INTRODUCTION
The pandemic has caused previously unexperienced necessary changes, such as temporal lockdowns, social distancing and remote learning/working. This huge, long- term challenge for teachers in all stages of education is associated with increased stress levels and lowered well- being, which puts this particular professional group in high-risk of mental health issues [1, 2].
However, projects focusing on teachers’ well-being during the pandemic are scarce. The current studies underline the impact of the pandemic on teachers in terms of quality of life, burnout and PTSD [3-5]. In the Spanish study concerning both compulsory and non-compulsory education teachers from the Basque Autonomous Community and Navarre [6], 50.6% of the sample (n = 1,633) indicated that they were suffering from stress, 49.5% – from anxiety, and 32.2% – from depression. Another Spanish study [7] examined 345 teachers, with a mean age of 44.62 years (SD = 9.53; 264 women), teaching in Spanish primary and secondary schools during lockdown. The average GHQ-12 score of the sample (M = 22.05; SD = 5.26) indicates symptoms of emotional problems. These emotional problems are predicted negatively by the time devoted to physical activity weekly and positively by the number of hours spent on teaching activity.
A Greek study concerning mental health of secondary school teachers (N = 226, 63.3% female) was conducted from March 24 to March 29, 2020. Gender seems to be a factor affecting the teachers’ fears (p = 0.001). It is noteworthy that women reported 2.5 times greater fear than men. However, the study did not include any specific scale to measure the mental health impact, and the questionnaire itself included 5 single-answer multiple-choice questions (regarding fear and stress, optimism about the outcome, depression, desire to return to work, concern over the implementation of distance learning) with answers given on a 6-point Likert scale. The results showed a correlation of stress/fear with depression (p = 0.001) as well as optimism (p = 0.001) [8].
Li et al. [9] carried out a study involving a large group of teachers in China (N = 88,611). The overall anxiety prevalence during COVID-19 pandemic was 13.67% (SE 0.12%). The prevalence was higher for women than for men (13.89% vs. 12.93%). Mild anxiety was most common (38.73%) in the 30-40 age group. It is worth referring these results to the earlier data. Huang et al. [10] reported the prevalence of anxiety to be 4.98% in 2013 according to China Mental Health Survey (CMHS) (N = 32,552 participants). The higher prevalence may be associated with threats related to the COVID-19 pandemic.
The present study aimed at understanding the nature of factors that protect and deteriorate mental well-being of Polish teachers in the COVID-19 pandemic. We have narrowed down the question of teachers’ mental health predictors to a set of sociodemographic, health, work- related and COVID-19-related factors.
METHODS
Participants
The study was carried out with a total sample of 2,757 teachers, where 2,372 (86%) were female and 381 (13.8%) were male. The mean age was 46.14 (SD = 9.35). Detailed socio-demographic characteristics of the study sample are presented in Table 1.
Table 1
Socio-demographic characteristics of the study sample (N = 2,757)
Almost 40% of the respondents (n = 1,069) were employed in primary schools (grades 4-8) and 34% worked in the secondary level education. The mean years of work experience in the sample was 20.94 (SD = 10.60). At the moment of completing the survey, most of the respondents (n = 1,488; 54%) were working on-site, 24% were still working remotely and 22.1% were engaged in the hybrid model. The mean duration of remote work in the previous 3 months was 7.83 weeks (SD = 4.61) (see Table 2).
Table 2
Work-related and COVID-19 related characteristics of the study sample (N = 2757)
Among the participants, 21.3% had been previously diagnosed with COVID-19 (n = 586), the majority (80.3%) experienced fears of negative consequences of COVID-19, and knew of cases of COVID-19 among close family members or friends (65.9%) or co-workers or students (89.6%). Nearly one in four respondents (24.9%) experienced death of a family member or friend by COVID-19. Thirteen percent (n = 359) reported deaths caused by COVID-19 among co-workers/students/parents of students (see Table 2).
The total score on the DASS-21 scale in the study sample of teachers during the COVID-19 pandemic (M = 15.41; SD = 13.43) does not differ from the results obtained in the validation study of the DASS-21 scale (M = 15.87; SD = 11.78) [13], Welch’s t-test: t(255.0893) = 0.54; p = 0.588. There was also no difference in the results in the Stress subscale, t-test: t(2967.0000) = –0.05; p = 0.958; Anxiety subscale, t-test: t(2967.0000) = 0.97; p = 0.334; or Depression subscale, Welch’s t-test: t(255.0506) = 0.32; p = 0.751 (see Table 3).
Table 3
Summary of the results on the DASS-21 scale in study sample (N = 2,757)
[i] DASS-21 cut off points according to Lovibond and Lovibond [11].
Procedure
This observational, cross-sectional national study, conducted in Poland from 3 March to 11 April 2021, involved teachers of all educational levels. Invitations to participate in the survey were disseminated through the principals of primary, secondary, vocational and technical schools, and other educational centres. All procedures were performed in accordance with the guidelines of the Maria Grzegorzewska University Ethics Committee and the 1964 Helsinki declaration as amended. The teachers who participated in the study filled out a semi-structured online questionnaires including their sociodemographic data, critical life events, the pandemic stressors relating to remote teaching. The teachers also responded to some detailed questions regarding their mental health.
Instruments
Teachers’ psychopathological symptoms were measured using the authorized translated Polish version of the screening tool, i.e. the Depression, Anxiety and Stress Scale (DASS-21) [11, 12], which consists in three subscales: depression, anxiety and stress. Each of the three subscales contains 7 items with a score from 0-3 on a Likert scale. The socio-demographic, health- and COVID-related data were collected using the self-administered online questionnaire. In order to collect information on the respondents’ general health status, they were asked about their chronic diseases (both mental and somatic), if any. Also, the information related to COVID-19 was sought, including COVID-19 diagnosis, fear of nega tive health consequences, change of work mode during the pandemic.
Data analysis
A series of stepwise multiple regression analysis were conducted to determine the variables that could predict factors relating to mental well-being. We used the hierar chi cal multiple regression analysis series to examine the unique contribution of sociodemographic and health status (first block of independent variables), work-related (second block) and COVID-19-related variables (the final model) in severity of depression, anxiety and stress (DASS-21). This sequential order of entry was based on an a priori hypothesis in which the additional variance of anxiety, stress and depression may be explained by the work-related characteristics, and experiences related to COVID-19 measured after accounting for the variance related to the participants’ demographics.
RESULTS
The descriptive statistics of the variables investigated are displayed in Tables 1-3.
The results of hierarchical regression analysis for sociodemographic, health, work-related and COVID-19-related factors as predictors of stress (DAss-21)
The analysis indicated that females were more stressed than males, when all variables were included in the model (β = –0.08, t = –4.26; p < 0.001). The age was also an important predictor of stress severity: the younger participants reported more stress than the older teachers (β = –0.17, t = –3.87; p < 0.001). Suffering from physical illness (β = 0.07, t = 3.52; p < 0.001), mental illness (β = 0.14, t = 7.67; p < 0.001), both physical and mental illness (β = 0.11, t = 6.06; p < 0.001) and being in the diagnostic process (β = 0.13, t = 7.14; p < 0.001) were the significant predictors of increased stress when compared to healthy participants (see Table 4).
Table 4
A summary of regression analysis for sample characteristics as predictors of stress (DASS-21) (N = 2,757)
Model 1: F(7, 2716) = 31.109, p < 0.001; R = 0.272, R2 = 0.074, R2 adj = 0.072. Model 2: F(19, 2704) = 13.329, p < 0.001; R = 0.293, R2 = 0.086, R2 adj = 0.079, ΔF(12, 2704) = 2.812, ΔR2 = 0.011, p < 0.001. Model 3: F(25, 2698) = 12.537, p < 0.001; R = 0.323, R2 = 0.104, R2 adj = 0.096, ΔF(6, 2698) = 9.258, ΔR2 = 0.018, p < 0.001
The analysis showed that of the work-related characteristics, location and remote method of teaching were significant predictors of the stress level. Participants who worked in a large city reported a higher level of stress than those who worked in a village (β = 0.12; t = 2.50; p = 0.012). Teachers working mostly remotely experienced higher stress level in comparison to those who worked on-site (β = 0.08; t = 3.15; p = 0.002).
In the final regression model, fear of negative consequences of COVID-19 was also a significant predictor of stress severity (β = 0.09, t = 4.75, p < 0.001). Also, cases of COVID-19 among co-workers and students (β = 0.06, t = 2.98, p = 0.003) as well as death of a family member or friend (β = 0.05, t = 2.52, p = 0.024) were predictors of increased stress.
The results of hierarchical regression analysis for sociodemographic, health, work-related and COVID-19-related factors as predictors of anxiety (DAss-21)
The analysis indicated that females showed higher anxiety in comparison to males, when all variables were included in the model (β = –0.07, t = –3.54; p < 0.001). The younger participants reported a higher level of anxiety (β = –0.14, t = –3.12; p = 0.002). Suffering from physical illness (β = 0.11, t = 5.62; p < 0.001), mental illness (β = 0.17, t = 8.97; p < 0.001), both physical and mental illness (β = 0.11, t = 6.05; p < 0.001) and being in diagnosis (β = 0.14, t = 7.70; p < 0.001) were significant predictors of an increased anxiety compared to healthy participants (see Table 5).
Table 5
A summary of regression analysis for sample characteristics as predictors of anxiety (DASS-21) (N = 2757)
Model 1: F(7, 2716) = 31.124, p < 0.001; R = 0.273, R2 = 0.074, R2 adj = 0.072. Model 2: F(19, 2704) = 12.315, p < 0.001; R = 0.282, R2 = 0.080, R2 adj = 0.073, ΔF(12, 2704) = 2.812, ΔR2 = 0.005, p = 0.200. Model 3: F(25, 2698) = 12.060, p < 0.001; R = 0.317, R2 = 0.101, R2 adj = 0.092,ΔF(6, 2698) = 9.258, ΔR2 = 0.021, p < 0.001
The analysis indicated that among the work-related characteristics, only working at the primary school level was a significant predictor of lower anxiety (β = –0.10; t = –2.31; p = 0.021). Participants who worked at the primary education level had a lower level of anxiety.
When all independent variables were included in the final regression model, diagnosis with COVID-19 and fear of negative consequences of COVID-19 were significant predictors of anxiety (respectively: β = 0.05, t = 2.73, p = 0.006; β = 0.11, t = 5.81, p < 0.001). Teachers who experienced death of a family member or friend (β = 0.04, t = 2.20, p = 0.028) and/or co-workers/students/ parents of students (β = 0.04, t = 2.16, p = 0.031) experienced higher anxiety in comparison to other teachers.
The results of hierarchical regression analysis for sociodemographic, health, work-related and COVID-19-related factors as predictors of depression (DAss-21)
When all independent variables were included in the regression model, all sociodemographic characteristics (except disability) were significant predictors of depression. Males reported lower severity of depression than females (β = –0.06, t = –3.12, p = 0.002). The age was also a significant predictor of depression severity: the younger participants reported higher depressive symptoms than the older ones (β = –0.15, t = –3.46; p < 0.001). Suffering from physical illness (β = 0.06, t = 3.21; p = 0.001), mental illness (β = 0.17, t = 8.92; p < 0.001), both physical and mental illness (β = 0.12, t = 6.37; p < 0.001) and being in diagnosis (β = 0.13, t = 6.91; p < 0.001) were significant predictors of increased depression in comparison to healthy participants (see Table 6). The analysis showed that of the work-related characteristics, work location and remote method of teaching were significant predictors of depression. Participants who worked in a large city reported higher depressive symptoms than those who worked in a village (β = 0.06; t = 2.46; p = 0.014). Teachers working mostly remotely reported higher levels of depression than those who worked on-site (β = 0.07; t = 2.79; p = 0.005). In the final regression model, fear of negative consequences of COVID-19 was also a significant predictor of the depression severity (β = 0.05, t = 2.76, p = 0.006). Death of a family member or friend (β = 0.06, t = 3.09, p = 0.002) was also a significant predictor of increased depression among teachers.
Table 6
A summary of regression analysis for sample characteristics as predictors of depression (DASS-21) (N = 2757)
Model 1: F(7, 2716) = 27.641, p < 0.001; R = 0.258, R2 = 0.067, R2 adj = 0.064. Model 2: F(19, 2704) = 11.991, p < 0.001; R = 0.279, R2 = 0.078, R2 adj = 0.071, ΔF(12, 2704) = 2.737, ΔR2 = 0.011, p = 0.001. Model 3: F(25, 2698) = 10.435, p < 0.001; R = 0.297, R2 = 0.088, R2 adj = 0.080, ΔF(6, 2698) = 5.157, ΔR2 = 0.010, p < 0.001
DISCUSSION
The most important mental health predictors are: (1) demographic and health-related: gender, age, place of work, somatic and mental illnesses; (2) work-related: remote teaching and school type; (3) COVID-19-related: COVID-19 case among co-workers and students, the experience of death among family or friends and anxiety of negative consequences of COVID-19.
The results showed that the mental health of the younger, female teachers working in the cities are in danger. In general, women and younger population appeared to be more affected by negative mental consequences of the pandemic than men [14-18]. Teachers, comparably to medical frontline workers, can be at a higher risk of experiencing symptoms of depression, anxiety, insomnia and stress [19]. The work location also proved to be an important predictor of severity of symptoms of stress and depression. Teachers who worked in a village had a lower severity of stress and depression compared to teachers who worked in a city. Indeed, most Polish cases were recorded in cities, particularly in the large ones, hence work in a large city may be associated with a greater risk of a COVID-19 infection.
Remote teaching is the next important predictor of stress and depression. In many countries, including Poland, teachers had no previous experience of remote teaching. They faced various technical problems related to the quality of the Internet connection and new tools they were forced to use. Another problem was the children’s access to the Internet and lack of their own computers. Online lessons require self-studying, often without teachers’ support, which may deteriorate the learning performance and increase anxiety levels of the students, and, secondarily, impair the well-being of teachers.
It should be noted, a number of domestic violence emergency calls received by child abuse helplines increased, not only in Poland but also worldwide [20, 21]. During online lessons, teachers may have considerable difficulty assessing students’ domestic problems, including material difficulties, parental addiction to psychoactive substances or domestic violence. At the same time Polish national reports indicate that teachers still need support conducting online classes. They have been experiencing difficulties expanding the repertoire of methods, tools and material used during classes. With regards to the didactic process, teachers mostly use informal, easily-accessible sources of digital materials and tools, for example those found on social media groups [22, 23].
Polish researchers indicated that the level of depression and anxiety experienced by adolescents increased during the pandemic [24]. The teachers in our sample noted significant risks for pupils related to distance learning, such as: risk of exclusion (in terms of technical, equipment and school competence), decline in students’ physical activity, and general overload related to long hours of work in the technological environment. The teachers observed severed relationships, isolation and apathy among students. Their awareness of the distance learning effects on the students’ mental health, such as depression and anxiety, may have caused feelings of helplessness.
Furthermore, teachers working remotely have to solve various educational and technical problems on their own without their colleagues’ support and principals’ supervision. As indicated in our study, remote work was associated with higher levels of stress and depression than classroom teaching in the COVID-19 pandemic. The mental well-being of teachers should be taken into account when planning a long-term strategy regarding reopening schools [25-27].
Until now, very few studies in Europe have analyzed the psychological impact of the pandemic and remote teaching on teachers, which is the main strength of this study. The results of other Polish and European studies are convergent to the study presented in this paper [28-31]. However, these findings need to be interpreted in the context of some limitations: (1) the consequences of cross- sectional study design (2) the sampling technique which relies on digital infrastructure and voluntary participation could increase selection bias. Although the sample size is large, the generalizability of our results is limited, since we used a non-probabilistic sample. In particular, the only teachers that had the opportunity to participate were the ones who had internet access. Replications should be conducted in different countries and regions, as the socioeconomic factors, availability of local mental health service, the stage of the pandemic and culture can play an important role in mental health during the pandemic.
CONCLUSIONS AND PRACTICAL IMPLICATIONS
Our results have several practical implications. Teachers’ support programs should be implemented on the basis of several pillars. First of all, mental health screening and interventions should be focused on specific groups of teachers. In particular, the focus should be on young women who work remotely. It would be useful to implement simple but empirically validated interventions for the whole teacher population. Finally, the offer of psychological support for teachers should be expanded, e.g., through support groups or at least regular supervision.
Overall, the COVID-19 pandemic has a significant impact on the psychological well-being of teachers. The deterioration of their mental well-being may have an impact on students, therefore prevention programs for teachers should be implemented [25-27]. In this way, further negative consequences of the pandemic can be limited, not only for the teachers, but also for the students’ population and the education system as well.
