachers' mental health during COVID‐19 lockdown in Chile. Additionally, we examined the impact of COVID‐19‐related worries on teachers' depression, anxiety, and stress symptoms, considering the mediating role of several protective factors. (H1) We expected to find a greater emotional impact among teachers with previous physical and mental health problems, that is higher levels of emotional symptoms (depression, anxiety, and stress) and higher levels of worries about COVID‐19 and its consequences, combined with lower levels of life satisfaction, lower emotional balance, and resilience. We also expect female teachers and those working in secondary education (high schools) to be particularly vulnerable to emotional distress. (H2) We expect to find that emotional symptoms would be associated with worries about COVID‐19 and their consequences and both would be negatively related to life satisfaction, emotional balance, and resilience. (H3) We expect that teachers' levels of depression, anxiety, and stress would be predicted by worries about COVID‐19 and its consequences, as well as lower levels of resilience, life satisfaction and emotional balance, controlling for demographic effects (gender, educational level, previous physical, or mental health problems). (H4) We expect to find an indirect effect of worries about COVID‐19 and its consequences on emotional symptoms mediated through resilience, life satisfaction, and emotional balance.

METHODS

Participants

In this study, a cross‐sectional descriptive design was used, and data were collected through anonymous self‐report questionnaires. Participants were Chilean teachers who were retrieved through Internet and social networks, using a snowball sampling method.

In total, 614 Chilean teachers (94.6% female) aged between 20 and 65 years (Meanage = 33.37; SD = 7.73) participated in this study. According to the latest OECD Review of School Resources in Chile, in 2015 there were about 225,000 teachers working in preprimary, basic to upper secondary and tertiary education including all school types (municipal schools, private‐subsidized and nonsubsidized schools, schools with delegated administration). By choosing a 99% confidence level and a sample size of 614 participant the margin of error would be ±5.2%, which is acceptable. Participants represent all areas of the Chilean educational system: 8.3% were preprimary teachers educating children up to the age of 5; 81.8% primary and secondary schoolteachers providing compulsory education for students aged 6 to 17 years; 7.2% tertiary teachers who taught at universities, professional institutes, or technical centers; 2.8% others.

Instrumentation

Participant characteristics were assessed by ad hoc questions about basic demographic information (gender and age), educational level (preprimary, primary, secondary, or tertiary teacher), prepandemic physical health, and/or mental health problems.

Symptoms of depression, anxiety, and stress were assessed with screening tool Depression, Anxiety, and Stress Scales (DASS‐21). The DASS‐21 comprises 21 items, 7 items for each of the 3 subscales (symptoms of depression, anxiety, and stress), which are rated on a 4‐point Likert scale, resulting in a total score ranging from 0 (absence of symptoms) to 21 (severe symptomatology). The DASS‐21 scale has shown adequate psychometric properties in Spanish speakers. In our study, the Cronbach's alpha, and McDonald's omega reliability values of the 3 subscales were excellent: depression (α = .93; ω = .88), anxiety (α = .87; ω = .88), and stress (α = .90; ω = .89).

Worries about COVID‐19 were assessed using the worries about COVID‐19 and its consequences Scale (W‐COV). The W‐COV was designed to assess worries about health, economic, and political circumstances, academic and/or occupational performance, social relationships, and psychological health. The scale consists of 16 items which were scored on a 5‐point Likert scale ranging between 1 (Almost never) and 5 (Almost always). In the original cross‐cultural validation study conducted in the Spanish‐speaking adolescent population, results from a confirmatory factor analysis suggested adequate structural characteristics of the 1‐factor solution of the scale (KMO = .83; RMSEA = .08; SRMR = .09; CFI = .91; GFI = .94). In our study, the reliability indices of the W‐COV were satisfactory: α = .88; ω = .89.

Life satisfaction was measured by the Satisfaction with Life Scale (SWLS). The SWLS assesses the cognitive component of life satisfaction, rather than focusing on affect. A total of 5 items are scored on a 7‐point Likert scale ranging between 1 (Totally disagree) and 7 (Totally agree). The SWLS has shown adequate properties in Spanish‐speaking teachers. , In our study, internal consistence indices were satisfactory: α = .87, ω = .86.

Positive and negative affect were assessed by the Scale of Positive and Negative Experience (SPANE). The SPANE measures pleasant (positive affect) and unpleasant feelings or affects (negative affect) on a 5‐point Likert scale (1 = never; 5 = always). The 12‐items are divided into 2 subscales (6 items each), but a global affect balance can be calculated. The SPANE has shown adequate psychometric properties in a Spanish‐speaking population. In this study, acceptable reliability indices were obtained: positive affect (ω = .93, α = .93), negative affect (ω = .88, α = .87), affect balance (ω = .72, α = .91).

Resilience was measured by the abbreviated version of the Connor‐Davidson Resilience Scale (CD‐RISC). The Spanish version of the CD‐RISC consisted of 10 items measuring general resilience. Participants scored items on 5‐point Likert scale ranging between 0 (Never) and 4 (Almost always). CD‐RISC has shown adequate psychometric properties. In our study, the reliability of the CD‐RISC was appropriate (ω = .84, α = .81).

Procedure

Participants were all informed of the purpose of the study, the confidentiality of their participation on a voluntary basis, and the anonymity of responses. After signing the informed consent form, we assessed participants using the Limesurvey platform, accessible from any electronic device (smartphone, tablet, laptop, etc.) during May and June 2020. This study was an anonymous online survey and participants were informed, before providing consent, that inclusion of personal information was discouraged. Data encryption was not deemed necessary or applicable, as data were completely anonymous. Data were kept in the online platform and only study investigators had access to them, while no paper‐based files were utilized.

Data Analysis

All statistical analyses were performed using SPSS V.26. Preliminary data analyses included descriptive statistics, independent t‐tests and analyses of variance (ANOVA) with Scheffé's correction to test for post hoc differences between groups. Effect sizes were calculated using Cohen's d (small ≈ 0.2, medium ≈ 0.5, high ≈ 0.8) and eta partial square η 2. (small ≈ 0.01, medium ≈ 0.06, high ≈ 0.14). We then calculated Pearson correlations and hierarchical stepwise regression models for symptoms of depression, anxiety, and stress. Mediation analysis was done with the PROCESS Macro (V.3.3) performing simple and multiple mediation analyses (models 4 and 6) with 10,000 boostraping samples and a confidence level of 95%.

RESULTS

Descriptive Statistics and Bivariate Correlations

Teachers showed high levels anxiety (mean = 13.84; SD = 10.67; 95% IC [12.99, 14.69]), depression (mean = 13.63; SD = 10.19; 95% IC [12.82, 14.43]), and stress (mean = 29.80; SD = 10.40; 95% IC [19.97, 21.62]). Specifically, 39.90% suffered from severe or very severe anxiety symptoms, 23.20% from depressive symptoms and 46.60% from stress (Figure 1).

Figure 1

Levels of Stress, Anxiety, and Depression in Teachers

What concerned them most about COVID‐19 were aspects related to work, mental health and the health situation. In line with the above, moderate life satisfaction was observed (mean = 25.25; SD = 6.00; 95% IC [24.78, 25.73]), accompanied by moderate emotional balance (mean = 3.06; SD = 7.58; 95% IC [2.46, 3.66]). Finally, it was observed that they generally presented low levels of resilience (mean = 26.39; SD = 5.75; 95% IC [25.94, 26.85]). Specifically, 57.20% showed a low level of resilience, and only 4.90% showed a high level of resilience.

With reference to the differences in means, women showed less resilience (t = 6.12, p ≤ .05, d = .40) and more anxiety than men (t = 2.20, p ≤ .05, d = .43). On the other hand, participants with physical or mental health problems had lower psychosocial adjustment (Table 1). Finally, primary and secondary school teachers were more anxious (F = 3.33; p ≤ .05; η 2 = .14) and stressed (F = 5.92; p ≤ .001, η 2 = .17) than university teachers. Overall, medium effect sizes were observed.

Table 1

Descriptive Analyses, Differences in Means and Effect Sizes of the Variables Under Study

	Previous Physical Health Problem					Previous Mental Health Problem
	Yes	No				Yes	No
	Mean (SD)	Mean (SD)	t	p	d	Mean (SD)	Mean (SD)	t	p	d
Anxiety	16.25 (10.49)	12.36 (10.52)	4.45	.001	.37	17.56 (10.29)	11.65 (9.89)	6.88	.001	.59
Depression	14.59 (10.14)	13.03 (10.17)	1.85	.065	.15	16.71 (9.93)	11.80 (9.98)	5.93	.001	.50
Stress	22.14 (10.24)	19.97 (10.43)	.51	.012	.21	24.41 (9.18)	18.66 (10.50)	7.10	.001	.58
Worries about COVID‐19	3.43 (0.64)	3.26 (0.68)	3.02	.003	.26	3.44 (0.63)	3.26 (0.68)	3.25	.001	.28
Life satisfaction	24.38 (6.34)	25.79 (5.72)	2.77	.006	.23	23.93 (6.22)	26.02 (5.75)	4.16	.001	.35
Emotional balance	2.10 (7.27)	3.64 (7.71)	2.50	.013	.20	1.02 (7.06)	4.23 (7.63)	5.32	.001	.45
Positive affect	20.11 (4.19)	20.89 (4.16)	2.24	.026	.19	19.70 (3.97)	21.12 (4.23)	4.10	.001	.35
Negative affect	18.01 (4.47)	17.24 (4.33)	1.85	.065	.17	18.68 (4.12)	16.86 (4.41)	5.06	.001	.42
Resilience	25.82 (6.12)	26.74 (5.49)	1.91	.057	.16	24.94 (5.77)	27.23 (5.57)	4.90	.001	.45

d, effect size; p, p value; t, t value.

We observed linear, moderate, and statistically significant associations between the variables studied (Table 2). Age was significantly and negatively associated with worries, symptoms of depression, anxiety, and stress, and positively associated with resilience and emotional balance. Emotional symptoms were positively and significantly associated with COVID‐19 worries, and negatively associated with life satisfaction, emotional balance, and resilience. Higher resilience was positively associated with life satisfaction and emotional balance.

Table 2

Intercorrelations Between Variables Studied

	1	2	3	4	5	6	7	8
Age	1
Anxiety	−.14**	1
Depression	−.22**	.67**	1
Stress	−.19**	.79**	.73**	1
Life satisfaction	.05	−.34**	−.46**	−.32**	1
Emotional balance	.19**	−.55*	−.70**	−.65**	.52**	1
Resilience	.16**	−.35**	−.45**	−.36**	.36**	.51**	1
Concerns	−.17**	.57**	.54**	.60**	−.29**	−.49**	−.26**	1

p < .05.

p < .01.

Prediction Models

The predictive power of the studied variables was analyzed conducting a hierarchical regression model (HRM). The criterion variables were anxiety, depression, and stress. The prediction model was conducted in 3 steps (Table 3): sociodemographic or control variables (age, gender, prepandemic physical or mental health problems, and education levels) were included first, followed by COVID‐19 worries and finally variables related to well‐being (life satisfaction, emotional balance) and resilience.

Table 3

Hierarchical Regression Model of Anxiety, Depression, and Stress

	Anxiety				Depression				Stress
Predictor	ΔR2	ΔF	β	t	ΔR2	ΔF	β	t	ΔR2	ΔF	β	t
Step 1	.13	17.50***			.12	15.87***			.12	16.62***
Gender			−.04	1.33			.06	2.25*			−.01	−.37
Age			−.02	.62			−.08	2.91**			−.04	−1.37
Scope of work			−.02	.70			.02	.78			−.03	−1.18
Physical health problem			.08	2.61**			−.02	.63			.01	0.20
Mental health problem			.14	4.38***			−.08	2.87**			.13	4.58***
Step 2	.24	248.75***			23	218.22***			.28	284.53***
Concerns	.08		.37	10.69***			.25	8.07***			.35	10.95***
Step 3		29.43***			.23	108.64***			.14	61.15***
Life satisfaction			−.03	.75			−.10	3.31**			.05	1.38
Emotional Balance			−.28	6.76***			−.45	12.07***			−.45	11.64***
Resilience			−.06	1.73			−.10	3.15**			−.03	.76
Durbin‐Watson	1.87				2.01						1.87
R 2 ajd	.45***				.57***						.53***

p ≤ .05.

p ≤ .01.

p ≤ .001.

ΔR2, change in R 2; ΔF, change in F; ß, regression coefficient; t, value of t‐test statistic.

The results of the hierarchical regression explained 45% of the variance of teacher's anxiety. In the last step, previous physical, mental health problems and worries about COVID‐19 presented a positive and statistically significant beta coefficient. Emotional balance presented negative and statistically significant beta coefficients. Thus, anxiety was predicted by preexisting physical and mental health problems, worries about COVID‐19, and lower levels of emotional balance.

On the other hand, in the depression prediction model, the variables included in the model predicted 53% of variance. In this last step, gender (being a woman), the presence of a previous mental health problem and concerns about COVID‐19 had a significant positive beta coefficient. Age (being younger), emotional balance, and resilience, had a significant negative beta coefficient. Thus, depression was predicted by being female, young, with a previous mental health problem, with more worries, lower life satisfaction, and lower resilience.

Finally, in predicting teacher stress, the model explained 53% of the variance. In the last step, the presence of a previous mental health problem and worries about COVID‐19 presented a significant positive beta coefficient, and the emotional balance, presented a significant negative beta coefficient. Thus, stress was predicted by having a previous mental health problem, greater worries, and worse emotional balance.

Mediating Models

We conducted a multiple mediation analysis with three mediators (emotional balance, life satisfaction, and resilience), controlling for the effects of sociodemographic variables (age, gender, previous physical or mental health problem, and education levels) for each of the dependent variables (symptoms of depression, anxiety, and stress). These models were not significant (p > .05) for the prediction of anxiety and stress, so, based on the previous results, we performed a simple mediation analysis for these variables, taking into account worries as an independent variable and emotional balance as a mediating variable. In the case of anxiety, the effects of previous physical and mental health problems were controlled for. In the case of stress, we controlled for previous emotional distress. Finally, we conducted a multiple mediation analysis for the prediction of depression with 2 mediators (emotional balance and resilience), controlling for the effect of previous psychological problems and gender (Table 4).

Table 4

Mediation Analysis and Multiple Mediation Analysis

IV	M	DV	R2	p	Effect of IV on M	Effect of M on DV	Direct Effect	Interdirect Effect	95% CI for Indirect Effects	Total Effect
Worries COVID‐19	Emotional balance	Anxiety	.38	≤.001	.40***	.05***	.55**	2.47***	1.85 to 3.18	.52***
Worries COVID‐19	Emotional balance	Stress	.27	≤.001	.40***	.04***	.49***	3.27***	2.65 to 3.95	.49***
IV	Ms	DV	R2	p	Effect of IV on Ms (a1, a2)	Effect of Ms on DV (b1, b2)	Direct effect (c)	Interdirect effect (c′)	95% CI for indirect effects	Total effect
Worries		Depression	.35	≤.001			.3.80***	0.61*		7.56***
	Resilience				−1.83***	−.20***		.36*	.11 to .69
	Emotional balance				−4.19***	−.66***		2.78*	2.12 to 3.48

p ≤ .05.

p ≤ .01.

p ≤ .001

DV, dependent variable; IV, independent variable; LLCI, lower level of the 95% confidence interval; Ms, mediators; R 2, coefficient of determination; ULCI, upper level of the 95% confidence interval.

Bootstrap samples = 10,000.

In the first mediation model, COVID‐19‐related worries were a good predictor of both emotional balance (β = .40, p ≤ .001) and anxiety (β = .55, p ≤ .001). Emotional balance was directly associated with symptoms of anxiety. (β = 0.05, p ≤ .001). The indirect effect of worries on anxiety through emotional balance was significant (effect = 2.47; CI = [1.85, 3.18]). The relationship between higher levels of worries about COVID‐19 and higher levels of anxiety was mediated by the teachers' emotional balance. Together, the direct and indirect effects predicted a total variance of 38%.

In the second model, COVID‐19 related worries predicted emotional balance (β = .40, p ≤ .001) and stress (β = .40, p ≤ .001). Emotional balance was also directly related to stress (β = .04, p ≤ .001). The indirect effect of worries about COVID‐19 on stress through emotional balance was significant (effect = 3.27; CI = [2.65, 3.95]). The positive association between worries about COVID‐19 and stress was mediated by the teachers' emotional balance, with the model predicting 27% of the variance.

Finally, in the third mediation model, COVID‐19 related worries had a direct negative effect on resilience and emotional balance, and a positive effect on depressive symptoms. In turn, resilience and emotional balance had a direct negative effect on depression. The impact of the indirect effects of COVID‐19 worries through resilience (effect = .36; CI = [.11, .69]) and emotional balance (effect = 2.78; CI = [2.12, 3.48]) on teachers' symptoms of depression were significant. Overall, the direct and indirect effects predicted a total variance of 32% when introducing the two mediating variables in the relationship between worries about COVID‐19 and depression.

DISCUSSION

The aim of the present study was to identify the potential risk and protective factors for stress, depression, and anxiety disorders among Chilean teachers during the first weeks of lockdown due to the COVID‐19 pandemic. This study will help to better understand how we may protect teachers in extraordinary situations and/or contexts so that they are able to continue their valuable educational service.

First, we expected to find a greater emotional impact among teachers with previous mental or physical health problems. Our results confirmed this first hypothesis, showing that teachers with previous health problems (physical or mental) experienced higher levels of emotional symptoms (depression, anxiety, and stress) as well as lower life satisfaction, lower emotional balance, lower levels of resilience and they were more concerned about COVID‐19 and its consequences in comparison to healthy teachers. These findings are consistent with previous studies reporting that people with prepandemic mental health disorders are more at risk of experiencing increased emotional distress due to high susceptibility to stress compared with the general population.

In addition, we expected female teachers and those working in higher education to be particularly vulnerable to emotional distress. In line with previous literature, female teachers in our study showed lower levels of resilience and experienced higher levels of anxiety than their male colleagues. , Traditionally, women take care of their children and/or older family members in addition to carrying out domestic work with a paid profession. The COVID‐19 pandemic has dramatically increased this double burden and is therefore very likely to have a negative impact on women's health and well‐being. , Contrary to our expectations, teachers who work at elementary and secondary schools experienced higher levels of anxiety and stress than high school teachers. This finding may stress the difficulties that primary and secondary schoolteachers face when they are forced to switch to improvised online learning and homeschooling for students of younger age, who are likely to need more instructions and guidance.

Our results support the second hypothesis that emotional symptoms are associated with worries about COVID‐19 and their consequences and both are negatively related to life satisfaction, emotional balance, and resilience. Thus, teachers who are more concerned about COVID‐19 and its consequence also experience higher levels of depression, anxiety, and stress. Furthermore, those teachers, who are more satisfied with their life, emotionally balanced and more resilient, are also less concerned about COVID‐19 and less emotionally distressed. Our findings are consistent with previous literature that has established a positive association between worries about Covid‐19 and mental health problems, but a negative association with resilience and well‐being. , ,

Furthermore, we expected that teachers' levels of depression, anxiety, and stress would be predicted by worries about COVID‐19 and its consequences, as well as lower levels of resilience, life satisfaction and emotional balance, controlling for demographic effects (gender, educational level, previous physical or mental health problems). Our results confirm this third hypothesis, showing that teachers with prepandemic mental health problems, high levels of COVID‐19 worries and lower levels of emotional balance reported more emotional symptoms, which is consistent with previous research. , Nevertheless, life satisfaction and resilience were only significant predictors of teachers' depression symptoms, which might indicate the more relevant role of emotional balance in the process of teacher's adaption to the additional stressors of the current global pandemic. ,

Supporting our central hypothesis, results from multiple mediation analyses only partially confirmed the indirect effect of worries about COVID‐19 and its consequences on emotional symptoms mediated through resilience, life satisfaction, and emotional balance. On the one hand, teachers who reported high levels of COVID‐19 worries but showed strong emotional balance experienced lower levels of depression, anxiety, and stress than those who worried about COVID‐19 and its consequences but showing low levels of emotional balance. Resilience also mediated the positive association between worries about COVID‐19 and emotional symptoms. It is worth mentioning that teachers from the present study sample showed rather low levels of resilience and only moderate levels of emotional ability in combination of high levels of overall emotional symptoms. Thus, our findings reveal the urgent need to foster teachers' noncognitive skills such as emotional balance and resilience in order to protect them from the negative impact of worries about the COVID‐19 and its consequences on increased mental health problems. ,

In sum, our findings indicate that worries about COVID‐19 and its consequences are positively associated with symptoms of depression, anxiety, and stress, but negatively related to resilience, emotional balance and life satisfaction in Chilean teachers. Furthermore, both resilience and emotional balance are protective factors of emotional symptoms, meaning that teachers with higher levels of these variables display lower levels of mental health problems. In addition, the positive association worries about COVID‐19 and its consequences and emotional symptoms is mediated by emotional balance.

Limitations

Despite the many strengths of this research, several limitations should be considered: First, the inherent limitation of collecting data through online survey puts limits to the generalizability of our findings in the study sample to the general population. However, participants did not charge or receive any financial reward for their participation in our study, which reduces the concern that they would intentionally respond inappropriately, as the primary motive for completing the survey was to express their opinion. In addition, we believe that self‐report measures were appropriate for the study sample given that adults are reliable sources of their internal states. We consider that another limitation may be the unequal sample size in terms of gender. The majority of participants were women. This may have influenced comparisons by gender. However, the authors have calculated effect sizes to control differences in sample size. In future research, however, mixed methods (qualitative and quantitative data), multiple reports from students might be used, in order to contrast the information about teachers mental health. Finally, cross‐sectional design does not allow for casual inferences, which can be addressed in longitudinal studies that evaluate changes in mental health conditions and needs following the pandemic outbreak.

Conclusions

This study makes an important contribution to the current COVID‐19 literature by analyzing a predictive model in which resilience and emotional balance act as a buffer between worries about COVID‐19 and its consequences and teacher's mental health. These and other factors of psychological strengths that mitigate the psychological impact of the pandemic crisis on teachers are crucial to promote their mental health in light of the various challenges facing school education during the current pandemic and in any other extraordinary situations and/or contexts.

IMPLICATIONS FOR SCHOOL HEALTH

What have we learned from the pandemic? Teaching is a challenging profession and the additional stressors related to the global health crisis are threating teachers' mental health. The future generation of teachers is facing a problematic work scenario, where teaching methods changes due to the digital revolution and social needs. How can politicians, school leadership, health professionals, and other stakeholders help teachers to cope with these challenges, enhancing their job motivation and wellbeing? It seems like personal resources such as resilience and emotional balance are crucial to ensure teachers health, especially in times of crisis. Teachers might benefit form specialized skill‐trainings that aim to promote their personal strengths such as social and emotional abilities. There is strong evidence that these social‐emotional education programs enhance teachers' wellbeing and reduce mental health problems, which in turn has positive implications for students and the whole school community.

Human Subject Approval Statement

The study has been approved by the Ethics Committee of the University of Valencia (Ref.n°:1595575567385) and the data collection followed the norms of the Declaration of Helsinki (World Medical Association, 2013).

Conflict of Interest

All authors declare no conflicts of interest in this paper.