Chinese Male Kindergarten Teachers: Factors Affecting Job Satisfaction and Organisational Commitment

,


Introduction
In early childhood education (ECE), male teachers often show their masculine traits (Jing, 2016;Martino, 2008) and are considered to play father-like roles (Ahmad et al., 2018;Ho & Lam, 2014), and in so doing, have a similar influence as the father in a family environment (Clark et al., 2009).For instance, Father Reading Every Day (a programme from the USA) researched nine primary schools in the UK, and the results show that children who participated in the project improved their reading, writing and numeracy skills compared with their peers who did not participate in the project (Forrest & Lloyd, 2014).This success might be attributed to the value that father-figures can bring to children's lives (Warin, 2018).Similarly, Warin (2018) points out that these benefits can also be gained in ECE, where male kindergarten teachers (MKTs) can provide alternative approaches to children's learning and activities in what is a largely female-dominated environment.
The imbalance towards females in ECE has been identified (Weaver-Hightower, 2011), with a shortage of male teachers in the US and Asian countries, such as Japan and South Korea (Wu, 2016), and they are also in the minority in China at just 2.22% (Yang, 2015).Policymakers in the UK (Lewis et al., 2022;Thornton & Bricheno, 2006) and other Western countries such as New Zealand (Livingstone, 2003), Australia (Martino & Kehler, 2006) and non-Western countries such as China and Nigeria (Mathwasa and Sibanda, 2021) are increasingly paying attention to the lack of male teachers in both kindergartens and primary schools.
The shortage of MKTs may be attributed to problems and unequal treatment they experience at work.For example, parental mistrust of males in ECE settings (Johnson et al., 2010), being unable to get along naturally with female colleagues (Mistry & Sood, 2015), or feeling lonely and neglected at work (Ashcraft & Sevier, 2006;Yang & McNair, 2019) are some potential factors.Therefore, this study draws attention to the relationships between the factors that can influence MKTs' job satisfaction (JS) and tests the relationship between JS and three dimensions of organisational commitment (OC), which include affective commitment, continuance commitment and normative commitment to raise awareness of what needs to be done to improve the recruitment and retention prospects for MKTs and how their work experience might be improved.

Literature Review
Previous literature has proposed few theories relating to the JS and OC of MKTs.To address this, this research builds a framework based on classical theories, including the Herzberg et al. (1959) Two-Factor Theory that summarises motivational factors and working conditions that might improve employees' JS from the two dimensions of satisfaction and dissatisfaction.In addition, the Hierarchy of Needs Theory (Maslow, 1943) is applied because of the way it can be used to help motivate employees to work by understanding their individual needs and Allen and Meyer's (1990) three dimensions of OC are drawn on, these being: "affective", "continuance", and "normative" (p. 6).

Relationships Between Influencing Factors and JS
ECE has long been considered a woman's job, and there are few men in this profession.Women occupy a dominant position which is consistent with the perceived public attitude towards this profession (Drudy, 2008).Johnson et al. (2010) state that male teachers, when trying to get along with young children, often face doubts from their parents, such as their work motivation, sexual orientation, and whether their personalities are suitable for the job.Compared with female teachers, MKTs suffer from stereotyping by society and face a higher level of questioning and supervision in their work (Cushman, 2005).However, Koperna (2019) suggests that, although male kindergarten or primary school teachers can often face bias and doubts from parents, parents are sometimes shown to have a more positive view of male teachers in which they focus more on personality than on gender.However, in China, Liu et al. (2018) emphasise that although some parents have realised the importance of MKTs, the whole society has not eliminated the prejudice against MKTs.They go on to say that it is also necessary for society and the government to increase the publicity of the importance of MKTs.
In terms of salary, men often occupy a dominant position and have an advantage in the labour market (Crompton,1997) by having a higher status, more power, and higher salaries than women (Owen, 2003;World Economic Forum, 2023).However, in ECE, kindergarten teachers' salaries are generally lower, which can lead to dissatisfaction (An & Bonetti, 2017;Rolfe et al., 2003).Therefore, low salary can be one of the barriers for men choosing a career in ECE (Bury et al., 2020;Rice & Goessling, 2005), particularly when men are often less satisfied with their salary than women (Şahin & Sak, 2016).
In addition to negative public attitude and pay dissatisfaction, there are few studies that focus specifically on MKTs' work pressure (Collie et al., 2011;Kyriacou, 2001).Work pressure can arise from low salary levels (Drudy, 2008), low social status (Carrington & McPhee, 2008) and low levels of recognition by the education sector (Cushman, 2005).Each of these have been shown to be influencing factors in the low levels of recruitment and retention of MKTs.In addition, a reasonable workload is one of the key factors in ensuring employees' work efficiency (Elliott et al., 2014), and therefore, unreasonable work intensity and workload may lead to employees' low levels of JS (Tentama et al., 2019).It is recognised that the working environment of kindergarten teachers can be intense (Čecho et al., 2019), and work pressure, work fatigue and persistently high levels of workload increase the risk of emotional disorders, job burnout and job dissatisfaction among kindergarten teachers (Hozo et al., 2015).
According to Garmendia et al. (2023), social support exerts a significant influence on employee satisfaction, encompassing aspects such as management style, leadership relationships, colleague interactions, and related factors.Therefore, interpersonal relationships between MKTs and colleagues have also been considered (Yin, 2018), and as a minority group in primary education, male teachers often feel social isolation (Ashcraft & Sevier, 2006).Stereotypes and prejudices from society mean that MKTs are often considered as not belonging to or accepted members of the primary educators' group (Gosse, 2011), and these stereotypes are more evident in the context of ECE (Yang & McNair, 2019).Huang (2007) also reports that MKTs often feel lonely at work due to a lack of common interests and conversation topics with their female counterparts.
Promotion and professional development are other factors that may influence MKTs' JS (Şahin and Sak, 2016).Their study compares the aspects of JS of male and female kindergarten teachers and indicates that male teachers are on average more satisfied than female teachers in terms of promotion, which may be due to a perception that men are more suited to leadership and management (Uwizeyimana & Mathevula, 2014).Regarding professional development, Tucker (2015) reports that primary school teachers in the US often complain about the few training opportunities to "examine, navigate, and address identity" (p.16).They want to receive professional training, as it could allow them to be more aware of their professional identity and the advantages they bring to the role (Tucker, 2015).Furthermore, men who actively display their profession through in-school training are more likely to promote a positive image of MKTs (Koch, 2015).
Finally, the physical work environment refers to people's workplaces, such as offices, factories, workshops and working facilities (Lin, 2003).Nordberg (2002) indicates that there are different work cultures in different industries, and the work culture in kindergartens is based on femaledominated behaviour patterns and opinions, and this is the norm since most kindergarten teachers are female.Børve and Børve (2017) report that employees need to design the kindergarten environment with gender characteristics in mind.For example, games and exercises involving male teachers and boys need to be macho, such as "toughness" and "roughness" (p.1078), while female teachers and girls-related exercises need to be gentle and feminine (Børve & Børve, 2017).Although there are male teachers in some kindergartens, there are few places for them in the physical environment construction, and female culture is often reflected in this design (Børve & Børve, 2017).Despite this perceived need, Brownhill and Oates (2017) challenge the so-called "imposition" (p.664) of masculine roles and the notion of the benefits of a male role model in favour of "celebrating the diversity and capacities" (p.668) of workers in early years settings.
Therefore, factors such as public attitude, pay satisfaction, interpersonal relationships, work pressure, promotion and professional development, and physical work environment can be cited as factors affecting MKTs' JS, which refers to a level of favourable judgement from people thinking about their work (Skaalvik & Skaalvik, 2007).

Relationship Between JS and Three Dimensions of OC
The following section introduces studies that explored the relationships between JS and affective commitment, continuance commitment, and normative commitment.
Research by Behera et al. (2023) shows that employee satisfaction is closely related to employee commitment, and since OC directly affects employee turnover, performance and organisational behaviours, it is crucial to satisfy employee JS.Regen et al. (2023) also reached a similar conclusion, indicating that JS and leadership style have an influence on OC, and this is further supported by the international Starting Strong Survey (Organisation for Economic Cooperation and Development (OECD, 2020) that focussed on the work of early childhood educators.Regarding MKTs, Zhang et al. (2023) collected questionnaires from MKTs (n=402) in China, and the results show that perceived organisational support has a positive and direct influence on career commitment, and career happiness plays an intermediary role between organisational support and career commitment.Their findings suggest that MKTs' JS affects their commitment and reduces turnover rate (Zhang et al., 2023).
Concerning affective commitment, this refers to employees' emotional dependence, identification, and commitment to the organisation (Shore & Wayne, 1993).Although the Starting Strong Survey (OECD, 2020) does consider emotional well-being of teachers within ECE, research on affective commitment (AC) in ECE is limited, so research on AC in other industries is also considered here as a reference point.Affective commitment is often used separately to investigate its relationship with JS, indicating that the higher the employee's JS, the higher their affective commitment to the organisation (Tarkar et al., 2019).Similarly, studies in other industries also report a positive relationship between these two, such as the police profession in Taiwan (Kuo, 2015), the IT profession in India (Patrick & Sonia, 2012), and the construction profession in Hong Kong (Leung et al., 2008).
For continuance commitment, this refers to when employees must stay in an organisation to avoid losing their position and the benefits they have received after years of investment (Colquitt et al., 2011).Positive relationships between JS and the three dimensions of OC have been reported, and the higher employee's JS, the harder it was for them to find another job with the same level of satisfaction, so there was a higher continuance commitment to the organisation (Clugston, 2000).Similarly, Cramer (1996) conducted a thirteen-month, twowave panel study on British engineers.More recently, Alsiewi et al. (2016) found that in educational settings there was a significant positive relationship between the quality of JS and continuance commitment.
Finally, normative commitment is defined as the employee's perceived obligation to stay in the organisation (Allen & Meyer, 1996) or their personal and organisational values that fit well (Stephens et al., 2004).McCormick and Donohue (2016) report that the consistency of individual values and the organisation's mission and values positively impact normative commitment.Volunteers are typically motivated to join an organisation because their personal beliefs align with the organisation's values, resulting in a generally high normative commitment to it (Souza & Dhami, 2008).Also, Imam et al. (2013) also show that JS positively impacts affective commitment in bank employees in Pakistan.The quantitative study by Alsiewi (2016) also found a statistically significant relationship between JS and normative commitment in primary schools.

Methods
Quantitative methods were used to explore relationships, and the hypotheses were formulated as follows and summarised as a conceptual model in Figure 1: H1: Is there a positive relationship between public attitude and JS? H2: Is there a positive relationship between pay satisfaction and JS? H3: Is there a positive relationship between work pressure and JS? H4: Is there a positive relationship between interpersonal relationships and JS? H5: Is there a positive relationship between promotion and professional development and JS? H6: Is there a positive relationship between physical work environment and JS? H7: Is there a positive relationship between JS and affective commitment?H8: Is there a positive relationship between JS and continuance commitment?H9: Is there a positive relationship between JS and normative commitment?
Based on the literature and theories above, this study was designed around six influencing factors: public attitude, pay satisfaction, work pressure, interpersonal relationships, promotion and professional development, and physical work environment.Additionally, this study references Allen and Meyer's (1990) three dimensions of OC: affective commitment, continuance commitment, and normative commitment.

Research Sample
The participants for the research were all in-service MKTs in China (n=194).An online questionnaire was shared via MKTs' WeChat groups, WeChat friend circles and QQ groups (instant messaging software services in China) to encourage potential participants through convenience and snowball sampling.

Ethics
All procedures, including the questionnaire design, were carried out in accordance with ethical requirements.Informed consent was gained, and the anonymity and confidentiality of participants was assured.

Research Instruments
The questionnaire in this study consisted of two main parts (JS and OC) and five sub-sections (demographic characteristics, JS, affective commitment, continuance commitment and normative commitment).Except for those in the demographic characteristics section, each question in the other sections was measured by the 7-point Likert scale.
Questions related to demographic characteristics were designed based on the characteristics of MKTs and kindergarten teachers, with reference to previous questionnaires on the JS of kindergarten teachers (Bosh, 2020;Gao & Zhang, 2012).The OC questionnaire in this research was divided into three parts according to Allen and Meyer's (1990) OC scale.The JS questionnaire was designed by reviewing Herzberg et al.'s Two-Factor Theory (Herzberg et al., 1959), Maslow's Hierarchy of Needs Theory (Maslow, 1943), Teachers' JS Questionnaire (Lester, 1987), the Minnesota Satisfaction Questionnaire (short form) (Weiss et al., 1977), and the Questionnaire on JS of Kindergarten Teachers (Gao & Zhang, 2012).

Reliability and Validity
The value of chi-squared is greatly affected by the number of variables and the amount of data.Normally, the greater the number of variables (the greater the matrix), the greater the chisquared value is.According to De Smith (2018), when the chi-squared degree of freedom ratio is between 1 and 3, the hypothesis model's fitting degree is acceptable.In this study, CMIN was used to represent the maximum likelihood ratio chi-squared value.For the normative commitment questionnaire, the minimum discrepancy (CMIN/DF) was 2.853, less than 3, and the root mean square error of approximation (RMSEA) was 0.098, lower than 0.1.Therefore, the CMIN value of the default model indicates that the model fit was acceptable.The value of chi-squared is greatly affected by the number of variables and the amount of data.Normally, the greater the number of variables (the greater the matrix), the greater the Chi-Square value is.According to De Smith (2018), when the chi-squared degree of freedom ratio is between 1 and 3, the hypothesis model's fitting degree is acceptable.In this study, CMIN was used to represent the maximum likelihood ratio chi-squared value.
The comparative fit index (CFI) and Tucker-Lewis index (TLI) are the value-added fit indexes.These usually compare the theoretical hypothesis model to be tested with the independent model to judge the degree of fit of the model.The values are usually between 0 and 1.The closer to 1, the better the model fitting degree.In this study, the CFI and TLI were 0.969 and 0.939, respectively, indicating that all the fit indexes were good, and the model was a good fit.
Concerning the other three parts of the questionnaire (JS questionnaire, affective commitment questionnaire and continuance questionnaire), the RMSEA of the JS questionnaire and affective commitment questionnaire were 0.104 and 0.107 respectively.The RMSEA of the continuance commitment questionnaire was lower than 0.1, and the other indexes, such as TLI and CFI, of the three questionnaires were all greater than 0.8.The CMIN/DF was lower than 5, so based on these values the validity of all the questionnaires was qualified.
The reliability for the whole questionnaire was 0.960, and the reliability of the JS questionnaire, affective commitment questionnaire, continuance commitment questionnaire and normative commitment questionnaire were 0.928, 0.852, 0.889 and 0.809, respectively.The Cronbach's Alpha of each part of the questionnaire was higher than 0.8, indicating that the reliability of the questionnaire was good.

Results
The following results show the relationship between the influencing factors and JS, and the relationship between JS and three dimensions of OC.In the first half of the model, secondorder confirmatory factor analysis (CFA) was used to test the relationship between six influencing factors and JS.In the second half of the model, the linear regression analysis was used to adhere to the relationship between JS and three dimensions of OC.
Concerning the first part of the model, the results show that the indexes of the model meet the requirements.The CMIN of the default model was 2.511, indicating that the degree of fit of the model was acceptable.The goodness of fit index (GFI) and the adjusted goodness of fit index (AGFI) were greater than 0.8, and the normed fit index (NFI) was greater than 0.9, indicating that the degree of fit for the hypothesis model was good.Although the RMSEA in this study was 0.088, other indexes were good.Therefore, this index might have little influence on the results even though the second-order equation model composed of six influencing factors and JS satisfied the reliability and validity.
The first part of the model tests the relationship between JS and six influencing factors, including public attitude (H1), pay satisfaction (H2), work pressure (H3), interpersonal relationships (H4), promotion and professional development (H5), and physical work environment (H6).Table 1 and Figure 2 below represent the results of CFA of the relationship between six influencing factors and JS.
According to Figure 2, there are two related questions under each influencing factor, because the model was first checked for unreliable questions when developing the questionnaire.If there is a serious deviation in the correspondence between the factor and the question, the question can be deleted.Additionally, if the factor loading between the measurement item and the factor is too low (for example, less than 0.5), this indicates that the relationship between the question and the factor is weak, and the question needs to be deleted (De Smith, 2018).
Where there were questions with weak relationships under each influencing factor these were removed, resulting in the support of two questions per influencing factor.As can be seen from Figure 2, when measuring the relationship, the absolute values of the standardised coefficients between the remaining influencing factors and the questions are all greater than 0.5 and are significant, which means that there is an effective measurement relationship here.For other model indicators, such as CMIN, GFI, AGFI, NFI, and RMSEA, these have earlier been presented as indicators of a good model fit.
Table 1 indicates that among the six influencing factors, interpersonal relationships and public attitude had the most significant impact on JS, with the estimated beta values being 1.021 and 1.063, respectively.Although the absolute values of both values are greater than 1, this may be due to the relatively small sample size (n=194).However, pay satisfaction had the least effect on JS, with a beta value of 0.701, showing that there was a statistically significant relationship between them.Furthermore, the beta values for the other variables (physical work environment, promotion and professional development, and work pressure) were 0.945, 0.957 and 0.963, respectively, indicating that these three influencing factors had a statistically significant relationship with JS.
For the first part of the model, in general, H1 to H6 were validated and supported effectively.This means six influencing factors had a positive and significant influence on JS.Public attitude had the most significant statistical impact on JS, while pay satisfaction had the least significant impact.Table 1.Regarding the second part of the model, the relationship between JS and three dimensions of OC, including affective commitment (H7), continuance commitment (H8), and normative commitment (H9) is tested.Table 2, Table 3 and Table 4 show the results of regression analysis of the relationships between JS and affective commitment, continuance commitment, and normative commitment, respectively.

Standardised Regression Weights of CFA
The demographic characteristics were considered as the control variable, including age, educational level (below high school graduate, high school graduate, associate bachelor, undergraduate, graduate), whether participants majored in ECE, kindergarten type (public or private), tenure and city tier (seven tiers of city size based on the city's economic development).
JS was taken as the independent variable, and affective commitment, continuous commitment and normative commitment were taken as the dependent variable for linear regression analysis.
Regarding the relationship between JS and affective commitment, based on Table 2, the R square value of the model was 0.629, which means that the demographic characteristics and JS of participants could explain 62.9% of the change in their response to affective commitment.The model passed the F test (F = 42.362, and p < 0.01), which means at least one of the participant's demographic characteristics and JS had an impact on their affective commitment.
Based on the results, it is evident that the JS of participants independently influenced affective commitment and played a positive predictive role (beta value = 0.755, p < 0.01) in determining affective commitment.In terms of the relationship between JS and continuance commitment, as indicated in Table 3, the R-squared value of the model was 0.676.This denotes that demographic characteristics and JS could account for 67.6% of the variance in participants' continuance commitment to kindergartens.The model successfully passed the F-test (F = 52.093,p < 0.01).Notably, Table 3 reveals that demographic characteristics did not predict continuance commitment, whereas JS positively predicted continuance commitment (beta value = 0.807).Regarding the linear regression analysis of JS and normative commitment, Table 4 illustrates that the R square value was 0.521, indicating that the change in the participants' demographic characteristics and JS explained 52.1% of the changes in normative commitment.The model passed the F test (F = 27.176,p < 0.01).Table 4 further shows that kindergarten type had a negative impact on normative commitment (beta value = -0.192,p < 0.01), and city tier and JS positively influenced normative commitment (beta value = 0.147, and p = 0.008 for city tier; beta value = 0.626, and p < 0.01 for JS).In summary, according to the results, H7 to H9 were validated and supported.JS had a positive and significant impact on affective commitment, continuance commitment and normative commitment.Additionally, kindergarten type and city tier exerted influence on normative commitment, although the magnitudes of these effects were relatively low, which means these two demographic characteristics had a low impact on normative commitment.

Public Attitude
Of the six influencing factors, the public attitude towards male teachers engaging in ECE is the most important factor affecting JS.This is consistent with Maslow's (1943) Hierarchy of Needs Theory (fifth level, self-actualisation needs) and the Herzberg et al. (1959) Two-Factor Theory, indicating that high employee satisfaction not only comes from salary but also from social recognition.
Additionally, Foster and Newman (2005) assert that some men interested in primary education are sometimes considered abnormal, homosexual and a threat to children.Skelton (2001) mentions that some parents may have prejudices about paedophilia and abnormality against MKTs.However, this accusation is demeaning of female-led work (Yang, 2015).The questions in this part of the questionnaire relate to whether children's parents respect participants and the attitudes of participants' families and friends towards their engagement in ECE, which were the factors that affect their JS according to statistical data.This is also consistent with the previous literature.

Pay Satisfaction
The salary issue has been shown to be a material factor with a statistically significant and positive impact on JS.It was supported in previous studies relating to JS in the field of ECE (Chen et al., 2017;Garcia, 2011), education (Winkelmann & Eberman, 2017), and in other industries (Sharma & Bajpai, 2011).
Although in this research, the beta-value of pay satisfaction was the lowest (0.7) compared with the other five factors, its influence on JS was statistically significant.The low salary satisfaction rated by participants corroborates previous research findings ( Barnett & Hustedt, 2003), which found that compared with other industries, the salary of kindergarten teachers is often lower than that of employees with the same qualifications.Furthermore, the higher the education level, the more significant the gap between the salary of teachers and those of the same educational level in other fields.

Work Pressure
Work pressure, promotion and professional development had the same beta value (.960), and as the third influencing factor, it had a positive prediction of JS.Work pressure affects JS (Parveen & Bano, 2019), burnout (Rakovec-Felser, 2011), and job performance (Sandilos et al., 2018).This research confirmed its positive relationship with JS, which is consistent with previous findings in the literature.
Employees' workload is an essential indicator affecting JS (Cole et al., 2004;Lea et al., 2012), and a reasonable workload is a critical factor in ensuring employees' work efficiency (Elliott et al., 2014).On the contrary, unreasonable work intensity and workload may lead to employees' decreased JS (Tentama et al., 2019).As shown in the quantitative relationship of this study, when the work pressure of the participants was low, they rated their JS high.Conversely, their JS decreased accordingly when they had high daily work stress.

Interpersonal Relationships
Interpersonal relationships at work are an important factor affecting Chinese MKTs' JS.This is consistent with the results of Wong and Zhang (2014), indicating that teachers who have good interpersonal relationships often have higher JS.Harmonious relationships can make them more willing to stay in their school (Renzulli et al., 2011).Also, Yang (2015) indicates that kindergarten teachers' happiness mostly comes from the children, and a good relationship with children is one of their motivations to stay in their job.
This study included questions about relationships with leaders, colleagues, and children.This is consistent with the content and results of some previous research designs.For instance, Sumsion (2000) interviewed a 30-year-old Australian male kindergarten teacher, who complained that the most he told children's parents was that he was already married and had children to reassure them of "his sameness rather than difference" (p.134), although he felt it should not be necessary to explain this.Moreover, Yin (2018) also claims that MKTs often feel alienated and isolated in school, which can lead to dissatisfaction with their work.

Promotion and Professional Development
The factor of promotion and professional development was ranked third and had a statistically significant influence on JS.Concerning promotion, although there is limited literature on the effect of promotion on JS in ECE, in other industries, the effect of promotion on employee JS (Shields & Ward, 2001) and job performance (Razak et al., 2018) have been indicated.As key incentive measures in human resources, promotion and career development play an important role in retaining talent, enhancing satisfaction, and fostering employee performance (Razak et al., 2018).
Concerning professional training, related courses can enhance teachers' satisfaction with the work experience (Liu & Jih-Lian, 2016).They contribute to teachers' improvement in professional knowledge and skills (Evans, 2002) and help teachers better understand the student-centred teaching model and promote students' efficient learning (Hanbury et al., 2008).This is consistent with the results of this study.Promotion and professional development have a positive impact on JS.Furthermore, Whalley (2011) emphasises that teachers who are motivated to develop their skills and qualities significantly influence their comprehension of children's physical and mental development, impacting their ability to provide an optimal learning experience.

Physical Work Environment
The relationship between environmental factors, such as working conditions, work environment and facilities, and JS has been identified in previous studies (Albashayreh et al., 2019;Al-Hamdan et al., 2017;Herzberg et al., 1959).However, in ECE there are few studies, and most focus on the classroom environment in kindergartens and its impact on children's participation and learning development (Aydoğan et al., 2015;Downer et al., 2007).Therefore, this is a new finding in this research.
According to the results of this study, the physical work environment shows the impact of environmental attributes on male teachers' JS.This is consistent with the results of previous studies in other industries.For example, Baernholdt and Mark (2009) report that a pleasant physical work environment positively impacts employees' working moods.When employees work in an unacceptable environment, their performance can be largely affected (Nowakowska et al., 2016).

Affective Commitment
Teachers' feelings towards a school can affect their work performance, and high affective commitment often helps teachers contribute their abilities and skills to the teaching process (Ting, 2011).This is consistent with the results of this study whereby MKTs' JS positively affects their emotional commitment to school, a recognition of the values of the kindergarten, a sense of belonging, and good interpersonal relationships.Again, whilst there is limited literature on JS and affective commitment in the field of ECE, the relationship between the two has been identified in other industries (Patrick & Sonia, 2012;Tarkar et al., 2019), educational settings (Demirtaş, 2018;Dinc & Kocyigit, 2017), and primary education (Dalgıç, 2014;Karadağ et al., 2011).

Continuance Commitment
Compared with the affective and normative commitments, participants in this study had the lowest evaluation of continuance commitment to their kindergarten.However, it had the highest significant statistical relationship with JS.Previous studies also indicate that JS plays an essential role in employees' continuance commitment, and there is a positive correlation between the two (Agarwal & Sajid, 2017).It is the same with this study, indicating the statistically significant relationship between JS and continuance commitment.
However, Hackett et al. (1994) state that JS is based on employees' passion for the work and recognition of organisational values, whilst the motivation to continue to serve the organisation because of material reasons indicates a low level of JS.However, in this study, pay satisfaction, as a measurement dimension, contributed to the overall JS.In other words, JS could mediate between continuance commitment and its influencing factors (e.g., pay satisfaction) (Yousef, 2002).

Normative Commitment
According to Allen and Meyer (1990), normative commitment is the final dimension of OC, and it indicates employees stay with an organisation since they "ought to" (p.34) (Parish et al., 2008).In this study, questions on normative commitment also present their loyalty and sense of responsibility to the kindergarten because of their JS.
The positive relationship between JS and normative commitment was identified in this study, and participants' moral commitment was often reflected in the form of a sense of responsibility or indebtedness to the kindergarten.Imam et al. (2013) report that JS positively impacts all dimensions of OC, and it can predict the employee's affective commitment, continuance commitment and normative commitment to the organisation.The findings of this study are consistent with this view, suggesting that participants' JS could predict their normative commitment to kindergarten.

Conclusion
The results showed that six influencing factors positively impacted JS, and there was a positive relationship between JS and three dimensions of OC.However, there were some limitations.
Firstly, three questions were designed for each influencing factor in the questionnaire.In the later analysis, some questions were deleted to improve its reliability and validity.Therefore, the addition of more questions would allow for better measurement of each factor.Secondly, the number of valid questionnaires returned was relatively small (n=194) when it came to establishing strong returns for the more nuanced questions.In future research, a larger sample of MKTs might provide a more convincing exploration of salary and working conditions, and material requirements and salary satisfaction.
From the perspective of policy and practice, combining the factors that affect MKTs' JS and OC, government cooperation with kindergartens to develop policies more conducive to improving levels of MKT recruitment and enhancing ECE job satisfaction is needed.In terms of ECE undergraduate and postgraduate study, there is still work to be done around dispelling myths and stereotypes around males in ECE.Universities and teacher training providers have a role to play in actively promoting the value of male teachers in ECE, supporting and encouraging male students to apply, and monitoring and adjusting the enrolment ratio of male to female students.In so doing, the benefits that both male and female kindergarten teachers bring to create a balanced offering to young children's development can be realised.
Secondly, kindergartens should strengthen the training of MKTs.The relevant training courses must pay attention to the characteristics and needs of male teachers to give full recognition and support to their advantages and functions.Kindergartens and training providers should explore training programs suitable for MKTs so that overall the training for all kindergarten teachers is strengthened regardless of gender stereotypes.
Finally, the advantages of MKTs should be explained to parents and communities to help achieve a more representative notion of the importance of ECE workforce diversity within contemporary society.

Figure 2 .
Figure 2. The second-order CFA model of factors affecting JS Note.PS represents pay satisfaction, and PWE represents physical work environment.WP, IR, PA and PPD represent work pressure, interpersonal relationships, public attitude and promotion and professional development, respectively.

Table 2 .
Linear Regression of JS and Affective Commitment

Table 3 .
Linear Regression of JS and Continuance Commitment.

Table 4 .
Linear Regression of JS and Normative Commitment