Teacher-Student Relationship and Teacher Subjective Well-being: Understanding the Employment Status as Moderator

: Teacher-Student Relationship and Teacher Subjective Well-being: Understanding the Employment Status as Moderator. Objective: This research aimed to see whether the teacher employment status could moderate the association between teacher-student relationship and teacher subjective well-being in secondary level of education. Methods: This study uses a quantitative research approach with a non-experimental correlational design. Correlation analysis and simple regression analysis was used to measure the moderation effect. Findings: The result showed that no significant moderation effect of the employment status on the effect of teacher-student relationship to the teacher subjective well-being. Conclusion: The findings of this study fail to confirm the moderation effect of employment status on the association between teacher-student relationship and teacher subjective well-being in secondary level of education.


 INTRODUCTION
The teaching profession is a job that has a big responsibility and demands careful preparation.High levels of stress resulting from teaching, doing other work in school such as administrative matters, with low income, are all associated with the well-being of teachers in carrying out their job (Johnson et al., 2005;Kauppinen et al., 2013).Teacher subjective wellbeing is an optimal condition for teachers in their work as seen from how confident the teachers are in delivering material to students and how bound and supported the teachers feel in their working environment (Renshaw, Long & Cook, 2015).Therefore, being a teacher is a job that requires extra energy on both emotional and interpersonal aspects (Milatz, Lüftenegger, & Schober, 2015).
In the last few years, the concern about subjective well-being of teachers in Indonesia has call the awareness of the society.Teachers in Indonesia often face various challenges, such as heavy workload, curriculum changes, juvenile delinquency and the gap in rights between honorary/temporary teachers and permanent teachers which can affect the welfare of these teachers.In 2019, the Indonesian Child Protection Commission (KPAI) received 153 reports or cases of violence happening in the school (Wartaekonomi, 2019).Of all the complaints received, 13% were violence committed by students against teachers.The types of violence committed include delinquency, physical abuse, verbal abuse, and bullying (Aprilianto, 2018;Rahadi, 2019;Rahayu, 2019).These student delinquencies will exacerbate the relationship that exists between the students and the teachers so that in the end it makes the teachers feel uncomfortable with their job.The gap in welfare based on teacher employment status is also a highlight in Indonesia's education setting.Temporary teachers are generally in the spotlight because their rights such as support, facilities, and the wages received are not adequate to support their welfare (Indraini, 2019;Rifa'i, 2019).
Teacher will feel low in subjective wellbeing when they start to feel unpleasant and unmeaningful in doing the work.This happens when the teachers start to think that their job is no longer meet their expectation.When the teachers do not feel successful in delivering material to students and/or feels that their work environment does not provide positive support for themselves, the teacher will experience a negative impact, namely decreased attitudes towards work and job performance (Schaufeli and Enzmann, 1998;Vandenberghe & Huberman, 1999;Schaufeli, 2003;Hakanen, Bakker & Schaufeli, 2006).For example, teachers who feel unsupported by their school, such as not having the opportunity to get training, will make the teacher feel neglected and affect teacher performance (e.g decrease in the effort made in completing assignments).The tendency of teachers to leave work will also increase when the teacher feels that their job is too stressful (Montgomery & Rupp, 2005).Students also feel the negative impact of the decline in teacher performance, such as decreased student achievement (Klusmann, Kunter, Trautwein, & Baumert, 2006;McLean & Connor, 2015).
The level of stress resulting from teaching is assumed to increase at the secondary/middle to high school level.One of the factors that cause an increase in stress levels is the high proportion of delinquency in middle/high school students compare than elementary or kindergarten (Aprilianto, 2018).As in the misbehavior committed by students, the teachers will assume that their relationship with the students is on a bad term so that the teachers feels uncomfortable with their work.According to Hargreaves (2000), there are differences in the quality of the relationship between teachers and students in early childhood education and elementary school level, compared with the teacher-student relationship at the secondary school level (SMP and SMA).Teachers feel they have a closer relationship with students at the early childhood and elementary school levels than with students at the secondary school level.Meanwhile, the quality of the teacher-student relationship has contributed to the quality of teacher life.Therefore, based on previous studies, further research is needed on secondary school teachers.
Besides the teacher-student relationship, the role of teacher employment status, both permanent and temporary, is also a factor that can predict teacher subjective well-being (Setiyawan, 2017).Teacher employment status is the status possessed by teachers as employees in schools which are distinguished based on whether the teacher is appointed as a permanent teacher or only as a temporary teacher (Meiza, 2017).A permanent teacher is a teacher who has been appointed by the government to become a civil servant and has an official assignment letter or is appointed as a permanent teacher at the private schools and is bound by a work contract.On the other hand, non-permanent (temporary) teachers are teachers who are appointed by school agencies.They have equal duties and responsibilities as permanent teachers.However, they do not get the same facilities as permanent teachers.Between the two employment statuses, there is a welfare gap that is seen based on the rights received.
Studies concerning the teacher employment status in Indonesia are varied and unique, because only Indonesia has differences in teacher employment status.For example, according to Setiyawan (2017), the role of teacher employment status can affect the level of happiness of teachers.Permanent teachers (like civil servants) tend to have a higher level of happiness than non-permanent teachers (temporary).This can be explained by the difference in rights received by both parties such as the amount of salary received.Permanent teachers get the fixed salaries ranging from 2 to 10 million per month which are influenced by applicable local regulations, while non-permanent teachers have an average salary of around 500 thousand rupiahs (Deny, 2018;Liputan6, 2018;Qerja, 2016).According to Ma & Zhang (2013), income is positively correlated with the level of happiness experienced by individuals, so that permanent teachers and temporary teachers will show different levels of happiness considering the amount of salary they receive.This finding is different result by Meiza (2017) who found that there was no difference in the happiness experienced by permanent teachers and temporary teachers.
Many variables can affect the well-being of teachers.There are variables such as salary, certification, employee status (permanent or honorary), and the relationship that exists between teachers and students (Zulkifli, Darmawan, & Sutrisno, 2014;Sari, Imron, & Setyadin, 2013;Hargreaves, 2000).The relationship between teachers and students is the most determining factor in teacher subjective well-being (O'Connor, 2008).This is because basically teacher and student considered as a unit in the learning process.As stated by Shann (1998), the relationship that exists between teachers and students is the most enjoyable and most important part of the teacher job.This can be the reason why someone choose teaching as their career or choose to stay in the job.Perhaps teacher feel less in well-being but the influence of the teacher's employment status can be covered by having a harmonious relationship with the students being taught.
From the background that has been explained above, the role of employment status can affect the associations between the teacherstudent relationship and the teacher subjective well-being at the secondary/middle school level.Based on the previous studies, this study aims to find out whether the teacher employment status can strengthen or weaken the impact of the teacher-student relationship on the teacher subjective well-being at the secondary school level.The hypothesis proposed in this study is teacher employment status moderate the effect of teacher-student relationship on teacher subjective well-being at the secondary school level.

 METHODS
This study used a quantitative approach with a correlational design to examine the associations of teacher-student relationship and teacher subjective well-being sensed by teachers.It also examined the roles of teacher employment status in strenghtening or weakening the effect of teacher-student relationship and teacher subjective well-being.This study took place in Jakarta and other cities, at secondary/middle school teachers.The minimum number of samples required for this study was estimated using a priori power analysis via G*Power application developed by Faul, Erdfelder, Buchner, and Lang (2009).Based on G*Power analysis, the minimum samples size of 188 required to meet the beneficial of regression analysis with two tail hypothesis and significance level of 0.05, the effect size of 0.15 and statisticcal power of 0.99.This study was conducted within a year (Januari -December 2020).This time was utilized to search for theoretical references to get an overview of the phenomena and solidify the research questions, collecting and analyzing the data.
The participants in this study comprised secondary/middle school teachers in Indonesia.The research sample was determined using convenience sampling techique.In convenience sampling, we choose the teachers who are easy to get as participants in this study (Forzano & Gravetter, 2012).We share the information about the study to our fellow teacher network and ask teachers to participate in the study by distributing the link of the questionnaires.We collected the data using an online questionnaire by the Google Form application.The online method in gathering data was carried out because the pandemic COVID-19 hits, and data collecting with direct contact was prohibited.
There were 289 secondary school teachers who participated in this study.Most of the participants were female (65.8%, n=187).They were many cities, but most of them came from Jakarta and the surrounding cities (Bodetabek area) (69%, n=196).The participants' age ranged from 20-60 years (M = 38.58,SD = 9.899).Related to teaching experience, most teachers have 1-5 years of teaching experience (32%, n=91).Of the total participants, more than half have permanent employment status (69%, n=196) and the rest had non-permanent or temporary employment status (31%, n=88).
The question items were outline in the online questionnaire, using Bahasa Indonesia.The instruments were gone through the translation process to Bahasa Indonesia and back translation to English, with the help from English Literature and Psychology.We also conducted the instrument trial to several secondary/middle school teachers to examined the test reliability and validity and to ensure the participants understand the words and sentences used in the questionnaires.
The instrument used to measure independent variable, which is teacher-student relationship, is the Student-Teacher Relationship Scale (STRS) which was first developed by Pianta (2001) and further revised by Aldrup et al. (2018).The instrument has 6 items with Likert-type scale of 1 to 4, where 1 represents "strongly disagree" and 4 represents "strongly agree", and measure the total score obtained from the participants.One of the items contained in the instrument is "My students respect me".The instrument (STRS) has a Cronbach Alpha coefficient of .836and a range of validity coefficients using the evidence of homogeneity method, namely the corrected item-total correlation (cr-IT), between .445 to .697.Based on a good reliability limit according to Urbina (2004), the STRS instrument is reliable in the sense that the measuring instrument is internally consistent or measures the same construct because the reliability coefficient of the STRS instrument is above .7.In addition, all items of the STRS instrument are valid with evidence of homogeneity because the cr-IT correlation index is all above .2(Nunnally & Bernstein, 1994).
The instrument used to measure dependent variable, which is teacher subjective well-being is the Teacher Subjective Well-being Questionnaire (TSWQ) developed by Renshaw et al. (2015).The instrument has 8 items with Likert-type scale which are divided into 2 aspects, namely school connectedness with 4 items and teaching efficacy with 4 items, with an answer scale of 1 to 4. In this instrument, 1 represents "almost never" and 4 represents "almost always", and measure the total score obtained from the participants.One of the items contained in the instrument is "I am a successful teacher".The instruments has a reliability coefficient of .865and a range of validity coefficients using the evidence of homogeneity method, namely the corrected item-total correlation (cr-IT), between .518 to .721.Based on Urbina (2004), the TSWB instrument is reliable in the sense that the measuring instrument is internally consistent or measures the same construct because the Cronbach Alpha coefficient is above .7.In addition, all items of the TSWB instrument are valid with evidence of homogeneity because the corrected item-total correlation (cr-IT), is more than .2(Nunnally & Bernstein, 1994).
Based on the research framework, a hypothesis can be formulated.Hypothesis stated that teacher employment status can moderate the effect of teacher-student relationship to teacher subjective well-being.To answer the hypothesis, we analyze the moderation effect using linear and multiple regression analyses with Hayes' Model (Hayes, 2013).Descriptive statistical tests and correlation analysis between variables were also conducted to see the general picture of participants and preliminary analysis before testing the effect of predictors on the dependent variable.The data processing process was assisted using the IBM SPSS version 24 application.

 RESULTS AND DISCUSSION
Data analysis of the instruments using descriptive statistics gave general description about participants.First, teacher-student relationship score was in the range of 13-24, with mean score of 20.84 (SD = 2.417).The score indicates that the higher the score obtained by participants, the higher the closeness of the teacher-student relationship perceived by the teacher.Based on the mean score, the finding show that teachers perceived themselves have close relationship with their students.They feel that students like and respect them, take their words seriously, and trust them when students share their personal problems.Second, teacher subjective well-being scores was in the range of 18-32, with mean score of 26.86 (SD = 3.180).The score indicates that the higher score obtained by participants, the higher subjective well-being as teachers.Based on the mean scores obtained, the finding also show that teachers have high subjective well-being.They can be themselves in the school, feel success in teach and do other things as a teacher, and have a sense of efficacy in helping students learn new things.
This study aims to examine the moderating role of teacher employment status in the effect of teacher-student relationship to teacher subjective well-being in secondary/middle school teacher.Teacher subjective well-being is about how teachers perceive their well-being in their occupation, in the sense of connectedness with other stakeholders and efficacy in teaching.This perception can be seen in how strong their feeling of being supported by the school and having a good relationship with the civitas in the school, and how they assess their ability regarding his teaching ability that he has done his job well and according to standards (Renshaw et al, 2015).How the school treats the teacher can affect teacher perception about the support and from the stakeholders in the school.Therefore, the role of the various stakeholders in school in supporting teachers is quite important.On the other side, teachers' perception of the work they do is also important to make them feel happy and satisfied with themselves.Meanwhile, the teacher-student relationship deals with the perception of the relationship that teachers have with their students both personally and in groups (Pianta, 1999).Students as one of the stakeholders in school have an important role in supporting teachers, as well as be supported by teachers.Based on the findings presented in Table 1, it can be seen that the teacher-student relationship has a significant positive correlation with teacher subjective well-being (r=0.588,p< 0.01).This result indicates that if there is an increase in the degree of closeness in the teacher-student relationships, it will be accompanied by an increase in teacher subjective well-being.In other words, if teachers feel that they have a close relationship with students, they will feel more positive about themselves as a teacher.
The result of this study is congruent with the study by Hargreaves (2000), who found that teachers who can manage to form warm relationships with their students and feel that these relationships can help students learning more effectively will feel happier than a teacher who can't.Another study by Virtanen, Vaaland, & Ertesvåg (2019) also found the same result as this study, which is improving teacher-student interaction can improve teacher subjective wellbeing, and vice-versa.In that study also found that teachers who have a good relationship with students will feel more connected with them.Teachers also have a higher belief in their ability to teach.These kinds of teachers with profiles that are a combination of high levels of warmth and control, also have effective ways of fostering student learning and development.Pianta (2016) in his study also found that teachers who displayed high performance in supervising students and paying attention to the signs of students can identify students who have academic and emotional problems.
Another study also found that the relationship that exists between teachers and students is the most enjoyable and most important part of the teacher's job (Shann, 1998).The result found in our study is also in accordance with previous studies as stated by Milatz et al. (2015) Aldrup et al. (2018), and Spilt, Koomen, & Thijs, (2011) which state that the teacher-student relationship is positively correlated with teacher well-being.This can be explained by the teacher's work which is mostly allocated by interacting with students.According to O'Connor (2008), interpersonal relationships that exist between teachers and students can be a source of happiness and pleasure for a teacher's daily life, and also a reason for a teacher to want to continue working as a teacher.
Teacher employment status is defined as a state of employe appointment that distinguishes employees from one another in a company or organization (Hendrajana, Sintaasih, & Priartrini, 2016).Generally, there are two types of teacher employment status, which are permanent and nonpermanent (honorer).Our study found that there is no significant correlation between teacher employment status and teacher-student relationship.This result indicates that if there is a change in teacher employment status, it will not be accompanied by an increase in the teacherstudent relationship.An explanation of this may be explained as follows.Teachers' perceptions of the degree of closeness of their relationship with students were not related to their employment status.This means that both permanent and nonpermanent teachers have the same variation in the level of teacher-student relationships.Further research is needed to explore the relationship between these two variables, especially since current study only targetting secondary school teachers.
The result of the correlation analysis also show that there is a positive significant correlation between teacher employment status and teacher subjective well-being.The result in this study indicates that if there is a change in employment status (from non-permanent to permanent), it will be accompanied by the increase in teacher subjective well-being.The finding is congruent with the previous findings in which the teachers who have permanent employment status would have better job satisfaction (Putra & Arlizon, 2021).A teacher who has permanent employment status, such as a civil servant or permanent teacher, will feel satisfied with their lives and feel connected to others and have high efficacy in their teaching skills.
Teacher employment status also shows differences in predicting teacher subjective wellbeing where teachers still tend to feel they have higher teaching efficacy and attachment to schools than non-permanent teachers.This can be explained because of the different rights received from each of these employment statuses such as different salaries, benefits, and opportunities for development (Hendri, 2018;Indraini, 2019;Meiza, 2017;Rifa'i, 2019).According to O'Connor (2008), being a teacher is a calling from within, and having a harmonious relationship with students will cover other factors that can affect a teacher's subjective well-being.Therefore, although non-permanent teachers tend to have lower teacher subjective well-being than permanent teachers, this difference can be covered by having a harmonious relationship with students.A normality test was conducted to see the distribution of data from both variables.The normality test which was carried out using the One-Sample Kolmogorov-Smirnov Test method showed that the data distribution of the two variables are still normally distribute (p>0.05).Based on the information presented in Table 2, we found there is no interaction effect of teachers employment status on the teacher-student relationship in predicting teacher subjective wellbeing (b3=0.24,95% CI [-0.0125, 0.4884], t=1.87; p=0.0625).The result indicates that the effect of teacher-student relationship on teacher subjective well-being is not moderated by teacher employment status.Based on these results, it demonstrates that the hypothesis is rejected since there is no significant impact of teachers employment status on the teacher-student relationship in predicting the teacher well-being.Therefore, teachers employment status does not strengthen nor weaken the degree of the teacherstudent relationship in predicting the teacher subjective well-being.The finding shows that whether teacher have permanent employment status or non-permanent/temporary employment status, doesn't affect their perception of teacher-student relationship, also doesn't affect teacher subjective well-being.This study shows that teachers employment status was not a variable that could strengthen nor weaken the effect of teacher-student relationship and teacher subjective well-being for Indonesia's education context.This finding was congruent with study by Meiza (2017) who found that there was no difference in the happiness experienced by permanent teachers and non-permanent/ temporary teachers.But other previous studies suggested that teacher employment status could affect teacher subjective well-being (Zulkifli, Darmawan, & Sutrisno, 2014;Sari, Imron, & Setyadin, 2013;Setiyawan, 2017).
On the other hand, the studies which could support the correlation between teachers employment status and student-teacher relationship are scarce as far as we concern.Our study was conducted nevertheless to pioneer the moderating effect of teachers employment status to the well-known correlation between studentteacher relationship and teacher well-being since teacher employment status is regarded as an endemic and unique variable of Indonesia's education context.Note.F = 78,118, p = 0,000, R2 = 0,357, STR = Student-teacher relationship, TSWB = Teacher subjective well-being.
A multiple regression analysis is conducted to give a more explanation about the associations between teacher-student relationship, teacher employment status and teacher subjective wellbeing.Based on the information presented in Table 3, it can be seen that the teacher-student relationship and teacher employment status as a predictor of teacher subjective well-being showed significant results at F(2.281) = 78.118,p < .0005.Both variables proved statistically significant in predicting the dependent variable, p < .05.The predictor of teacher-student relationships has a beta coefficient of = .582in predicting teacher subjective well-being.This Nugraha et al., Teacher-Student Relationship and Teacher Subjective Well-being... shows that for every one-point increase in the total score for teacher-student relationships, the total score for teacher subjective well-being will increase by .582points.On the other hand, the predictor of teacher employment status has a beta coefficient of = .107in predicting teacher subjective well-being.This shows that teachers will still have a difference in teacher subjective well-being of .107greater than non-permanent teachers.These results prove that the research hypothesis is accepted.The magnitude of the effect size generated from this model is .357so it can be interpreted that 35.7% of the variance of teacher subjective well-being can be explained by the teacher-student relationship and teacher employment status.Specifically, the teacherstudent relationship has an effect size of .3387,and teacher employment status has an effect size of .0114 on teacher subjective well-being.This means that 33.87% of the variance of teacher subjective well-being can be explained by the teacher-student relationship and 1.14% of the variance of teacher subjective well-being can be explained by teacher employment status.Overall, there is an additional 0.69% variance explained from the overlapping effect of the two independent variables on the dependent variable.
This finding shows that to teacher-student relationship has greater effect size on teacher subjective well-being than teacher employment status.As confirmed with other studies, teachers interact with students everyday.The most important part and the most enjoyable part of being a teacher is having close relationship with students (Shann, 1998).One of the best ways to observer teacher subjective well-being is to observe the effect of teacher-student relationship in class.
The result of this study, i.e., no moderating role of teacher employment status in the relationship between teacher-student relationship and teacher subjective well-being, indicates that perhaps those other factors must be considered in this relationship.Several variables can strengthen or weakened the relationship between teacher-student relationship and teacher subjective well-being, such as school climate, school type (public or private), teachers' perception of their working environment, or teacher relationship with their co-workers and parents.
This study uses the positive aspects of closeness only with the instrument created by Pianta (2016) and further adapted it into the Indonesian context.This aspect of closeness describes the teacher as a safe place for students to be able to communicate openly, warmly, and harmoniously (Aldrup et al., 2018).As the study shows, there is the contribution of the teacherstudent relationship and teacher employment status to teacher subjective well-being.However, teacher-student relationships have more contribution to teacher subjective well-being than employment status.It might be the case where teachers with non-permanent status can have high teacher subjective well-being because their subjective well-being can be explained by their relationships with their students.
There are several things that can be identified as advantages of this study.First, this study is one of the first studies that examines the conditions of teacher subjective well-being in Indonesia, using the framework of teaching situation, which are school connectedness and teaching efficacy.This research is part of larger theme in exploring teacher subjective well-being in Indonesian education context.The researches that we conduct aimed to explore more the factors that affect the subjective well-being of teachers in Indonesia.Another research that has been conducted by our team, examined about the moderating role of teaching experience in the effect of teacher-student relationship to teacher subjective well-being (see Farhah, Saleh, & Safitri, 2021), Second, we managed to adapt and produce a valid and reliable instrument to measure both constructs used in this study, namely teacherstudent relationship and teacher subjective wellbeing, in the Indonesian context which could be used for further research addressing a similar topic.The findings from this study are considered justifiable and can be used to estimate the teacher population condition in Indonesia.Third, this study gives evidence that the teacher-student relationship affect teacher subjective well-being which supports the findings from previous studies on the relationship between the two.
Also, this research was conducted in the early time of pandemic COVID-19 so we have to make some methodological adjustments to adapt the situation.Because of the pandemic, all schools and universities are forced to use distance learning form to continue the learning activities.This dramatic changing situation also impact in the choice of data collection techniques.Although we planned to conduct offline and online data collecting, but in the end, we use online questionnaires.There was concern about the validity of the participants' identity, whether they really teachers or not.We anticipate the concern by matching the data they gave (full name and school where they teach) with the online data of teachers listed in the Kemendikbud teacher database.

 CONCLUSIONS
This study aims to examine whether the employment status of a teacher moderates the effect of the teacher-student relationship on teacher subjective well-being at the secondary school level.It is found that there is a significant positive correlation between teacher-student relationship and teacher subjective well-being.However, the moderation analysis shows that there is no significant moderation effect of the employment status on the effect of teacher-student relationship to the teacher subjective well-being.Therefore, the first hypothesis of the study is accepted but the second hypothesis is rejected.The absence of a moderating role of teacher employment status indicates that there is no influence of teacher employment status on the effect of teacher-student relationships on teacher subjective well-being.This finding shows that whatever the employment status of teachers, permanent or non-permanent, does not affect how they perceive their closeness to students which has an impact on teacher subjective wellbeing.
The implication from this study is that schools can provide opportunities for teachers, both permanent teachers and non-permanent teachers to improve their ability to foster closeness between teachers and students.A good relationship between teachers and students can have a positive impact on improving teacher subjective well-being.In carrying out this study, we identified a limitation related to the inequality in the distribution of the moderator group.In this study, the moderator variable is a dichotomy variable, namely permanent teachers and nonpermanent teachers.Based on the data, the ratio obtained by participants who have permanent teacher status (65.8%) higher than participants who have non-permanent status (34.2%).Although the statistical assumptions of the model used in this study are in accordance with the Hayes Macro PROCESS model, this inequality can cause disruption of the homogeneity of variance between the moderator and the independent variable (Gravetter & Wallnau, 2013).The disruption of the homogeneity of variance between the teacher employment status and teacher-student relationship can affect the interaction between the two of the moderating effect on the variable.There is a practical suggestions that can be made in future research based on the limitation of this study.Further research can ensure that the moderator group is divided equally in order to get more optimal results.

Table 3 .
Multiple regression analysis