Affective Commitment of Public-School Principals : How do Organizational Justice and Job Satisfaction Affect the Commitment?

: This research aims to examine the effect of organizational commitment and job satisfaction on high school principals’ affective commitment. The research employed a quantitative survey. The samples researched were 90 senior high school principals in Jakarta. The data were collected through a series of questionnaires. The data were then analyzed using descriptive statistics and path analysis. The results indicated that: 1) organizational commitment gave a positive direct effect on t he principals’ affective commitment; 2) job satisfaction gave a significant direct effect on the principals’ job satisfaction, and 3) organizational commitment gave a positive direct effect on the principals’ job satisfaction. The present research implies that the provincial education board is strongly suggested to keep performing fair procedural justice prescribed by regulations to maintain the principals working satisfactorily and feeling attached to their schools


Introduction
In 2045, it is predicted that Indonesia will encounter a new challenge to prepare its leading and outstanding generation, which is expected to both compete globally in social interactivity (Visaria & Dommaraju, 2019), and or bigger social transitions in varieties of business, organizations and new companies to develop in the future time (Bass, 2000;Davis, Crothers, Grant, Young, & Smith, 2012). The generation is the one whose excellent ability and skill to conquer the digital era. Preparing such a generation will deal with the quality of education nowadays. Education quality will play a role in their productive age in the future (Thang, Lim, & Tan, 2018;Yao, 2018;Burr, Caro, & Moorhead, 2002;Ozanne, 2010;Morrow-Howell & Greenfield, 2010;Kim, Lee, & Sims, 2016;Choi & Dinse, 1998).
Schools as an institution to develop skills and knowledge for the sake of providing quality education (Malik, 2018) should be able to answer this challenge of preparing for the digital era generation. School management should work with the government, parents, and society in this case. Severe and high commitment from all the parties is needed to realize visions and missions of building a quality digital era generation. The principal position plays a critical role in unifying the vision and mission of all parties.
School principals have a variety of functions to realize schools' visions and missions to build a quality generation. They play roles as educators (Albu, 2013), managers or administrators or supervisors (Andersson et al., 2015), role models (Sahenk, 2010), innovators (Athanasoula et al., 2010), motivators/supporters (Ingemarson et al., 2014), and communicators (Bakirci et al., 2012) which all included as soft skills (Ngang et al., 2015). School principals' roles have technically been formulated in Policy Number 13 of the Year 2007 by The Ministry of National Education (2007). They are required to be competent in personality, social, managerial, supervision, and entrepreneurship.
As one of the competencies to acquire, managerial competence should get particular attention. This competence is closely related to how school principals manage human resources or others. In fact, not all school principals are aware of this managerial competence. In some cases, many school principals are accused of violating in managing their schools. Such violation further indicates that the principals' commitment should be assessed and confirmed to sustain quality education. The principals need to be well-equipped with the belief that they need to remain at their schools and emotionally engage themselves in the schools. This commitment is strength for an organization to move forward as it gives organization energy to survive (Bloemer et al., 2013) and a reflection of personal point of view to support strong and safe working culture (Fruhen et al., 2019).
Some factors need to be considered to keep the affective commitment of school principals. Some of the factors that determine one person's affective commitment are organizational justice and job satisfaction. The feeling of wanting to stay in an organization undeniably has to do with justice or treatment that one person receives from an organization; in this case, the public school principals' affective commitment will be affected by the justice they receive from the schools and authorities. The feeling of being fairly treated by the organization is expected to make a good leader for school advancement. Heponiemi et al. (2011) claim that promoting justice in an organization is significant in terms of management, results, and processes.
Several studies have portrayed how organizational justice significantly affects affective commitment. Ha & Ha (2015) has proven that organizational justice affects the affective commitment of college student-athletes directly and positively; profoundly, all the justice dimensions (distributive, procedural, interactional justice) positively affect affective commitment. Similarly, Malla & Malla (2022) highlight that based on their survey of employees from various industry companies, distributive and informational justice significantly contributes to the employees' affective commitment. Another study conducted by Primawidi & Mangundjaya (2020) has shown that organizational justice primarily influences the affective commitment of employees, in which the most dominant dimension are procedural and distributive justice.
Besides organizational justice, job satisfaction is also claimed to influence affective commitment. Employees who feel satisfied with their job are predicted to be more loyal to their workplace and advance their performance and achievement. Some studies have indicated how job satisfaction may influence affective commitment. After studying several employees working in the food and beverages industry, Nugroho & Tanuwijaya (2022) concluded that job satisfaction significantly improves employees' affective commitment. Another study conducted on civil engineers has reported that job satisfaction significantly and positively influences affective commitment (Sohail & Ilyas, 2018). Wen et al. (2018) state that further signified that job satisfaction led employees to be more engaged in the workplace.
The above-listed studies have already presented interesting data on how people in their profession show their affective commitment. However, the affective commitment of public-school principals is worth researching as they act as a manager at educational institutions that attempt to create qualified human resources. Specifically, the research aims at investigating the effect of organizational justice and job satisfaction on the affective commitment of school principals of public high schools in Jakarta. The current research's

Research Method
The present research employed a quantitative approach with a survey design intending to deeply analyze the public school principals' affective commitment predicted by organizational justice and job satisfaction. It follows the idea of Fraenkel et al. (2015), that the survey method is used to define a population's characteristics. For examining the direct effect among three variables, the model was proposed: exogen variables consist of organizational justice (x1) and job satisfaction (x2), while endogen variables consist of affective commitment (y). The details are as follows.
1. Organizational justice had a positive direct effect on the affective commitment of public-school principals, formulated as: H 0 : βу1 ≤ 0 H 1 : βу1 > 0 2. Job satisfaction had a positive direct effect on the affective commitment of publicschool principals, formulated as: H 0 : βу2≤ 0 H 1 : βу2> 0 3. Organizational justice had a positive direct effect on job satisfaction of public-school principals, formulated as: H 0 : β 12 ≤ 0 H 1 : β 12 > 0 The constellation model is displayed below in Figure 1. = organizational justice X 2 = job satisfaction Y = affective commitment The research was conducted in Jakarta's public senior high school in August 2019. The accessible population was all public senior high school principals in Jakarta comprising of 117 persons. A simple random sampling technique was utilized, in which 90 samples were then assigned using the Slovin formula for sample size (Tejada & Punzalan, 2012). The calculation is as follows. n = N 1+Ne 2 n = 117 = 90,52 (then 90 principals were chosen) 1+ (117 x (0,05) 2 ) Note: n = sample size N = population e = margin on error at 0,05 significance level The detail sampling calculation based on areas in Jakarta is presented in Table 1. The data were collected using questionnaires, examining the relations among the research variables. The blueprints are listed in Table 2. The data collected from the samples were analyzed using path analysis. It was specifically used to examine the effect among variables based on a theoretical foundation. The technique used requires a causal framework as a prerequisite in path analysis. Before hypothesis testing, normality, significance, and linearity tests were applied.

Results and Discussion
The present research involved 90 public school principals. The respondent characteristics are reported in Table 3. The public school principals were selected based on the following criteria: working area, education, age, and working period. The working area of the respondents consisted of six areas covering Central Jakarta, West Jakarta, East Jakarta, North Jakarta, South Jakarta, and Seribu Islands, in which the most respondents came from East Jakarta (34.4%), and the least was from Seribu Island (1.1%). From the criterion of education level, the public school principals possess bachelor, master, and doctoral degrees, most of which come from master's degrees (71.1%) and doctoral degrees (2.2%). Based on their age, the criterion was chosen with the following range: 46-50, 51-55, and 56-60 years old, in which most of the principals were between 51-55 years old (45.6%). The last criterion was the working period, which covered 7 -17, 18 -28, and 29-39 years, most of which came from 18-28 years of working (60%).

Descriptive Analysis Description of affective commitment (Y)
The data on affective commitment had 40 valid questions. With a scale between 1-5, the theoretical score gained was between 40-200, while the empirical score was 142-189. Therefore, the score range was 47 with a mean of 167.67, median of 168.00, mode of 171.00, variance of 109.62, and standard deviation of 10.47. Based on the data above, the highest frequency was found in the class interval 163-169 with 25 respondents (27.78%), while the lowest was in the interval 142-148 and 184-190 with five respondents (5.56%). The histogram graphic of affective commitment (Y) data distribution is presented in Figure 2.

Description of organizational justice (X 1 ) variable
The data of the organizational justice variable had 38 valid questions. On a scale of 1-5, the theoretical score gained was between 38 and 190, while the empirical score was between 114 and 154. Therefore, the score range was 40 with a mean of 133.52, median of 135.00, mode of 135.00, variance of 75.26, and standard deviation of 8.68. Based on the data above, the highest frequency was found in the class interval 132 -137 with 27 respondents (30.00%), while the lowest was in the interval 150 -154 with four respondents (4.44%). The histogram graphic of data distribution of organizational justice (X1) is presented in Figure 3.

Figure 3. Histogram graphic of organizational justice (x 1 ) Description of job satisfaction (X 2 ) variable
The data of the job satisfaction variable had 34 valid questions. On a scale between 1-5, the theoretical score gained was between 34 to 170, while the empirical score was between 112 to 159. Therefore, the score range was 47 with a mean of 132.75, median of 133.00, mode of 134.00, variance of 108.50, and standard deviation of 10,42. Based on the data above, the highest frequency was found in the class interval 133 -139 with 22 respondents (24.44%), while the lowest was in interval 154 -160 with three respondents (3.33%). The histogram graphic of data distribution of job satisfaction (X 2 ) is presented in Figure 4.

Figure 4. Histogram graphic of job satisfaction (x 2 ) Normality Test
A normality test is a requirement to administer before path analysis is applied. To proceed to the path analysis test, the data should be normally distributed. In this case, the normality test was employed using Lilliefor's formula. The hypothesis is formulated as follows: H0: The data come from a normally distributed population H1: The data do not come from normally distributed population The normality test results on the three variables are reported in Table 4.  Table 4 shows that the first normality test applied is the normality of affective commitment to organizational justice (Y on X1). It gained Lcount 0.0519, and the significance is 0.0934 at α = 0,05. It is clear that Lcount ≤ Ltable, then it is assumed that the estimation error of affective commitment on organizational justice (Y on X1) comes from the normal distribution population. The second normality test is an affective commitment on job satisfaction (Y on X2). The Lcount is 0.0526, and the significance is 0.0934 at α = 0,05. It means that Lcount ≤ Ltable, then it is proven that the estimation error of affective commitment on job satisfaction (Y on X2) is normally distributed. The last is the normality test of organizational justice on job satisfaction (X1 on X2). It is seen that Lcount is 0.0624, and the significance is 0.0934 at α = 0,05. It indicates that Lcount ≤ Ltable; therefore, it is proven that the data of organizational justice on job satisfaction (X1 on X2) is from the normally distributed population.

Significance Test and Regression Linearity
The hypothesis testing in this present research was initially applied by regression analysis. The regression analysis was used to predict the relation model. The hypothesis proposed to test the significance and linearity of regression is as follows: The regression equation of Y on X1 is Ŷ= 71,475 +0,720 x1. For the significance test, it gained Fcount of 48,72, higher than Ftable 6,93 at α = 0,01. As Fcount > Ftable, the regression equation is concluded as very significant. Fcount 1,238 is lower than Ftable 1,65 at α = 0,05 for the linearity test. This indicates that the calculation result is linear.
The regression equation of Y on X2 is Ŷ= 86,423+ 0,612 x2. For the significance test, it gained a Fcount of 51,84, higher than the Ftable of 6,93 at α = 0,01. In Fcount > Ftable, the regression equation is concluded as very significant. Fcount 1,143 is lower than Ftable 1,64 at α = 0,05 for the linearity test. It indicates that the calculation result is linear.

Hypothesis Testing Goodness of Fit Test
A goodness of fit test is needed to determine whether the hypothetical model proposed is fit or consistent with the empirical data. The test was applied by comparing the theoretical correlation matrix with the empirical, theoretical matrix. Based on the calculation using Lisrel, the goodness of fit test results are as follows.
Goodness-of-Fit Statistics Degrees of Freedom for (C1)-(C2) 0 Maximum Likelihood Ratio Chi-Square (C1) 0.0 (P = 1.0000) Browne's (1984) ADF Chi-Square (C2_NT) 0.0 (P = 1.0000) The Model is Saturated; the Fit is Perfect! The calculation results showed that the value of Chi-square = 0,0 with a degree of freedom = 0 and p = 1. The value of p = 1 > 0,50 or not shows a significant difference between the theoretical and empirical matrix. Therefore, H0 is accepted, inferring that the model is suitable or matches the empirical data, then the fit is categorized into perfect.

Path Coefficient
Path coefficient calculation is administered based on the equation structure in the constellation model. The correlation coefficient at each path can be seen in Table 6.  Table 6 shows that the score of path coefficient py1 is 0,289, count = 2,76, with table = 1,99 at α = 0,05. Therefore, this indicates count> table, which means H0 is rejected, and H1 is accepted. It can be interpreted as a positive, direct, significant effect of X1 (organizational justice) on Y (affective commitment). For the path coefficient py2=0,294, tcount = 2,72, and ttable = 1,99 α = 0,05. Thus, because count> table, then H0 is rejected, and H1 is accepted. The result reveals a positive and significant effect of X2 (job satisfaction) on Y (affective commitment). The last path coefficient is p21 with a score of 0.640, count = 7.81, and table = 1,99 at α = 0,05. Hence, because count> table, then H0 is rejected, and H1 is accepted. It indicates that X1 (organizational justice) has a favorable direct, and significant impact on X2 (job satisfaction). The summary of path coefficient test is presented in Figure 5. The statistical hypotheses have calculated to investigate the effect of organizational justice on affective commitment, job satisfaction on affective commitment, and organizational justice on job satisfaction.  Table 7 exhibit the hypotheses testing that, first, P y1 = 0,289 with t count = 2,76, while t table =1,99 (α =0,05; dk = 86), then H 0 is rejected and H 1 is accepted. It infers that organizational justice variable has a positive direct effect on the principals' affective commitment. Second, P y2 shows the score of path coefficient 0,294 with t count = 2,72, and t table = 1,99 (α =0,05; dk = 86), so that H 0 is rejected and H 1 is accepted. The calculation result unveils affects positively and directly on affective commitment. Third, P 12 results in 0,640 with t count = 7,81, and t table =1,99 (α =0,05; dk = 86), then H 0 is rejected and H 1 is accepted. The result denotes that organizational justice positively directly affects the school principals' job satisfaction.

Organizational Justice Direct Effect on Affective Commitment
The hypothesis testing denoted that organizational justice had a significant direct impact on practical commitment. The correlation coefficient value was 0,597, and the path coefficient value was 0,289.
This first finding revealed that organizational justice received by the school principals determined their affective commitment. Employees' affective commitment is highly determined by organizational justice, whether their boss or authority fairly treats them (Lee & Lin, 2014). The more justice or fair treatment they feel and receive from schools' management or the authority, the more they feel attached and show affection to their job as the school principals. Colquitt et al. (2011) claim that people who had experienced much higher justice tend to get the stronger affective commitment. Additionally, support from supervisors or authorities felt by employees will also mediate the connection between organizational justice and affective commitment (Li, 2018).
In the end, as the schools'' principals were treated with justice, more engagement, better involvement, and well performance of the principals shall be attained. Positively, these significant work attitudes will certainly promote schools' quality. That is why realizing organizational justice is expected to foster a school principal to be more committed (Buluc & Gunes, 2014) and gain trust (Cui et al., 2018) to achieve school advancement. Furthermore, the statistical computation declares that the highest predictor of organizational justice is the principals receiving fair procedural treatment. The respondents whose ages were above 50 years old, with a working period of more than 18 years, and who earned a Master's degree (S2) gave good responses to procedural justice given by their organization. This finding indicates procedural justice is the leading facet affecting the school principals' affective commitment. The research of Tjahjono et al. (2020) signified a similar thing that procedural justice shows a positive effect on employees' affective commitment. Then the schools' management, or the authorities, should employ good strategies to maintain procedural justice to maintain the principals' affection for their job. Designing intervention (Primawidi & Mangundjaya, 2020) as a part of persevering such a commitment (Deressa et al., 2022) is decisive to make.

Job Satisfaction Direct Effect on Affective Commitment
From the path analysis computation, the job satisfaction variable is viewed to have the highest impact on affective commitment, with a correlation score of 0,609 and a path coefficient of 0,294. The computation result bespeaks a significant direct impact of job satisfaction on the principals' affective commitment.
The impact further describes that the school principals' affective commitment will be enhanced if they are satisfied with their job. No wonder a positive correlation exists between job satisfaction and affective commitment (Valaei & Rezaei, 2016). These two variables impact each other significantly (Sohail & Ilyas, 2018). Positive emotion emerges once these principals see that their job is successful. Perpetuating this emotion assuredly will intensify the principals' sense of affection for their job and commitment to their schools. It is underlined by Colquitt et al. (2009) that job satisfaction highly predicts affective commitment, which leads employees to stay in an organization. Al Samman & Mohammed's (2021) research has also proven the relationship between these two variables, as job satisfaction is also claimed to be a substantial predictor of affective commitment (Malla & Malla, 2022).
Speaking of the indicators to determine the main job satisfaction catalyst, the statistical computation suggests that the highest indicator to give a direct positive effect is satisfactory feeling for school achievement. Most respondents above 50 years old who have been working for 18 years old with a master's degree show very positive responses on the job satisfaction they experience. Such a satisfactory feeling may take place at the level of achievement acceptance by the work environment, including the stakeholder. School achievement, such as improving school facilities, is one of the most contented things they feel at the schools. At their age, they still can show innovation and creativity for their schools' progress. Then it is crucial to understand that age, gender, education, and work experience will significantly affect organizational commitment (Labrague et al., 2018) including affective commitment. However, it was found that such commitment does not always have a direct effect on organizational performance (Khunsoonthornkit & Panjakajornsak, 2018).

Organizational Justice Direct Effect on Job Satisfaction
The third hypothesis examines the direct effect of organizational justice and job satisfaction. The statistical computation results showed that the coefficient correlation was 0,640, and the path coefficient was 0,640. Then, the hypothesis testing concludes that organizational justice received by the school principals positively influences their job satisfaction.
This relation denotes that a much higher level of organizational justice experienced by the school principals will incontrovertibly increase their job satisfaction. Job satisfaction, as an evaluation or perception of a job, might be resulted from working experience, including treatment given by the organization, in this case, the schools and authorities. Nugraha & Mujiati (2020) have demonstrated that organizational justice significantly determines work contentment. To be more profound, the social theory exchange has shown that employees with a high perception of organizational justice will experience high job satisfaction, which ultimately impacts their performance (Mashi, 2017).
The primary determinant of organizational justice to predict this relation is procedural justice. Most school principals above 50 years old who have been working for 18 years old and holding master's degree exhibit very positive responses to the procedural treatment they received from the schools and authorities. It further portrays that their fair treatment by the authorities affects their working contentment. The result is in harmony with a study conducted by Lambert et al. (2020) and Jameel et al. (2020) that procedural justice positively relates to job satisfaction. A significant but not positive correlation between procedural justice and job satisfaction is also found (Indahyati & Sintaasih, 2019). Notwithstanding not positive relation, the importance of procedural justice cannot be disregarded. Then,