Emotions, Illness Symptoms, and Job Satisfaction among Kindergarten Teachers: The Mediating Role of Emotional Exhaustion

: Kindergarten teachers’ emotions are an essential factor in their physical and psychological wellbeing. Previous studies mainly focused on the relationship between kindergarten teachers’ emotions and their students’ emotions while ignoring the important relationships between kinder-garten teachers’ emotions and their own wellbeing (e.g., teachers’ health, job satisfaction, burnout). Therefore, this study explores teacher emotions as predictor variables, illness symptoms, and job satisfaction as criterion variables, and emotional exhaustion as a mediator. In total, 1997 kindergarten teachers completed the Teacher’s Emotion Scale, the Occupational Emotional Exhaustion Scale, the Illness Symptoms Scale, and the Job Satisfaction Scale. Results revealed that enjoyment negatively predicted illness symptoms and positively predicted job satisfaction via the mediating role of emotional exhaustion. The opposite relationships were found with anger, also conﬁrming the mediating role of emotional exhaustion. Anxiety positively predicted illness symptoms, completely mediated by emotional exhaustion, but no relationship was found with job satisfaction. The function of emotions in teachers’ physical and mental health, implications for kindergartens’ research and practice, and suggestions for future research are discussed.


Introduction
Teacher emotions constitute an important part of the teaching and learning process [1].Appraisal theory provides a meaningful perspective to understand teachers' emotions [2][3][4].Appraisal theory proposes that emotions result from individuals' interpretation and cognitive judgement of their circumstances rather than the circumstances themselves [5][6][7][8].Previous studies applied appraisal theory in exploring the impact of teachers' emotions on students' performance [9,10].Frenzel et al. [11,12] showed that teachers' positive appraisals in the classroom are essential in achieving good student performance, motivation, and discipline, and forming positive teacher-student relationships.Similarly, previous research confirmed that teachers' emotions are related to students' emotions, emotional competence, motivation, academic performance, classroom discipline, and social behavior [13].For example, Wang et al. (2019) found that teachers who expressed their true feelings were more committed to their work and reported better physical health [14].Teachers' emotions also affect cognitive performance, health, job satisfaction, burnout, motivation, and teaching approaches [10,[15][16][17].An Australian study showed that teachers generally have high emotional needs in terms of work, burnout, and job satisfaction [18].Emotional burnout, a core component of job burnout, has a significant predictive effect on job satisfaction and health [19,20], and this appears to be related to teachers' emotions as well.For example, teachers with more positive emotions tend to report lower burnout and higher job satisfaction [21].Because of the heavy workload [22] and relatively high turnover rate, kindergarten teachers [23] are regarded to be a more vulnerable group compared to other teachers.Kindergarten teachers' fatigue, job burnout, and poor physical condition in response to the heavy workload can threaten their teaching quality.Therefore, the question of how to maintain kindergarten teachers' job satisfaction and physical health has drawn increasing attention from researchers [14,24].

Teachers' Emotions
According to a multicomponent approach, emotions can be defined as a synchronous and coherent pattern of central nervous and peripheral physiological responses.These responses are reflected in movement tendencies and facial, vocal, and gestural expressions in one's subjective experience [25,26].More specifically, teachers' emotions are discrete emotions that teachers experience in their professional lives, particularly while providing instruction [27].The present study adopts this multicomponent definition of emotions, viewing kindergarten teachers' emotions as subjective experiences.According to the dimensional approach to emotions, emotions are defined on the basis of where they lie on two or more continuous dimensions, such as valence and arousal [28], positive and negative activation [29], or approach-avoidance [30].Others argue that emotions are discrete and should, therefore, be considered separately as unique experiential states [31].Previous research on emotions and learning took a dimensional approach that effectively treats all discrete emotions (e.g., anger, sadness, joy) as globally negative or positive affective states [32].However, as Gooty et al. [33] indicated, when discrete emotions are combined into overall negative or positive dimensions for analytical purposes, the different antecedents, functions, and outcomes of different emotions are not considered [34].Therefore, the present study considers teachers' enjoyment, anger, and anxiety.These discrete emotions were selected because enjoyment and anger outrank other discrete emotions in terms of their frequency.In addition, anxiety was included because it has received considerable research interest in the teaching context due to its high relevance to physical health and psychological wellbeing [35].Thus, this research focuses on three discrete teacher emotions: enjoyment, anger, and anxiety.
A variety of stimuli may trigger emotions experienced by teachers in the classroom.For example, teachers experience positive emotions such as joy, enjoyment, and pleasure when students progress or are responsive and cooperative.On the other hand, teachers' anger and frustration are most often related to students' misbehavior or lack of classroom discipline [10].Many studies empirically confirmed the contribution of positive and negative affect in favoring job perceptions and attitudes [36,37].The teaching profession is one that requires the most emotional input, and this is especially true for kindergarten teachers [38][39][40].Because kindergarten teachers deal with children who are not physically and mentally mature, they need to exhibit patience and enthusiasm.This emotional labor can result in relatively high levels of burnout [41].Research also indicated that long-term stress and negative emotions contribute to teachers' lower job satisfaction and decreases the stability of teaching teams, as about 30% of teachers leave the profession within five years [42,43].Teachers who do not resign are more likely to feel anxiety and depression symptoms [44].These negative emotions continue to consume emotional resources, adding to the emotional exhaustion of kindergarten teachers [31].Negative emotions further affect teachers' evaluations of their job satisfaction, resulting in a more negative overall assessment of the work [45].Therefore, to promote high-quality teaching in kindergarten, it is necessary to pay more attention to the development of key emotions in kindergarten teachers.

Illness Symptoms
Previous research on kindergarten teachers' illness symptoms focused on hearingrelated symptoms [46,47], as kindergarten teachers experience high levels of noise in their occupational environment.However, the impact of the heavy workload and excessive professional pressure cannot be underestimated [48,49].A study comparing kindergarten teachers with elementary-and middle-school teachers found that pressure on kindergarten teachers is more pronounced than that of other teachers [50].Working as a kindergarten teacher involves physically demanding behaviors such as straining one's head down, kneeling, stooping, squatting, and bending.In addition, excessive professional pressure can lead to the deterioration of teachers' health [51].Research found that kindergarten teachers are more prone to occupational illnesses such as laryngitis, cervical spondylosis, and lower-back pain [52,53].Previous research also demonstrated that teacher burnout affects teachers' health [54].Hence, we conducted this study to focus on the illness symptoms of kindergarten teachers.

Job Satisfaction
Job satisfaction refers to one's attitude towards one's working environment and other job-related circumstances [55], which is fundamental to evaluating the stability of a team [56,57].Teachers' job satisfaction denotes teachers' emotional relations with their job and perception of the outcomes of their teaching tasks [58].Intrinsic sources of job satisfaction for teachers include making a positive difference in others' lives and feelings of belonging [59], and the positive emotion of enjoyment.On the other hand, extrinsic sources of job satisfaction include salary, status, school support [60,61], and school administrators' leadership styles [62].
We applied the broaden-and-build theory of positive emotions to explore the relationship between kindergarten teachers' emotions and job satisfaction.Broaden-and-build theory describes positive emotions as "broadened thought-action repertoires" that build "enduring personal resources" [63,64].According to this theory, when teachers are in a positive emotional state, they broaden their thinking and behavior patterns, thereby expanding and building upon their personal resources.The construction and accumulation of personal resources can promote teachers' job satisfaction [65,66].For example, a research study conducted in Turkey with 581 teachers demonstrated that positive emotions were correlated with higher levels of job satisfaction [67].Other studies demonstrated that engaging in activities in classroom settings and establishing positive teacher-student relationships enhance teachers' wellbeing at work [68,69].However, most kindergarten teachers reported low job satisfaction caused by lower salaries and excessive work pressure compared to other professions [70,71].Thus, it is essential to investigate the factors that may mediate the relationship between kindergarten teachers' emotions and their job satisfaction.

Emotional Exhaustion
Teacher burnout is traditionally conceptualized as resulting from long-term occupational stress [72] and the unpleasant and negative emotions that result from various aspects of the teaching experience [19].Emotional exhaustion is the core component and the most visible symptom of work-related burnout [73].It is evidenced in extreme fatigue or loss of vitality resulting from the excessive depletion of emotional resources, lack of enthusiasm towards work, and lack of interest in external stimuli [74].Teaching is ranked as one of the most stressful professions in various cultural and educational contexts [75].Emotional exhaustion significantly negatively impacts both teachers' wellbeing in terms of their selfrated physical and mental health [21], job satisfaction [59], and student performance [76].Emotional exhaustion is also related to high absenteeism, retirement, and turnover [40,77].
According to conservation-of-resources theory (COR), emotional exhaustion in teachers may mediate the relationship between emotions and job satisfaction and health conditions [78].This theory emphasizes the objective elements of threat and loss, and the process of protecting this loss of resources.It proposes that emotional exhaustion results from a sense that one cannot deal with an impending loss of resources through one's own coping efforts [79].For example, kindergarten teachers consume emotional resources in work, and to maintain physical and mental health they must actively construct and protect their existing emotional resources [80].In line with COR, emotional exhaustion or the loss of these emotional resources is associated with job attrition.Previous studies indicated that kindergarten teachers with higher levels of stress and emotional exhaustion expressed a greater intention to leave the job [81].As kindergarten teachers often need to exaggerate their positive emotions and suppress negative emotions to help children understand specific topics or attract their attention, these actions can lead to emotional exhaustion [82].In addition, COR predicts that individuals must invest resources to prevent the loss of these resources, but this is increasingly difficult for teachers.In the past few decades, education researchers from multiple countries reported an increase in work tasks in the teaching profession, and teachers are required to complete more tasks in a shorter timeframe.These demands greatly shorten teachers' leisure time, impeding teachers from recovering physically and mentally [77,83].Teachers with high levels of emotional exhaustion are more likely to show low levels of job satisfaction [84].On the basis of the COR view, negative emotions such as anxiety could consume emotional resources, leading to greater emotional exhaustion, and therefore lower job satisfaction and heath.On the other hand, the enjoyment that teachers experience in teaching can provide them with more emotional resources to cope with emotional exhaustion, thereby positively contributing to job satisfaction and physical health [85].

Literature Review
Two databases were included in the literature search (PsycINFO and Web of Science), which was carried out using search terms "kindergarten teacher" or "preschool teachers", "job satisfaction", "illness symptoms", "physical health", and "emotion".Research results are summarized in Table 1.Articles were selected on the basis of two inclusion criteria: (1) Studies were required to examine empirical relationships, thereby excluding qualitative or theoretical articles.(2) Job satisfaction and illness symptoms needed to be outcome variables in the studies.According to the current literature review, emotional regulation ability [21], emotional exhaustion [59], emotional labor [86,87], social support [86,88], workplaces [88], teachers' perceived work climate [89], turnover intentions [90], and stress [91] were examined in prior empirical studies as a potential predictors of teachers' job satisfaction.Three articles explored the relationship between teachers' emotions and job satisfaction.Of these three articles, two adopted the classification of teachers' emotion as positive versus negative emotion [67,91], and one article used teachers' enthusiasm (enjoyment, excitement, and pleasure) as a predictor [36].These studies revealed the positive relationship between positive emotions and job satisfaction.However, Atmaca et al. [36] did not find a significant relationship between negative emotions (sadness, anger, and fear) and job satisfaction, but found that negative emotions were related to job burnout.None of the above studies included anxiety as a discrete emotion.
For studies examining the antecedents of illness symptoms, stress and burnout were often considered by researchers as proximal contributors [20,92,93].Studies using teachers' emotional labor as a predictor found significant association between emotional labor and physical health [14,94].Findings also showed a relationship between teachers' negative emotions and illness symptoms [95], and that anxiety disorders negatively affected their physical health [35].In the current literature review, only 6 of the 17 articles included preschool teachers as the participants.Thus, kindergarten teachers' emotions were somewhat neglected in previous studies.As previous research suggests that kindergarten teachers experience burnout and emotional exhaustion to a greater extent than teachers in other grades [50], it is necessary to include kindergarten teachers in research on this topic, as we do in the current study.A second gap in the research that was identified in the present literature review was the need for a more integrated model-based approach to these variables that are often independently studied.In order to expand knowledge, instead of looking at the pairwise relationships among teacher emotions, illness, job satisfaction, and emotional exhaustion, the present study concurrently examines their complex associations.These two features of the present study (i.e., sample and analyses) contribute to its significance.With respect to the role of emotional exhaustion, according to appraisal theory [5][6][7][8], individuals' judgments of events trigger emotions, while conservation-of-resources theory [79] suggests that, once individuals use up their resources, they are motivated to acquire new resources.However, if resources are not effectively replenished, emotional exhaustion is experienced.Therefore, this study explored the relationship between kindergarten teachers' emotions and both their illness symptoms and job satisfaction.Further, this study investigates the mediating role of emotional exhaustion.

Research Hypothesis
We hypothesized the following: Hypothesis 1.The direct effect of kindergarten teachers' emotion on illness symptoms and job satisfaction is significant.Enjoyment negatively predicts illness symptoms and positively predicts job satisfaction, while anger and anxiety positively predict illness symptoms and negatively predict job satisfaction.Hypothesis 2. Emotional exhaustion mediates the relationship between discrete emotions and illness symptoms and job satisfaction.

Ethical Statement
This study was approved by the research ethics committee of our university, and all procedures complied with the ethical standards of the latest version of the Helsinki Declaration.Participants had been informed about the nature of the study before they took part in the survey, including the aim of the study, and their anonymity was maintained in the survey and data analysis.All participants willingly gave their informed consent to participate after being informed about the purpose of the study.All analyses were based on anonymous data.

Participants and Data Collection
Data were collected via a Chinese online research panel, Wenjuanxing (https://www.wjx.cn/, 7 June 2021), which is functionally equivalent to Amazon Mechanical Turk.This method has the advantages of Internet-based survey methods (e.g., increased simplicity in gathering data, anonymity of respondents, voluntary participation).A total of 1997 kindergarten teachers voluntarily participated in the study.Each participant was informed about the purpose of the study before the program commenced.Table 2 provides the demographics of the kindergarten teachers in the sample.There were 61 male teachers and 1936 female teachers, ranging from 15 to 60 years of age (M = 31.41,SD = 8.77).A total of 1485 worked in urban areas, and 512 worked in rural areas.Regarding the teachers' establishment, 426 were tenured, and 1571 were not tenured.A total of 1376 individuals worked in public kindergartens, and 621 participants were in private kindergartens.Regarding years of teaching, 870 had worked for 0-5 years, 442 had worked for 6-10 years, 225 had worked for 11-15 years, 162 had worked for 16-20 years, 95 had worked for 21-25 years, and 203 participants had worked for more than 25 years.Teacher emotions were assessed by the Teacher Emotions Scales [27].We used the group-specific scale of enjoyment (e.g., "I enjoy teaching these students"), anger (e.g., "Sometimes I get really mad at these students"), and anxiety (e.g., "I feel tense and nervous while teaching these students").Each emotion included four items, and a 4-point scale was used ranging from 1 (completely disagree) to 4 (completely agree).The Chinese translation of this scale followed a multiple-stage translation process, including an independent backand-forth translation by two professional translators, comparisons, and revisions by two psychology experts, and a pretest with Chinese teachers.The internal consistencies were Cronbach's αs = 0.84, 0.75, and 0.81 for enjoyment, anger, and anxiety, respectively.Fit statistics of the confirmatory factor analysis were χ 2 /df = 5.878, CFI = 0.981, TLI = 0.972, RMSEA = 0.049, SRMR = 0.030.For enjoyment, anger, and anxiety, AVE were 0.578, 0.424, 0.520, CR were 0.845, 0.736, 0.810, respectively.

Data Analysis
Data were analyzed in three steps.First, we calculated descriptive statistics and the Pearson correlation coefficients.Second, using structural equation modelling (SEM), fit indicators of three structural models and the χ 2 difference test of models were calculated.As χ 2 tends to be significant when calculated on data from large samples, we additionally used ∆CFI for the evaluation of nested measurement models.Lastly, we used 5000 bootstrap samples and the 95% bias-corrected confidence interval (95% CI) to examine the significance of the mediation effect [97].The statistical significance level was set at p < 0.05.All analyses were conducted using SPSS Version 25.0 and AMOS 26.0.The quality of model fit was assessed by the χ 2 /df, comparative fit index (CFI), Tucker-Lewis index (TLI), root-meansquare error of approximation (RMSEA), and standardized root-mean residual (SRMR).A good fitting model is indicated by χ 2 /df < 5, RMSEA values of < 0.08, SRMR values < 0.06, CFI and TLI, values > 0.95 and we can accept χ 2 /df < 8 [98,99].

Common Method Biases
Harman's single-factor test was used to diagnose common method bias [100].The results of principal component factor analysis without rotation showed that there were 5 factors with eigenvalues greater than 1.The variance explained by the first factor was 33.267%, below the threshold of 40%.Therefore, common method bias did not affect the outcome of this study.The Durbin-Watson (D-W) test was used for diagnosis of autocorrelation.All D-W values were between 1.959 and 2.078 (ps > 0.05).Results of multicollinearity tests were acceptable for enjoyment (tolerance = 0.655, VIF = 1.526), anxiety (tolerance = 0.497, VIF = 2.014) and anger (tolerance = 0.418, VIF = 2.393).

Descriptive Statistics and Correlation Analysis
The descriptive statistics and correlation matrix of all variables are shown in Table 3. Teacher emotions, emotional exhaustion, and job satisfaction data were normally distributed (skewness and kurtosis within ±1).Illness symptoms showed a slight positive skew and the distribution was slightly leptokurtic.Enjoyment was significantly negatively correlated with emotional exhaustion and illness symptoms, and significantly positively correlated with job satisfaction, while anxiety and anger were significantly positively correlated with emotional exhaustion and illness symptoms, and significantly negatively correlated with job satisfaction.

The Relationships between Teacher Emotions, Illness Symptoms, and Job Satisfaction
The proposed relations between the variables were evaluated using SEM in AMOS 26.0.Three models were estimated: (1) a model containing only direct paths, (2) a full mediation model, and (3) a model containing both direct and indirect paths.The tested models are shown in Figure 1, which illustrates the hypothesized relations among these variables.
The model fit indicators of these specified structural models and the results of the χ 2 difference tests are presented in Table 4, and all path coefficients are presented in Table 5.In Model 1, χ 2 /df = 2.35, CFI = 0.996, TLI = 0.989, RMSEA = 0.026, SRMR = 0.018, teacher emotions were independent variables, illness symptoms and job satisfaction were used as dependent variables, and gender, working location, age, and kindergarten type (i.e., public or private kindergarten) were used as control variables.Hypothesis 1 was supported in that enjoyment was negatively related to illness symptoms (β = −0.15,p < 0.001) and positively related to job satisfaction (β = 1.05, p < 0.001).The second hypothesis was partially confirmed.Anxiety (β = 0.22, p < 0.001) and anger (β = 0.24, p < 0.001) were positively related to illness symptoms.In addition, anger (β = −0.56,p < 0.001) was negatively related to work satisfaction, and anxiety (β = −0.05,p > 0.05) was not related to job satisfaction.On the basis of results of χ 2 difference tests, ∆χ 2 (4) = 10, p < 0.001, Model 3 was better than Model 1.
Emotional exhaustion (β = 0.33, p < 0.001) positively predicted illness symptoms, while the coefficients of enjoyment (β = 0.03), anxiety (β = 0.06), anger (β = −0.01)were not significant (ps > 0.05).In addition, the results of the bootstrap analysis showed significant indirect effects (see Table 6), indicating that emotional exhaustion plays a full mediating role between the three emotions (enjoyment, anxiety, and anger) and illness symptoms.The coefficient for the relationship between emotional exhaustion (β = −0.32,p < 0.001) and job satisfaction was significant, and the coefficients of enjoyment (β = 0.88, p < 0.001) and anger (β = −0.30,p < 0.001) were still significant, while anxiety was not significant.Results of bootstrap analysis showed significant indirect effects (see Table 6), indicating that emotional exhaustion plays a partial mediating role between enjoyment, anger, and job satisfaction, accounting for 16.6%, 44.7% of the total effect.

Discussion
In the present study, a mediating model was tested to examine the relations between kindergarten teachers' emotions and emotional exhaustion, and their illness symptoms and job satisfaction.Results showed that enjoyment and anger predicted illness symptoms and job satisfaction via emotional exhaustion.Anxiety predicted illness symptoms via emotional exhaustion but did not show a significant relationship with job satisfaction.

Direct Effects of Teachers' Emotion
Obtained results are partially consistent with the hypotheses.Teachers' emotions predicted both their job satisfaction and illness symptoms.As expected, enjoyment negatively predicted illness symptoms and positively predicted job satisfaction.This result confirms the proposition of broaden-and-build theory [63].Positive emotions (such as enjoyment) appear to broaden kindergarten teachers' instantaneous cognitive ability and attention, enhance their ability to identify opportunities and adapt to new situations, allowing for them to build enduring personal learning resources [64].These findings are in line with previous research showing that teachers who experience more positive emotions are more enthusiastic and feel more confident when facing obstacles, whereas those who experience more negative emotions report a higher level of job burnout [40], which then negatively affects their health [54].Additionally, past research found that persistent anxiety symptoms lead to undesirable physical and psychological conditions [101].However, in the present study, anxiety was not related to job satisfaction.This may have been due to the overall low anxiety level found in our sample of teachers and the unique nature of anxiety.Keller and colleagues proposed that anxiety, as a trait measure, is inherently vague [102].In the present study, angry and anxious teachers demonstrated a negative interpretation bias, but this bias was more associated with anger than with anxiety.Additionally, anger was generally associated with more significant results than those of anxiety [103].Hence, it is not clear how anxiety influences job satisfaction among kindergarten teachers.A future study is needed to further explore how to distinguish between the negative emotions of anxiety and anger among teachers, and to further clarify the independent function of anxiety.
Teachers' emotions are closely related to illness symptoms.First, the social functions of emotion include fulfilling an individual's needs in adapting to a certain environment [104].From the perspective of appraisal theory, the effects of emotion with different antecedents, functions, and results are also worth discussing.Teachers feel different emotions when overburdened with teaching tasks and when experiencing the challenges of teaching performance evaluations [2,6].Enjoyment can help people in maintaining their physiological health and in promoting positive emotions [105].On the other hand, anxiety and anger can affect teachers' daily lives and work, hinder their development, and pose a major threat to their health [4].Second, teaching is a profession with a high consumption of emotional resources, and teachers' emotional reactions to work can further affect their attitudes toward work [106].This is also in line with Fredrickson's broaden-and-build theory [64], which proposes that positive emotions expand one's attentional scope and thought-action repertoire.
Our results also suggest that teachers' enjoyment helps to improve their job satisfaction, while anger reduces their job satisfaction.These findings are consistent with previous findings showing positive emotions are associated with higher levels of job satisfaction [67], and negative emotion are associated with lower job satisfaction [42,43].These results suggest that as enjoyment, pleasure, and excitement can be naturally projected into the enthusiastic teaching behavior, anger has the opposite effect on performance [36].Therefore, it is important for kindergarten teachers to properly express their emotions, as the effective expression of emotions is beneficial to the job satisfaction and physical health of kindergarten teachers.In regards to the recommended coping strategy for negative emotions, past research suggests that teachers from different regions have different ways of coping with negative emotional states such as depression [107].Kindergarten teachers tend to reflect the emotions that are expected from them, such as happiness, calm, and a willingness to help others, while suppressing their true emotions [108].This is especially true in the context of Chinese culture, as emotion is deemed as a homeostatic learning process where the body, mind, and heart coordinate to regain emotional equilibrium [109].Therefore, kindergarten teachers who are unwilling to communicate openly with colleagues and parents, and instead suppress their emotions, may face severe emotional exhaustion.The accumulated anger, anxiety, and other negative emotions gradually decrease their enthusiasm for work, and eventually lead to the decision to resign [76].

Mediating Role of Emotional Exhaustion
The present study found that emotional exhaustion plays a mediating role in the prediction of illness symptoms by anger, enjoyment, and anxiety, and also contributes to the prediction of job satisfaction by anger and enjoyment.This result confirmed findings from previous work indicating emotional exhaustion reflects one component of work-related burnout [73], and that burnout leads to negative effects on teacher's physical condition [80].Similarly, past research had showed that high levels of emotional exhaustion contributes to a lower level of job satisfaction [59,84], highlighting the essential role of emotional exhaustion.Conservation-of-resources theory can be used to explain the mediating role of emotional exhaustion between teachers' emotions and their job satisfaction and illness symptoms.In a stressful state, individuals tend to use existing resources to obtain new resources in order to reduce the net loss; at the same time, individuals actively construct and maintain their current resource reserves to cope with possible future resource loss scenarios [79].Teachers' emotional exhaustion can occur in a high-stress work environment when they deplete their emotional resources [109,110].Our findings suggest that emotional exhaustion mediates the relationship between emotions, and illness symptoms and job satisfaction.Consistent with our expectations, this result highlights the importance of the role of emotional exhaustion and shows that it is negatively associated with job satisfaction [85].
Thus, emotional exhaustion significantly increases the prevalence of illness and harms job satisfaction [111].Many studies indicated a positive correlation between kindergarten teachers' anger and emotional exhaustion [4,112].As mentioned before, kindergarten teachers are required to suppress their anger due to their occupational norms and expectations.Hence, if kindergarten teachers cannot reach the level of deep acting (an emotional labor strategy that adjusts the inner emotional experience and changes external emotional expression), their emotional exhaustion can increase and cause adverse consequences, such as physical symptoms and higher energy consumption [113,114].The regulation of emotional exhaustion can be achieved by specifically targeting and increasing teachers' enjoyment through psychological intervention and reducing their anger [102].Another negative emotion, anxiety, can also cause emotional exhaustion, negatively affecting physical and mental health [115].Current results support the mentioned research and confirm the mediating role of emotional exhaustion between emotions of enjoyment, anxiety, and anger, and illness symptoms.Methods for effectively alleviating and reducing the level of emotional exhaustion received a great deal of attention [116,117] and can thus be applied in this context.The present study indicates that increasing teachers' enjoyment, and reducing teachers' anger and anxiety can effectively reduce their levels of emotional exhaustion, subsequently improving their physical health and job satisfaction.

Limitations and Educational Implications
The present results indicate that kindergarten teachers' enjoyment, anger, and anxiety directly affect their illness symptoms, while enjoyment and anger affect their job satisfaction.Results further highlight the importance of properly managing emotions.For example, teachers can be trained to use effective and adaptive emotion regulation strategies to reduce or hinder the experience of negative emotions.Furthermore, the results of this study clearly show that emotional exhaustion plays a significant mediating role among teachers' emotions, illness symptoms and job satisfaction.According to COR, positive emotions are a protective resource that can reduce teachers' emotional exhaustion.Therefore, enjoyment is beneficial for teachers' physical health and job satisfaction, and may lead to better teaching quality and efficiency.Likewise, protective resources such as culture, social equity, social networks, advice, respect, and support from supervisors can reduce the negative effects of anger and anxiety.Therefore, we also recommend that education administrators and society at large focus more resources and attention on kindergarten teachers.To ensure the psychological and physical health of kindergarten teachers, steps should be taken to improve their income and social status, and to provide professional training opportunities.However, as kindergarten teachers may experience different emotions than primary school teachers do, one must be cautious in generalizing the current findings to teachers who work with older students.
These findings offer further implications for kindergarten teachers' research and practice.Enjoyment emotions can help teachers in developing their teaching performance, increase their job satisfaction, and form good teacher-student relationships [118,119].Interventions that increase individuals' positive emotions such as gratitude interventions [120] could, therefore, strengthen the positive outcomes for teachers.Kindergarten teachers who experience emotional frustration in teaching should efficiently regulate these negative emotions (i.e., anger and anxiety) to maintain effective teaching without depleting their cognitive resources.Second, it is a normal phenomenon for anger and anxiety to be aroused in the teaching process, and employing emotional regulation strategies to lessen the impact of these negative emotions may be beneficial.
There are several limitations in the current study.First, the data used in this study were derived from the self-evaluations of kindergarten teachers.Although their anonymity was maintained, the self-reports of individual teachers could have still been affected by social desirability, entailing responses that met their social expectations.Future research could consider introducing evaluations of kindergarten leaders, parents, and young children to improve the objectivity and diversity of evaluations.Second, this study used a one-time cross-sectional study to explore the relationship among kindergarten teachers' emotions, illness symptoms, and job satisfaction, which impeded determining the causal impact of variables.Future studies can use a longitudinal design to investigate causal relationships among variables.

Conclusions
In this study, we investigated the relationships between kindergarten teachers' emotions, and their illness and job satisfaction, and the mediating role of emotional exhaustion.Results demonstrated that the enjoyment and anger of kindergarten teachers both had significant direct and indirect associations with illness and job satisfaction through emotional exhaustion.Anxiety was positively related to symptoms of illness through emotional exhaustion, although it was not related to job satisfaction.Thus, this research highlighted the different functions of specific emotions of kindergarten teachers.Whereas enjoyment maintains teachers' positive attitudes and physical health, anger and anxiety result in the opposite outcomes.Thus, kindergarten teachers should explore ways to accentuate and strengthen their positive emotions, and to improve their abilities to effectively and appropriately process negative emotions.This helps in reducing the emotional exhaustion of kindergarten teachers, spark their enthusiasm for their educational careers, and improve their physical health.

Figure 1 .
Figure 1.Effect of kindergarten teachers' emotions on illness symptoms and job satisfaction: mediating test of emotional exhaustion.Note: *** p < 0.001; results in brackets are p values of Model 2. M1 containing only direct paths; M2 containing only indirect paths; M3 is a model containing both direct and indirect paths.

Figure 1 .
Figure 1.Effect of kindergarten teachers' emotions on illness symptoms and job satisfaction: mediating test of emotional exhaustion.Note: *** p < 0.001; results in brackets are p values of Model 2. M1 containing only direct paths; M2 containing only indirect paths; M3 is a model containing both direct and indirect paths.

Table 1 .
Summary of study findings.

Table 2 .
Category of kindergarten teachers in the sample.

Table 3 .
Descriptive and correlation analyses of study variables (N = 1997).

Table 4 .
Fit statistics of tested models and results of χ 2 difference test.

Table 5 .
Effect of kindergarten teachers' emotions on illness symptoms and job satisfaction: mediating test of emotional exhaustion.

Table 6 .
Bootstrap analysis of mediating effect.