According to the World Health Organization (2006), mental health is a state of well-being that enables individuals to successfully cope with the normal stresses of life and work productively and contribute to their community. Mental health and its assessment are one of the broadest and most complicated issues in psychology given the many contributing factors (Keyes et al., 2002; Franken et al., 2018), and adverse symptoms, for example, depression, anxiety, and somatization, reflect aspects of mental health status (Chekroud et al., 2017).

Teaching has one of the highest rates of stress and burnout, which interfere with mental health (Grayson and Alvarez, 2008; Miller et al., 2019). The issue of mental health and well-being in teachers is critical not only for their teaching performance but also for the students (Dods, 2016; Miller et al., 2019). On the one hand, plenty of research has reported a positive relationship between teachers’ mental health status and their job satisfaction and efficacy beliefs (Allan et al., 2016) and negative relationships between mental health status and burnout and depression (Jennings et al., 2017; Capone and Petrillo, 2018), which sometimes even leading to teachers’ leaving their professions (Bowles and Arnup, 2016). On the other hand, teachers’ mental health is correlated with the students’ developmental outcomes (McLean and Connor, 2018). A recent study found that secondary teachers’ better mental health was associated with students’ better well-being and lower psychological distress (Harding et al., 2019).

In regard to the status of preschool teachers’ mental health, the results are mixed. Although a recent meta-analysis suggested that in China, preschool teachers’ mental health conditions are at least as good as, if not better than, those of other kinds of teachers, for example, primary and secondary teachers (Fan et al., 2016); other research has reported different results, indicating that education is a work context in which professionals (teachers) are likely to suffer from burnout, which may be associated with low levels of mental health (Zhang et al., 2014). Compared with K-12 teachers, preschool teachers scored higher than Chinese national controls on the SCL-90 (Chen, 2009), but more detailed information concerning preschool teachers is absent. Accordingly, the present study aims to determine the status of preschool teachers’ mental health.

Empathy, the capacity to feel and understand the emotion of others, is widely accepted as a multifaceted construct that contains both cognitive and affective components (Preston and de Waal, 2002; Buck et al., 2017). Among a variety of empathy multicomponent assumptions, Davis’s (1983) point of view is representative. According to him, empathy contains four components: personal distress, empathic concern, fantasy, and perspective taking. Among these components, perspective taking, which is the trend and ability to walk in others’ shoes, is a typical cognitive component. Although both personal distress and empathic concern are regarded as emotional components, they are believed to have opposite functions, with personal distress making individuals focus on themselves and empathic concern enabling individuals to attend to the others. In addition, fantasy is individuals’ abilities and trends to transpose themselves into fictional situations. Another representative multicomponent assumption is Preston and de Waal’s (2002) russian doll model of empathy, which assumed that empathy is a layered structure, with personal distress and emotional contagion in its emotional core, the more advanced form of empathy (a mixture of emotional and cognitive components, e.g., empathic concern) in the middle layer, and cognitive component (e.g., perspective taking) in the outer layer. Decety and Lamm (2006) also built a social neuroscience model of empathy, which argues that empathy involves both emotion sharing (bottom-up information processing) and the regulation and modulation to this experience (top-down information processing); and the feeling of empathic concern is the outcome of the experience being regulated.

Empathy enables teachers to feel and understand what the child is feeling, communicate about the feeling with him/her, and then responding in ways that meet his/her needs. In preschool settings, each aspect of empathy plays an important role in guiding interactions between the teacher and child. For example, when encountering a crying child, a preschool teacher may have multiple empathy reactions: she/he may feel anxious and upset and may urge the child to stop crying (the feeling of personal distress), she/he may feel warm and soft-hearted and want to help the child (the feeling of empathic concern), or she/he may also consider cognitively what has happened to the child (perspective taking). In addition, the preschool teacher’s responding with the child may also be automatically affected by the fictional characters in books or films that she/he have encountered.

As early education settings are highly emotionally demanding, empathy is usually regarded as a trait and a skill necessary for the teaching and caring for children, and teacher empathy emerges as a highly desirable trait (Peck et al., 2015). However, what is the role of empathy in teachers’ well-beings and mental health? Previous research concerning the relationship between empathy and mental health is mixed. Both positive and negative relationships were found between empathy and mental health-related issues.

On the one hand, empathy was proved to correlated with individuals’ better social relationship (Coutinho et al., 2014), higher well-being (Carnicer and Calderón, 2014), greater professional satisfaction (Halpern, 2003), and emotional self-efficacy (Goroshit and Hen, 2014), and all these issues have a close relationship with one’s mental health (Wacker and Dziobek, 2018). There are also evidence that empathy is a protective factor of burnout in physicians (Lamothe et al., 2014; Thirioux et al., 2016). Empathy can also improve one’s emotional state, as empathy is proposed to be a method of interpersonal emotion regulation, which have proved to be healthy and efficient (Zaki, 2019). On the other hand, researchers have also noticed the relationship between empathy and mental health symptoms. For example, empathy was found to be positively related with depression (Zhang et al., 2013) and anxiety (Fulton, 2012; Gambin and Sharp, 2018). The risk role of empathy in mental health was especially obvious in helping professionals; for example, empathy was found to have a close relationship with burnout and secondary trauma in caregivers (Kim, 2017), physicians (Tone and Tully, 2014), social workers (Macritchie and Leibowitz, 2010), and so on.

The mixed results may stem from the complication of both empathy and mental health. Although the multicomponent assumption of empathy has been wildly accepted, most of the research mentioned above does not especially examine how different dimensions of empathy correlated with mental health-related issues. Further investigation of the studies mentioned above implied that most of the research supporting the protective effect of empathy in mental health is biased to the cognitive aspect of empathy (Dekel et al., 2017), whereas research found that a risk role of empathy in mental health problems is focused on the emotional aspect of empathy (Williams, 1989; Dekel et al., 2017). Accordingly, we hypothesize that emotional empathy is prone to mental health problems whereas cognitive empathy plays a protective role in the among this research.

In this research, we will examine this complex relationship between empathy and mental health. Moreover, we also want to examine this relationship in preschool teaching profession, which is a typical helping profession (Erera, 1997) and a kind of emotional labor (Philipp and Schüpbach, 2010).

Teaching experience is a critical variable that should not be neglected when considering the mental health and empathy of teachers. Researchers have found that teaching experience, not age, correlates with variables related to empathy and mental health in educational settings. Researchers have found a relationship between working experience and mental health in both educational and non-educational contexts. For example, in physicians, professional experience negatively related with mental health (Yekta et al., 2014; Dai et al., 2015); in teachers, teaching experience contributed to the negative emotions concerning their professional identities (Mevarech and Maskit, 2015). Experience is also found to be related with empathy. For example, in psychiatrists, subjects with more years of experience had lower empathic concern scores (Santamaría-García et al., 2017); in preservice preschool teachers, years of study is also found to correlate with various dimensions of empathy (Huang et al., 2018).

Katz (1972) has emphasized that teaching experience is one of the core variables of preschool teachers’ characters. Based on Katz’s theoretical frame, the growth of preschool teachers generally proceeds in four stages: survival, consolidation, renewal, and maturity. It is theoretically assumed that in different developmental stages, teachers have different psychological characteristics; however, from the empirical perspective, to date, we do not know how teaching experience affects preschool teachers’ mental health and empathy across Katz’s four stages. We hypothesize that in different stages of teacher development, empathy plays different roles in preschool teachers’ mental health.

The original sample consisted of 4343 preschool teachers, as 42 respondents’ data were deleted from the analyses because of incomplete information, and the final sample consisted of 4301 preschool teachers from 544 preschools. These preschools were distributed throughout the 19 administrative districts in Beijing City and its surroundings. All the preschools were public, and in 2015, when the data for the present study were collected, there were 913 public preschools and approximately 21,000 preschool teachers in the Beijing area (Beijing Municipal Government, 2017); thus, approximately 51.58% of the preschools and 20.48% of preschool teachers participated in the present study. The participating preschools were randomly selected. According to the Chinese Ministry of Education (2012), all public preschools in China can be classified into four categories—model preschools, first-class preschools, second-class preschools, and third-class preschools—according to more than 10 criteria, for example, the preschools’ site size and hardware level, numbers of students and teaching staff, and the quality of courses and activities. The local educational authorities conduct comprehensive assessments of each preschool according to these criteria every 3 years. In the present study, 11% of the preschool teachers were from the model preschools, 22% were from the first-class preschools, and 31% were from the second-class preschools. The teachers in each preschool were selected randomly and asked to complete the questionnaires through mobile phones or computers.

The age range of the final sample was 18.10 to 59.06 years, with a mean age of 30.07 years (SD = 8.44), and the participants’ teaching experience ranged from 0.12 to 44.00 years (M = 8.11; SD = 8.10). To compare the teaching experience of the preschool teachers, we divided the sample into four groups according to Katz’s (1972) theory concerning preschool teachers’ career development: less than 2 years of teaching experience (n = 1237), 2–5 years of teaching experience (n = 1138), 5–10 years of teaching experience (n = 638), and more than 10 years of teaching experience (n = 1294). The detailed demographic characteristics of the respondents are illustrated in Table 1. In both the whole sample and each teaching experience group, most of the participants were female (93–98.4%) and ethnic minorities (94.1–94.9). The other demographic information of the participants is presented in Table 1.

The data were collected in Spring 2016, and this research was a part of a larger cross-sectional study examining mental health and well-being in teachers. All questionnaires consisted of 200 questions and required approximately 40 min to complete. The questionnaires were compiled using an electronic program and released through the Beijing Preschool Education Comprehensive Service Platform, which was operated by the government, and the main users of this platform are preschool teachers and administrators. The electronic questionnaires were distributed to preschools and were answered voluntarily and anonymously by the preschool teachers. And in the introduction of this questionnaire, the participants are informed that all the information about the participants will be kept confidential and only be used for research purposes. To attend the survey, the only personal identity information that the participants are required to offer is their telephone number. The participants were asked to respond to the questionnaires through mobile phones or computers online. After completing the questionnaire, the participants drew prizes from 5 to 100 yuan, with a winning rate of 80%. In addition, before the survey, the participants do not know the reward, and after all the questions were completed, the web program automatically went to a lottery page, and then the participants have the opportunity to draw.

The participants were told that these questionnaires were used to understand their feelings and common problems at work, and they were not aware of the intention of the present study. This study was approved by the Ethics Committee of the School of Preschool Education, Capital Normal University.

According to Table 2, the sample means for each subscale of the SCL-90-C were between 1 and 2, indicating that the preschool teachers exhibited mental health symptoms at levels between “a little bit” and “moderately.”

To profile the teachers’ performance on SCL-90-C, a 9 (types of mental health symptoms) × 4 (teaching experience groups) repeated measures MANOVA was performed. Among the variables, the category of mental health symptoms was within-subject variables, and the participants’ teaching experience group was the between-subject variable. The results are shown in Figure 1. The sphericity assumption for repeated measures MANOVA was violated: Mauchly’s W₍₃₅₎ = 0.341, χ²₍₂₎ = 4620.12, p < 0.001. Because violating this assumption inflates the Type I error rate, we used the Huynh–Feldt epsilon to adjust the degrees of freedom and provide a more accurate Type I error rate (Stevens, 2002). The results indicate that there are significant interactive effects between all the nine mental health symptoms and the four teaching experience groups. The interactive effect between the symptoms and teaching experience was significant, F(24,4297) = 23.13, p < 0.001, ηp2=0.016, d = 1.0. Next, two sets of simple-effect analyses were conducted to analyze the interactive effect from two perspectives.

In the first simple-effect analysis, nine ANOVAs were used to determine the changing characteristics of the nine mental health symptoms across the four teaching experience groups. The results indicated that for all the nine mental health symptoms, there were significant differences between each experience groups, and with the teaching experience increased, all the nine symptoms increased (see Table 2).

In the second simple-effect analysis, we split the data into four teaching experience groups and conducted repeated measures across the nine symptoms in each group. The results suggest that in each of the four teaching experience groups, obsessive-compulsiveness behavior, interpersonal sensitivity, somatization, and depression are the four most serious symptoms. Although in each teaching experience group obsessive-compulsiveness behavior is the most serious symptom, there were some slight differences for the other symptoms: in the group with less than 2 years of teaching experience, somatization was the second most serious symptom, and interpersonal sensitivity was the third most serious problem; in the groups with 2–5 and 5–10 years of teaching experience, the severity of somatization increased and ranked as the second most serious symptom for preschool teachers, and in this teaching experience group, depression was as serious as interpersonal sensitivity and ranked as the third most serious symptom. In the two groups with more than 5 years of teaching experience, the severity of depression outweighed the severity of interpersonal sensitivity and was the third most serious mental health symptom.

In short, the teachers in the present study exhibited mental health symptoms at levels between “a little bit” and “moderately,” and they scored higher than the Chinese norm for all the nine mental health symptoms. The most serious symptoms in preschool teachers are obsessive-compulsiveness behavior, interpersonal sensitivity, somatization, and depression. The severity of all these symptoms increased with experience, and it seems that somatization and depression’s severity grew faster.

A 4 (dimensions of empathy) × 4 (teaching experience groups) repeated measures MANOVA was performed to profile the teachers’ empathy, and among the variables, the empathy dimensions were within-subject variables, and the participants’ teaching experience group was the between-subject variable. The sphericity assumption for repeated measures MANOVA was violated: Mauchly’s W₍₅₎ = 0.585, χ²₍₂₎ = 2300.11, p < 0.001. The results indicate that there are both significant main effects of empathy dimension and interactive effects between dimensions of empathy and teaching experience, for empathy dimension, F(3,4297) = 3205.34, p < 0.001, ηp2=0.427, d = 1.0; and for interactive effect, F(9,4296) = 15.865, p < 0.001, ηp2=0.011, d = 1.0.

In the following simple-effect analyses, four ANOVAs (see Table 2) were conducted to compare levels of empathy across the four teaching experience groups, and the results suggested that the four dimensions of empathy showed different changing trends across the four teaching experience periods. For fantasy, no difference was found between any of the teaching experience groups, but for empathic concern, the first (mean difference = 0.07, p < 0.01) and third teaching experience groups (mean difference = 0.08, p < 0.01) had significantly higher scores than the fourth group. For perspective taking, the first teaching experience group had higher scores than the second (mean difference = 0.08, p < 0.01) and fourth teaching experience groups (mean difference = 0.11, p < 0.01). For personal distress, the first teaching experience group had significantly lower scores than the other three groups (mean differences = −0.12, −0.12, and −0.16, ps < 0.01). It seems that in the preschool teachers, empathic concern and perspective taking showed a general decreasing trend, whereas personal distress showed an increasing trend.

Pearson correlations were conducted to explore the relationship between dimensions of empathy and mental health symptoms (see Table 3). On the basis of correlation analyses, two sets of hierarchical regression analyses were conducted to determine whether empathy predicts mental health symptoms in preschool teachers. The first set of regression analyses used the nine mental health symptoms as dependent variables (see Table 4). For these hierarchical regressions, the demographic variables were entered in Step 1, the length of teaching experience was entered in Step 2, and the four variables of empathy were entered in Step 3. For all nine hierarchical regressions, the models with the set of population measures were significant, Fs ≥ 19.33, ps < 0.001, and R²s ≥ 0.019, and we found that age significantly predicted all nine mental health symptoms. When the variable for preschool teaching experience was entered in the second step, with the exception of somatization, age did not account for any additional variance in the models, 0.02 ≤ ΔR² ≤ 0.09, 0.018, F (change) ≥ 16.75, p < 0.001. In addition, in this step, teaching experience positively predicted all mental health symptoms, even after the effect of age was controlled. In the third set of models, the four empathy dimensions were entered and showed different predictive patterns for the nine mental health symptoms. On the one hand, both fantasy and personal distress positively predicted the nine mental health symptoms; thus, it seems that the more an individual is prone to fantasy or personal distress, the more susceptible he or she is to these mental health symptoms. On the other hand, empathic concern and perspective taking negatively predicted the nine mental health symptoms; thus, it appears that these two components are protective factors for mental symptoms.

To explore the changing pattern of empathy in mental health symptoms based on teaching experience, hierarchical regressions were conducted in each teaching experience group, with the nine mental health symptoms as the dependent variables (see Table 5). In each hierarchical regression, the demographic variables were entered in the first step, and the four dimensions of empathy were entered in the second step. The results revealed that the predictions of the four dimensions of empathy were different across the four teaching experience groups. As all first models were significant (all Fs > 29.01, ps < 0.001) and to avoid an unwieldy table, we present only the second set of models in Table 5. The results revealed that fantasy and personal distress positively predicted all mental health symptoms in each teaching experience group; however, the roles of empathic concern and perspective taking were different across the four teaching experience groups. Although empathic concern negatively predicted anxiety, phobic, anxiety, paranoid ideation, and psychoticism in all the four teaching experience groups, it selectively negatively predicted somatization, obsessive-compulsiveness behavior, interpersonal sensitivity, depression, and hostility only in participants with more than 5 years teaching experience (in the third or four teaching experience group). In contrast, although perspective taking also negatively predicted the similar symptoms with empathic concern in all teaching experience groups, it only negatively predicted somatization, obsessive-compulsiveness behavior, and phobic anxiety for the participants with teaching experience less than 5 years (in the first and second teaching experience groups). That is to say, empathic concern plays a protective role against mental problems mainly in new preschool teachers, whereas empathic concern plays a protective role mainly in experienced preschool teachers.

By examining mental health in a sample covering a wide range of teaching experiences, the present study reveals the general condition and dynamic aspect of Chinese preschool teachers’ mental health.

First, the results indicate that as teachers’ experience increases, preschool teachers exhibit a trend toward increases in mental health symptoms. This result was supported by a meta-analysis, which indicated that during the period of 1998–2013, Chinese teachers’ mental health becomes worse (Yang et al., 2019). However, despite its extreme importance, the mental status of teachers is still worrying.

In addition, the characteristics of the nine symptoms suggested that for each of the four teaching experience groups, obsessive-compulsiveness behavior, somatization, depression, and interpersonal sensitivity were the most serious symptoms, with the severity of somatization and depression increasing faster. Among these symptoms, obsessive-compulsiveness behavior is the most prominent mental health problems for preschool teachers with various teaching experience. It seems that obsessive-compulsiveness behavior is a typical mental problem in teaching professionals, as obsessive-compulsiveness behavior was also found as the most serious mental health problem among Chinese primary and high school teachers (Zeng and Li, 2011). Interpersonal sensitivity is also a typical mental problem especially for the new preschool teachers. Teaching, by nature, is a complex interpersonal interaction, and new teachers have to rebuild their interpersonal relationships, which may contribute to the preschool teachers’ interpersonal sensitivity. In addition, the increase in severity of somatization and depression with teaching experience may reflect the process of emotional exhaustion and job burnout, as these two indexes are sensitive to stress and emotional labor (Seery and Corrigall, 2009), which are both typical characters in preschool education (Miller et al., 2019).

The present study found that among preschool teachers, the four dimensions of empathy showed different patterns as teaching experience increased: fantasy remained stable, empathic concern and perspective taking showed a slow decreasing trend, and personal distress showed an increasing trend.

According to a research on student preschool teacher (Huang et al., 2018), the mechanisms underlying the changing pattern of empathy are dehumanization and detachment, which are indexes of job burnout (Haslam and Loughnan, 2014). Preschool teaching typically involves emotional labor, of which emotional engagement is a core part (Philipp and Schüpbach, 2010). Exposure to children’s emotions for an extended period could decrease preschool teachers’ sensitivity to emotional cues, and empathy, if it is too strong or lasts too long, could increase the risk of emotional exhaustion (Wróbel, 2013; Wacker and Dziobek, 2018). Accordingly, the changes observed in the preschool teachers may reflect a protective mechanism.

The present study found a relatively complex relationship between empathy and mental health in preschool teachers. These results only partially supported our first hypothesis. This study indicated that empathy dimensions correlate with mental health differently; however, this study also suggested that fantasy and personal distress are risk factors to preschool teachers’ mental health, whereas empathic concern and perspective taking play protective roles in preschool teachers’ mental health. In addition, the second hypothesis was supported as the result indicated that the relationships between dimensions of empathy and mental health symptoms were different across the four teaching experience groups.

The present research also indicates a dynamic relationship between empathic concern, perspective taking, and preschool teachers mental health status across different teaching experience groups. A possible explanation to this result is the psychological nature of teaching experience.

According to Katz’s (1972) theoretical framework, preschool teachers’ development contains four stages: survival, consolidation, renewal, and maturity stage. The most prominent feature of the earlier career stage, such as the survival and consolidation stages, is that teachers cannot break through the self-centeredness and mainly focus on their own survival and adaptation. Preschool teachers have to use the cognitive tool, for example, perspective taking, to understand the feeling and thoughts of others and consequently to adopt the working task and personal relationship. With the increase of teaching experience, preschool teachers enter the stage of renewal and maturity stages; and empathic concern, which is automatic and affective in nature, may help the teachers feel and understand others’ emotions better. From this perspective, this research highlights the role of teaching experience in an emotional context.