Organizational Justice and Teachers’ Turnover Intention in Primary and Secondary Schools: The Importance of Sustainable Salary Management

Abstract

Primary and middle school teachers are a critical part of China’s foundational education; teacher turnover heavily affects the stable development of foundational education. Aiming at the phenomenon of teacher turnover intention in primary and secondary schools, interviews and questionnaires were used to explore the relationship between teachers’ organizational justice, salary satisfaction, and turnover intention in the management of primary and secondary schools. The research subjects were more than 500 primary and secondary school teachers from Zhejiang and Jiangsu provinces. The data analysis results show that the sense of interactive fairness is the highest, followed by procedural fairness and the lowest sense of distributional fairness, with a certain sense of unfairness among the three dimensions of organizational fairness among primary and secondary school teachers. The average salary satisfaction score is 2.64, which is lower than the midpoint of 3, and there is a certain degree of dissatisfaction. The results of the correlation analysis showed that the pairwise correlations between the variables reached a very significant level. The results of regression analysis showed that organizational justice had the greatest impact on turnover intention, accounting for 20% of the variance. Salary satisfaction plays a partial mediating role between various dimensions of organizational justice and turnover intention, indicating that procedural fairness, interaction fairness, and distributional justice not only have a direct impact on turnover intention, but also indirectly affect turnover intention by affecting salary satisfaction. The results of this study can provide empirical data support for educational administrative departments and school managers in economically developed regions to improve their compensation management level and implement effective and sustainable compensation management policies and programs, as well as provide first-hand information for cross-country comparative studies of teacher compensation management in the field of education administration.

1. Introduction

Teacher turnover has increasingly become a great social concern in China. A survey conducted by Liu and Li found that 5.58% of primary and middle school teachers reported high turnover intention [1]. In recent years, a survey on teacher turnover conducted by the Education Bureau of a provincial capital city in an economically developed eastern area of China reported that nearly 60 senior teachers in primary and middle schools resigned during the summer vacation, twice as many as three years ago, thus generating great social concern [2]. Why did so many teachers choose to leave their posts?

1.1. Analysis of Reasons for Resignation

Based on the results of the semi-structured interviews (Approval No. 2021001) and the literature study, the reasons for primary and secondary school teachers’ intention to leave are summarized as follows:

Firstly, the absolute value of teachers’ salaries is low, and the actual total income of teachers is not high. According to the China Labor Statistics Yearbook 2016, in 2015, the average salary of teachers in China was CNY 66,592, ranking 9th among 19 categories of industry and placing it at a medium level [3]. According to the China Labor Statistics Yearbook 2020, in 2019, the average salary of teachers in China ranked 12th among the 19 categories of industry, which was slightly above the average level [4]. However, due to institutional reasons, the salary growth of most teachers was slow. In mainland China, because of the lack of government investment, school leaders have always turned a blind eye to the phenomenon of teachers mentoring students to earn money. However, in the eastern coastal areas, there is no tolerance for paid student mentoring in professional title evaluation and annual performance appraisals. Therefore, capable teachers who set up classes for students so as to earn money to make up for their insufficient income have to take great occupational risks and endure significant psychological pressure. Once reported by others, their only option is to quit; some teachers do not want to hide, and simply leave their jobs.

Secondly, the sense of fairness within the teaching industry is low. (1) The income gap between teachers with different professional titles is large. All of a teacher’s income, including their basic salary, allowance, and performance salary, is related to their professional title. Two teachers with equal teaching strength and the same degree of effort, but with different professional titles, could have a pay difference of more than CNY 1000 per month. There is a CNY 1000 income gap between the senior and the first grades, and even more of a difference between the senior and the second grades. (2) Professional title evaluation is not fair and reasonable enough. This is mainly reflected in the low proportion of senior professional title holders and the uneven distribution of professional titles. Higher educational departments allocate professional titles according to the grade and scale of a school. For example, key provincial and municipal schools are allocated more titles than ordinary rural primary and secondary schools. In some schools, professional titles are given to school leaders first, and to the teachers who are close to these leaders, making other teachers feel hopeless and unfairly treated [5,6]. (3) The unfair distribution of performance pay is another key reason for this sense of unfairness. In some schools, there are no objective and fair performance evaluation standards. The right to make performance policies and distribute performance rewards belongs to school leaders, and the distributions are more beneficial to administrators than to front-line teachers. Furthermore, the distribution results are not open and transparent. Teachers know only their own results; they are not given the basis and standard for assessment and do not know the results of leaders and others.

Thirdly, the sense of fairness among different regions is low. First of all, from the perspective of regional differences, teachers’ salaries vary greatly among regions. In 2015, the average teachers’ salaries in Beijing and Tibet were over CNY 110,000, and that in Shanghai was over CNY 100,000, while those in Henan, Hunan, and Jiangxi provinces were just over CNY 50,000, with the difference being more than double. The average teachers’ salaries in 2019 were CNY 180,000 in Beijing, CNY 140,000 in Tibet, and CNY 130,000 in Shanghai, but only CNY 70,000 in Henan, Hunan, and Jiangxi provinces [3,4]. From the perspective of industry differences, even in economically developed areas, the overall salary levels of teachers are basically close to the national average salary level; a teacher’s income does not have a comparative advantage. Compared with people with the same academic qualifications in other industries, teachers’ salaries rank lower even if the teachers have relatively higher academic qualifications. In addition, people in other industries can obtain extra pay by engaging in the same job task in their spare time, but teachers cannot. Furthermore, private schools offer generous salaries to attract outstanding teachers, so teachers sometimes choose to quit and join these schools for various reasons. Additionally, some local governments in economically developed areas have issued policies to attract outstanding teachers, which also encourages teachers to leave their posts for these areas. In summary, differences in the sense of fairness within the teaching industry are also a reason for teacher turnover.

1.2. Justice Theory

Justice theory is an important incentive theory [7,8,9]. It suggests that employees will feel happy, work hard, and have a relatively low turnover intention if they have a sense of fairness regarding the distribution of benefits according to their own judgment [10,11,12]. However, if they have a sense of unfairness, they will be resentful, work with negative attitudes, and have higher turnover intention [10,13,14,15]. Organizational justice refers to people’s feelings toward the reasonable nature of pay distribution, and it includes distributive justice, procedural justice, and interactive justice [16,17,18].

Distributive justice refers to people’s sense of justice about pay distribution [19,20]. This sense is first affected by the absolute values of salaries. The higher the actual pay, the higher its utility value, the more attractive it is to individuals, and the relatively higher the sense of justice. Additionally, it is influenced by the relative values of salaries [10]. Individuals will compare the ratio of their own income and input with those of others within the same organization [21,22,23]. If the ratio is basically the same, they should possess a sense of organizational justice. However, if there is a big difference, two outcomes are possible: One is that if an individual’s actual income is less than the income they feel they deserve, they will feel that they have suffered a loss and feel aggrieved [10,24,25,26,27]. The other is that if their actual income is more than the income they feel they should receive, they will think that they have obtained an undeserved benefit and feel guilty [10,28]. In addition, individuals will compare the ratio of their own income and input with those of individuals from other organizations with similar educational backgrounds and abilities [28,29,30]. If the ratio is basically the same, there will be a sense of interorganizational justice, but if there is a big difference, there will be a sense of interorganizational unfairness [31,32].

Procedural justice refers to people’s sense of the justice of distribution standards and procedures during the process of organizing pay distribution [10,33,34]. Some researchers have proposed that procedural justice should follow six principles: (1) the consistency rule, which means the allocation standards should be consistent even if the objectives are different or the time is different; (2) the reduction of bias rule, so as to avoid personal biases affecting the distribution of results; (3) the accuracy rule, which means the source of decision-making information should be accurate and reliable; (4) the correctability rule, which means that when a decision is wrong, there should be an error correction mechanism to correct it; (5) the representative rule, which means distribution standards and procedures should represent the will and interests of most people; (6) the moral and ethical rule, which means distribution procedures should conform to social moral norms and ethical standards [35,36,37].

Interactive justice (leader justice) refers to the degree of respect and information disclosure given to employees by superiors in the process of resource allocation [9,38,39], and it includes interpersonal justice and information justice. Interpersonal justice means that superiors should respect and be polite to subordinates during distribution processes or when deciding distribution results. Information justice refers to the fact that leaders should explain the distribution procedures and results to subordinates—that is, it refers to the degree of distribution information disclosure [40].

Distributive justice is an important predictor of personal outcome variables (salary satisfaction, satisfaction promotion, and performance evaluation satisfaction) [41,42,43]. Procedural justice is an important predictor of organizational outcome variables (organizational commitment, job satisfaction, organizational citizenship behavior, absenteeism, and turnover) [42,43,44,45,46]. Interactive justice is an important predictor of leader-related variables (trust and evaluation of authority) [9].

Therefore, this study proposes the following:

Hypothesis H1.

Organizational justice is positively related to the turnover intention of primary and secondary school teachers.

1.3. Salary Satisfaction

Salary satisfaction refers to the positive or negative emotional reactions of employees to pay and pay management systems [47,48]. Pay is an important incentive factor that can not only meet people’s life needs and make people rich, but may also be regarded as an important symbol with which to measure personal achievement, social recognition, social status, and respect [49]. Raising an individual’s pay not only improves the living standards of themselves and their families, but also offers a symbol for realizing self-value. Organizations regard pay as an incentive means to improve morale and realize organizational value. Low salaries will destroy the morale of an organization, and employees will regard them as a negation of their self-value and a kind of “insult” to the realization of organizational value. They will feel that their efforts have not been fairly rewarded, which will affect their work performance and intentions to leave. In addition to pay, the pay management system has an even more important influence on employees. This system determines the objective and fair degree of pay given to employees—that is, the degree to which the pay is consistent with individual work performance [35]. Folger and Konovsky showed that procedural justice and distributive justice had significant effects on salary satisfaction [39]. After controlling for other variables, the influence of distributive justice on salary satisfaction was explained to be 18.87%, and that of procedural justice on salary satisfaction to be 8.6%, with the influence of distributive justice on salary satisfaction being greater than that of procedural justice. A large number of domestic and foreign studies have shown that distributive justice, procedural justice, and interactive justice have a significant impact on salary satisfaction, and distributive justice has a greater impact on salary satisfaction than procedural justice [50,51,52,53].

Therefore, this study proposes the following:

Hypothesis H2.

Organizational justice is positively related to the salary satisfaction of primary and secondary school teachers.

1.4. Limitations of Previous Studies and Direction of This Study

Rewards based on work performance, especially fair rewards, can improve salary satisfaction and reduce turnover intention. Fields and others showed that distributional equity had a significant impact on turnover intention [54,55]. Liu, Ma, and Han showed that organizational justice, distributive justice, procedural justice, and interactive justice had significant predictive effects on turnover intention [56,57,58]. Ma and Kong found in their study that all dimensions of organizational justice had an indirect impact on turnover intention through mediating variables [58]. However, the above studies were not targeted at the teaching industry. Previous studies on the influence of teachers’ salary satisfaction on turnover intention only studied the influence of teachers’ personal and school organizational factors on salary satisfaction from a single perspective. In reality, primary and secondary school teachers in China are employed by government-affiliated institutions, and their salaries are paid uniformly by the government. Although the pay management of primary and secondary school teachers is affected by the distribution standards and procedures of schools, the actual income, basic salary, professional title indices, and performance salaries of these teachers are mostly influenced by national financial policies and personnel policies of education. The main salary results are determined by the state education administration and budget departments at all levels. The study of teacher payment justice and its influencing factors should comprehensively, completely, and systematically consider individuals, organizations, and national policies. Further, we should study the reasons why teachers leave through multiple dimensions, such as the existing teacher payment structure in China and its growth mechanism, organizational justice, and individual salary satisfaction, and operative and constructive opinions and suggestions should be put forward. However, this aspect of research is still to be completed.

Therefore, this study proposes the following:

Hypothesis H3.

Salary satisfaction has a mediating effect between organizational justice and the turnover intention of primary and secondary school teachers.

The hypothesis model is shown in Figure 1.

2. Method

2.1. Participants

2.1.1. Interview Survey

The interview research was approved by the Academic Ethics Committee of Jing Hengyi Institute of Education of Hangzhou Normal University (Approval No. 2021001). During the interview study phase, an informed consent form for accepting the interview was sent to each subject.

We interviewed 20 primary and middle school teachers in Zhejiang and Jiangsu provinces, with 10 being middle school teachers and 10 being primary school teachers. Among them, 10 teachers (5 were from primary schools and 5 were from junior high schools) had 10–15 years of teaching experience, and the other 10 teachers (also 5 from primary schools and 5 from junior high schools) had more than 15 years of teaching experience. The distribution of professional titles was as follows: 8 first-grade teachers and 12 senior teachers. The interview questions were as follows: (1) Have any teachers left your school recently? (2) What do you think are the main reasons for teachers in your school to quit? (3) Do you have any good ideas and suggestions to reduce the turnover rate of teachers? (4) Are you satisfied with your salary (industry comparison, absolute satisfaction evaluation)? (5) Is your salary commensurate with your workload (or can your salary reflect your workload)? A detailed inquiry was made in a timely and appropriate manner according to the respondents’ answers. Finally, we sorted the interview results, revised the questionnaire items in accordance with the interview results, and summarized the teachers’ reasons for resignation intentions in combination with the literature research.

2.1.2. Questionnaire Survey

In this study, primary and middle school teachers in economically developed areas of Jiangsu and Zhejiang provinces were investigated, and a total of 500 questionnaires were distributed, from which 485 valid ones were collected. Among the participants, 223 were primary school teachers (46%) and 262 were junior middle school teachers (54%); 40 were rural teachers, 327 were village teachers, and 118 were urban teachers; 121 were male teachers, 363 were female teachers, and 1 response to this question was missing; the mean age was 37.96 years old (SD = 7.81); 4 had a technical secondary school degree or below, 50 had a junior college degree, 417 had an undergraduate degree, and 14 had a postgraduate degree; 52 had undetermined professional titles, 104 had second-grade titles, 205 had first-grade titles, and 124 people had senior titles; 47 had teaching experience of less than 5 years, 90 had 5–10 years, 214 had 11–20 years, 133 had more than 20 years, and 1 response to this question was missing.

2.2. Measures

2.2.1. Organizational Justice Scale

The present study adopted the organizational justice scale developed by Liu [54], and its theoretical basis was derived from Adams’s theory of distributive justice [7], Thibaut and Walker’s theory of procedural justice, and Bies and Moag’s theory of interactive justice [9]. The applied scale consists of 20 items and involves three factors: procedural justice (10 items), interactive justice (5 items), and distributive justice (5 items). Before the formal survey was completed, all questionnaires were revised according to the interview results and the teachers’ actual situations (as were the scales described below). The questions were answered using a five-point rating scale ranging from “completely disagree” (1) to “completely agree” (5). The Cronbach α coefficients of procedural justice, interactive justice, and distributive justice scales were 0.96, 0.92, and 0.93, respectively.

2.2.2. Salary Satisfaction Scale

The salary satisfaction scale’s items were extracted from the Minnesota Satisfaction Questionnaire (MSQ) compiled by Weiss and other researchers in China [59,60,61]. It has seven items (e.g., “I am satisfied with the basic salary”, “I am satisfied with the bonus income”). It has a five-point rating scale ranging from “completely disagree” (1) to “completely agree” (5). Higher scores show higher levels of satisfaction. The internal consistency of the scale was found to be satisfactory (α = 0.95).

2.2.3. Turnover Intention Scale

The resignation intention questionnaire compiled by Meyer and Allen was used [62]; it has three items, such as “Sometimes I really want to leave my present job”. It has a five-point rating scale ranging from “completely disagree” (1) to “completely agree” (5). Lower scores show lower satisfaction. The scale had a sufficient reliability of 0.75.

2.3. Statistical Analysis

SPSS 20.0 was utilized for statistical analysis.

3. Results

3.1. Descriptive Statistics and Correlation Analysis

As shown in

Table 1. Descriptive statistics, reliability, and correlation analysis (N = 485).

Variable Name	M	SD	1	2	3	4	5	6	7	8
1. Age	37.96	7.81
2. Educational background	2.91	0.40	−0.03
3. Professional title	2.93	0.85	0.73 ***	0.10 *
4. Procedural justice	3.03	1.15	−0.17 ***	−0.17 ***	−0.10 *	0.96
5. Interactive justice	3.64	0.94	−0.11 *	−0.12 **	−0.02	0.73 ***	0.92
6. Distributive justice	2.85	1.17	−0.08	−0.14 ***	−0.04	0.74 ***	0.55 **	0.93
7. Salary satisfaction	2.64	1.16	−0.11 *	−0.19 ***	−0.10 *	0.73 ***	0.55 ***	0.78 ***	0.95
8. Turnover intention	2.78	1.10	−0.03	0.15 **	0.01	−0.44 ***	−0.37 ***	−0.41 ***	−0.42 ***	0.75

Notes: education code: 1 = below technical secondary school, 2 = junior college, 3 = undergraduate, 4 = above postgraduate; title code: 1 = internship, 2 = second-grade, 3 = first-grade, 4 = senior; the diagonal data represent the internal consistency reliability of the scale; * p < 0.05, ** p < 0.01, *** p < 0.001. The same notes apply below.

, for the measurement of organizational justice, 3 represented the median point, with scores higher than 3 indicating a sense of justice, scores lower than 3 denoting a sense of unfairness, and scores close to 3 showing a state of uncertainty. Among the three dimensions, interactive justice had the highest average score of 3.64, indicating a higher level of justice; procedural justice ranked second, at 3.03, which indicated uncertainty and meant no fairness or unfairness was felt; and distributive justice ranked last, recording a mean of 2.85, indicating a sense of unfairness.

The average salary satisfaction was 2.64 points, which was lower than the midpoint of 3, showing a certain degree of dissatisfaction.

The average score for turnover intention was 2.78, thus generally denoting a relatively low level of turnover intention.

The correlation coefficients between the background variables were as follows: The correlation coefficient between the age of primary and secondary school teachers and their professional titles was 0.73 (p < 0.001), demonstrating an extremely significant correlation; the correlation coefficient between academic qualifications and professional title was 0.10 (p < 0.05), displaying a significant correlation.

The correlations between the background variables and organizational justice (procedural justice, interactive justice, distributive justice) were as follows: The correlation coefficient between age and procedural justice was −0.17 (p < 0.001), demonstrating an extremely significant correlation; the correlation coefficient between age and interactive justice was −0.11 (p < 0.05), displaying a significant correlation; age showed no correlation with distributive justice. The correlation coefficient between educational background and procedural justice was −0.17 (p < 0.001), demonstrating an extremely significant correlation; the correlation coefficient between educational background and distributive justice was −0.14 (p < 0.001), showing an extremely significant correlation; the correlation coefficient between educational background and interactive justice was −0.12 (p < 0.01), showing a very significant correlation. The correlation coefficient between professional title and procedural justice was −0.10 (p < 0.05), showing a significant correlation; professional title was not significantly correlated to interactive justice or distributive justice.

The correlations between the background variables, organizational justice, and salary satisfaction were as follows: The correlation coefficient between age and salary satisfaction was −0.11 (p < 0.05), showing a significant correlation; the correlation coefficient between educational background and salary satisfaction was −0.19 (p < 0.001), showing an extremely significant correlation; the correlation coefficient between professional title and salary satisfaction was −0.10 (p < 0.05), showing a significant correlation. The correlation coefficients between procedural justice, interactive justice, and distributive justice and salary satisfaction were 0.73 (p < 0.001), 0.55 (p < 0.001), and 0.78 (p < 0.001), respectively.

The correlations between background variables, organizational justice, salary satisfaction, and turnover intention were as follows: Age and professional title were not correlated with turnover intention. The correlation coefficient between educational background and turnover intention was 0.15 (p < 0.01), showing a very significant correlation. The correlation coefficients between procedural justice, interactive justice, and distributive justice and salary satisfaction were −0.44 (p < 0.001), −0.37 (p < 0.001), and −0.41 (p < 0.001), respectively, with all showing extremely significant correlations. The correlation coefficient between salary satisfaction and turnover intention was −0.42 (p < 0.001), showing an extremely significant correlation.

3.2. Cumulative Percentage Analysis

As shown in

Table 2. Cumulative percentage analysis of variables (N = 485).

Cumulative Percentage (%)	1	2	3	4	5
Independent variables (IV)
Procedural justice	3.9	23.5	53.2	79.2	100
Interactive justice	1.6	5.8	28.9	69.5	100
Distributive justice	8.7	30.7	61.9	85.2	100
Mediate variables (MV)
Salary satisfaction	11.3	36.9	65.8	87.2	100
Dependent variables (DV)
Turnover intention	10.3	30.1	65.6	87.8	100

, from the perspective of the three factors of organizational justice, 61.9% of teachers reported a sense of unfairness with regard to distributive justice, of which 8.7% had an intense sense of unfairness and 22% had a relatively intense sense of unfairness; in terms of procedural justice, 53.2% of teachers had a sense of unfairness, of which 3.9% had a strong sense of unfairness and 19.6% had a relatively strong sense of unfairness; in terms of interactive justice, 28.9% of teachers had a sense of unfairness. Only 14.8% had a relatively strong sense of distributive justice, 20.8% had a relatively strong sense of procedural justice, and 30.5% had a relatively strong sense of interactive justice.

From the perspective of salary satisfaction, 65.8% of teachers reported being dissatisfied with their pay, of which 11.3% were very dissatisfied, 25.6% were relatively dissatisfied, and 28.9% were slightly dissatisfied. In total, 12.8% reported being satisfied with their pay.

From the perspective of turnover intention, 34.4% of teachers reported the intention to resign, within which 12.2% had a very high intention and 22.2% had a relatively high intention.

3.3. Regression Analysis of Impact of Demographic Variables, Organizational Justice, and Salary Satisfaction on Turnover Intention

In order to further examine the influence of demographic variables (age, educational background, professional title), organizational justice, and salary satisfaction on turnover intention, the stepwise multiple regression method was adopted. As shown in

Table 3. Regression analysis of impact of demographic variables, organizational justice, and salary satisfaction on turnover intention (N = 485).

	Beta	t	ΔR2	R2	F-Value
Organizational justice	−0.33	−5.05 ***	0.20	0.22	42.20 ***
Salary satisfaction	−0.16	−2.45 **	0.01
Age	−0.11	−2.60 **	0.01

Notes: The regression coefficients in the table are standardized regression coefficients; ** p < 0.01, *** p < 0.001. The same notes apply below.

, organizational justice, salary satisfaction, and age were included in the regression equation, and all three had very significant negative predictive effects on turnover intention, effectively explaining 20%, 1%, and 1% of the variation in turnover intention, respectively. Further, the total variance explanation rate of the standardized regression equation reached 22%.

3.4. Analysis of the Mediating Role of Salary Satisfaction between Organizational Justice and Turnover Intention

As shown in

Table 4. Analysis of the mediating role of salary satisfaction between organizational justice and turnover intention.

Equation		β	R2	F
Model 1 (1)	Procedural justice → Salary satisfaction	0.73 ***	0.53	551.19 ***
(2)	Procedural justice → Turnover intention	−0.44 ***	0.20	118.60 ***
(3)	Procedural justice → Turnover intention	−0.30 ***	0.22	66.59 ***
	Salary satisfaction → Turnover intention	−0.20 ***
Model 2 (1)	Interactive justice → Salary satisfaction	0.55 ***	0.31	213.48 ***
(2)	Interactive justice → Turnover intention	−0.37 ***	0.13	75.04 ***
(3)	Interactive justice → Turnover intention	−0.19 ***	0.20	60.95 ***
	Salary satisfaction → Turnover intention	−0.31 ***
Model 3 (1)	Distributive justice → Salary satisfaction	0.78 ***	0.61	747.07 ***
(2)	Distributive justice → Turnover intention	−0.41 ***	0.17	98.86 ***
(3)	Distributive justice → Turnover intention	−0.22 ***	0.19	58.14 ***
	Salary satisfaction → Turnover intention	−0.25 ***

Notes: The β in the table serves as the standardized regression coefficient. *** p < 0.001. The same notes apply below.

, procedural justice was taken as an independent variable. Equation (1) revealed that the regression coefficient of procedural justice to salary satisfaction was 0.73 (p < 0.001), reaching an extremely significant level; Equation (2) showed that the regression coefficient of procedural justice to turnover intention was −0.44 (p < 0.001), reaching an extremely significant level; Equation (3) showed that turnover intention regressed to procedural justice and salary satisfaction simultaneously, and the regression coefficient of the intermediary variable (salary satisfaction) was −0.20 (p < 0.001), reaching an extremely significant level. The regression coefficient of the independent variable (procedural justice) decreased, and the three conditions of the intermediate variable were all met. The regression coefficient of the independent variable in Equation (3) was −0.30 (p < 0.001), reaching an extremely significant level, which indicated that the mediating variable (salary satisfaction) played only a partial mediating role. The independent variable (procedural justice) not only had a direct effect on turnover intention, but also had an indirect effect through the intermediary variable (salary satisfaction).

As shown in Table 4, interactive justice was taken as an independent variable. Equation (1) showed that the regression coefficient of interactive justice to salary satisfaction was 0.55 (p < 0.001), reaching an extremely significant level; Equation (2) showed that the regression coefficient of interactive justice to turnover intention was −0.37 (p < 0.001), reaching an extremely significant level; Equation (3) showed that turnover intention regressed to interactive justice and salary satisfaction simultaneously, and the regression coefficient of the intermediary variable (salary satisfaction) was −0.31 (p < 0.001), reaching an extremely significant level. The regression coefficient of the independent variable (interactive justice) decreased, and the three conditions of the intermediate variable were all met. The regression coefficient of the independent variable in Equation (3) was −0.19 (p < 0.001), reaching an extremely significant level, which indicated that the mediating variable (salary satisfaction) played only a partial mediating role, and the independent variable (interactive justice) not only had a direct effect on turnover intention, but also had an indirect effect through the intermediary variable (salary satisfaction).

As shown in Table 4, distributive justice was taken as an independent variable. Equation (1) showed that the regression coefficient of distributive justice to salary satisfaction was 0.78 (p < 0.001), reaching an extremely significant level; Equation (2) showed that the regression coefficient of distributive justice to turnover intention was −0.41 (p < 0.001), reaching an extremely significant level; Equation (3) showed that turnover intention regressed to distributive justice and salary satisfaction simultaneously, and the regression coefficient of the intermediary variable (salary satisfaction) was −0.25 (p < 0.001), reaching an extremely significant level. The regression coefficient of the independent variable (distributive justice) decreased, and the three conditions of the intermediate variable were all met. The regression coefficient of the independent variable in Equation (3) was −0.22 (p < 0.001), reaching an extremely significant level, which indicated that the mediating variable (salary satisfaction) only played a partial mediating role, and the independent variable (distributive justice) not only had a direct effect on turnover intention, but also had an indirect effect through the intermediary variable (salary satisfaction).

4. Discussion

This study verified that organizational justice, salary satisfaction, and age have a significant impact on the turnover intention of primary and middle school teachers. The stronger the sense of organizational justice and salary satisfaction, the weaker the turnover intention; further, the older the teacher, the weaker the turnover intention. Among these factors, the sense of organizational justice was found to have the greatest influence on turnover intention, and the variance explained reached 20%.

Salary satisfaction was found to play a partial mediating role between the factors of organizational justice and turnover intention, demonstrating that procedural justice, interactive justice, and distributive justice not only have direct impacts on turnover intention, but also have an indirect impact on turnover intention by affecting salary satisfaction.

4.1. Analysis of the Phenomenon of Low Satisfaction and Justice Caused by the Current Pay System in Primary and Secondary Schools in Reality

To understand teachers’ salary satisfaction, we must first analyze the drawbacks of the existing pay system for primary and secondary school teachers in China. The salary structure for teachers includes three parts: their basic salary, subsidy, and performance salary.

4.1.1. Teachers’ Basic Salary Is Low

Basic salary is also called post-rank salary, and it includes post salary and pay grade salary. It is a two-dimensional wage system introduced in 2006 and is based on position and rank. Post salary is determined by the professional rank of the teacher. Pay grade salary is determined by the number of years since the teacher was last promoted to a higher professional position. Teachers’ existing professional ranks in secondary/primary schools include senior teachers, vice-senior teachers, first-grade teachers, second-grade teachers, and third-grade teachers. However, teachers with senior titles are rare due to the limited quotas assigned. For this reason, most teachers end their careers with a vice-senior title. In other words, there are nominally five grades, but in reality, there are only three: vice-senior teachers, first-grade teachers, and second-grade teachers.

Teachers’ Professional Title Promotion Is Slow

Teachers can not be promoted until they have completed a fixed number of years. According to the basic standards for the assessment of teachers’ professional titles in primary and secondary schools, undergraduates can be assessed as second-grade teachers once they have become regular teachers after one year of probation. Teachers with a bachelor’s degree need to have been engaged in teaching for more than 4 years in second-grade positions before they are qualified to compete for a first-grade title, and these teachers must then have been engaged in teaching for more than 5 years in first-grade positions before they are qualified to compete for the senior titles. It takes at least 10 years for a teacher to be assessed as a senior teacher—that is, for them to be simply qualified to compete for the assessment. That is to say, a teacher with enough years of teaching may still be far from a senior title. Because there is no such post in some places, a teacher who has attained his or her desired title may not be employed based on this title.

Ratio of Professional Titles Is Low

Education personnel departments determine the proportion of teachers holding each professional title in a school according to school level (provincial and municipal standardized schools, municipal general schools, county-level standardized schools, county-level general schools, central schools, primary schools), the school scale, the number of teachers, etc., and then verify the quotas of each professional title in the school. Once there are no positions left within the quota, a new round of professional title assessments cannot be carried out until the older teachers retire and the positions become vacant. Therefore, although a teacher may be qualified to compete for the assessment, they may not be able to compete for a higher-level professional title immediately. This may also depend on whether their institution has vacant positions within their quota or not. In the process of assessment, in addition to extraordinary teachers, more elderly teachers are generally given priority; it is often said that it is a must for teachers to have enough years of teaching to be qualified to compete for higher professional titles. Among the indicators of assessment, seniority is heavily weighted, and many teachers have endured hardships all their lives but have no chance to obtain senior professional titles. The proportion of teachers with middle and low professional titles is high, and their basic salary, subsidies, and performance salaries are all paid by the financial department based on professional titles. Thus, if they can not attain higher professional titles, it is hard for them to raise their overall salary level. Although there are senior and vice-senior professional titles, the proportion of these titles is lower, so most teachers still have little chance of promotion.

The High Cost of Living in Economically Developed Areas Is Not Taken into Account

After a teacher’s post rank is determined, their basic salary (post salary and pay grade salary) is determined according to the national unified Comparison Table of Teachers’ Salary Grading Standard. The advantage of this is that differences between economically developed areas and underdeveloped areas are leveled, but the problem is that the high costs of living in big cities and coastal economically developed areas are not taken into account.

Teachers’ Salary Levels Are Fixed and Room for Growth Is Lacking

According to the Comparison Table of Teachers’ Salary Grading Standard, post salaries and pay grade salaries are fixed, which means teachers’ salary levels cannot increase with inflation and increasing local fiscal revenue. After several years, the basic salary, which originally accounted for most of a teacher’s income, has become a fraction of their income, with it accounting for only one-fifth of teacher income in provinces such as Beijing, Tianjin, and Zhejiang [63]. Although the Comparison Table was adjusted once in 2015, it quickly began to lag behind economic development.

4.1.2. The Granting of Allowances and Subsidies to Teachers Lacks Institutional Guarantees

Allowances and subsidies include seniority allowance; class hour allowance; head teacher allowance; and special allowances for excellent teachers, backbone teachers, academic leaders, and other outstanding teachers. In some places, in order to make up for the shortages of the basic salary, certain bonuses will be paid in the form of additional allowances and subsidies. Since a large part of allowances and subsidies are not included in the fiscal budget in the form of fixed items, this proportion of a teacher’s income is uncertain.

4.2. Analysis of Professional Titles, Organizational Justice, Salary Satisfaction, and Turnover Intention of Primary and Secondary School Teachers

The correlation coefficient between the ages of primary and secondary school teachers and their professional titles was 0.73 (p < 0.001), demonstrating an extremely significant correlation; the correlation coefficient between academic qualifications and professional titles was 0.10 (p < 0.05), displaying a significant correlation. Age is highly correlated with professional title. According to China’s education and personnel policy, teachers must complete the legally required number of working years before they are qualified to compete for professional title assessment and higher titles. Moreover, it takes a long time for a young teacher to continue to improve his or her educational ability and quality after graduation from a university until he or she becomes an excellent teacher. In addition, schools often take an overall sense of justice into consideration, and professional title assessment also favors older teachers. There is a significant correlation between academic qualifications and professional title. Teachers with high academic qualifications have both relatively better teaching ability and better theoretical research ability, so they are more likely to win in professional title assessments.

In this study, for primary and secondary school teachers, among the three factors of organizational justice, interactive justice scored 3.64, and the sense of justice reached a high level. This shows that in the process of pay distribution, school leaders are respectful and polite to teachers when determining the distribution results; leaders can basically treat subordinates equally and without prejudice; teachers’ work performance can be recognized by leaders, and teachers can get support and help from leaders with their work; and leaders are willing to listen to teachers’ opinions. Procedural justice scored 3.03, and the sense of justice was low, reaching an uncertain level. This indicates that the majority of teachers are not clear about the distribution standards and procedures of schools; the distribution processes are insufficiently open and transparent; teachers are not quite clear whether distributions are carried out according to the system and standards; teachers do not know whether the distributions are fair and equal; distribution standards and procedures are decided by the school administration without extensive consultation with teachers; school leaders do not care much about teachers’ ideas on distribution, therefore thinking there is no need to communicate; and when distribution processes or results find that teachers have not been satisfactory, leaders rarely come forward to give explanations. Distributive justice scored 2.85, meaning there was a certain sense of unfairness. This indicates that the majority of teachers believe that compared with their workload and responsibilities, their salary levels are unreasonable and do not reflect their hard work; the ratio of their income to their contributions is obviously different from that of their colleagues, thus not adequately demonstrating their contributions to their respective schools; compared with people with similar educational backgrounds and ability in other organizations, teachers’ income levels are obviously different. According to a survey conducted by Pan on perceptions of organizational justice among primary and middle school teachers in Chongqing and Chenzhou, distributive fairness scored 3.35, reaching a high level, which was inconsistent with the results of this study; meanwhile, procedural justice scored 3.14, which was close to the results of this study [64]. In Yang’s survey of primary school teachers in Anshun, Guizhou, procedural justice scored 3.77, interactive justice scored 3.75, and distributive justice scored 3.41, all reaching high levels, which was obviously different from the results of this study [65]. The following conclusion can be drawn: In addition to more negative perceptions of interactive justice among primary and middle school teachers in coastal areas, these teachers also perceive the levels of procedural justice and distributive justice to be low, while inland teachers tend to have a stronger sense of justice. This may be related to the high costs of living in coastal areas. Teachers have no other source of income than their wages, and they are not allowed to give students remedial classes or attend training schools to work part-time to increase their income. Coastal teachers belong to low-income groups in their local areas, and they pay more but receive less. Therefore, they are more likely to feel a sense of unfairness. However, the further inland a teacher lives, the lower their cost of living. In addition to salaries, teachers in inland areas can give students remedial classes and attend training institutions and private schools to work part-time to increase their income. Inland schools and education administrative departments are not strict in this respect. Therefore, teachers in these areas belong to the class of middle- and high-income earners, and their income can enable them to lead a decent life; therefore, they are more likely to have a relatively strong sense of justice.

A total of 61.9% of teachers reported a sense of unfairness with regard to distribution, of which 8.7% had an extremely strong sense of unfairness and 22% had a relatively strong sense of unfairness. Among the three factors of organizational justice, primary and secondary school teachers’ sense of unfairness with regard to distributive justice was the most prominent. Many people in society think that teachers’ work is relatively easy compared with that of ordinary wage earners. Teachers are better paid than the working class and have three months of winter and summer vacations every year. All these benefits have made the profession attractive to others. Therefore, in other people’s view, it is teachers’ own desires that lead to their low sense of distributive justice. In fact, most people’s views on teachers are simply superficial and lack in-depth investigation and understanding of the educational profession and of teachers. Since teachers have high levels of education, they often invest more time in educational activities, but the hours that can be calculated as class hours are lower. According to the National Education Supervision Report 2008 (Abstract) [66], the average weekly working hours of primary and junior high school teachers in China’s compulsory education stage amount to 42.4 h. The results of actual research report much higher numbers. Although the weekly class hours of primary school teachers amount to only 8.3 h, the average weekly workload actually reaches 53.8 h, and the average daily workload is about 10.76 h [67]. To ensure distributive justice, it is necessary to make the distribution of salaries and welfare fair and reasonable and develop a rational performance pay distribution scheme that enables the ratio of a primary or secondary school teacher’s income to input to be basically the same as those of their colleagues, thus reducing their sense of unfairness. In addition, we should keep the income of primary and secondary school teachers basically consistent with those of employees in other industries, so as to create a sense of justice among different industries. The current challenge is determining how to raise the salary levels of teachers (especially in economically developed areas).

In the study, 53.2% of teachers reported a sense of unfairness with regard to procedural justice, of which 3.9% had a very strong sense of unfairness and 19.6% had a relatively strong sense of unfairness. Procedural justice mainly involves the assessment of professional titles, the calculation of class hours, and the distribution of performance pay. Although assessments in many schools appropriately favor excellent teachers, seniority is favored more generally. Backbone teachers who have been teaching for about 10 years generally have attained first- and second-grade titles. Some schools, because they do not have quotas for higher professional titles, not only will not hire teachers who have successfully been granted these titles, but also do not provide chances to compete for assessment at all. Schools calculate teachers’ class hours by multiplying the actual class hours on the schedule by the number of weeks of the semester. Instances of temporary overtime, such as extracurricular activities, theatrical performances, and competition guidance, do not count toward this workload. When developing specific pay distribution schemes, assessment often favors cadres, veteran comrades, and teachers with high professional titles. As a result, while young and middle-aged teachers undertake a large number of teaching tasks, they are paid the least. According to the Teachers’ Blue Book: Report on the Development of Primary and Secondary School Teachers in China (2014) [21], a survey of Beijing showed that teachers’ salaries are obviously affected by their years of teaching. Among teachers with an annual income of less than CNY 60,000, those with 1–3 years of teaching were the most common, accounting for 61.11%; for those with CNY 60,000–80,000, teachers with 4–10 years of teaching experience were the most common, accounting for 51.67%; and for those with more than CNY 80,000, teachers with more than 21 years of teaching experience were the most common, accounting for 27.23%. To improve procedural justice, schools should maintain consistency in their resource allocation processes, reduce decision-making biases, raise their decision-making accuracy, and design correction procedures for incorrect results to ensure that the assessment results represent the opinions of most people and conform to social, moral, and ethical standards.

In this study, 28.9% of teachers reported a sense of unfairness with regard to interactive justice, of which 5.8% had a relatively strong sense of unfairness. In order to improve the level of interactive justice, it is necessary to enhance interpersonal justice and information justice. In the process of resource allocation, leaders of primary and secondary schools must focus on respect during interpersonal communication with teachers, make public announcements about their decision-making processes, strengthen their interpretation and communication skills, and improve the degree of information disclosure.

In this study, the average salary satisfaction was 2.64, showing a certain degree of dissatisfaction. A total of 65.8% of teachers reported being dissatisfied with their salaries, of which 11.3% were very dissatisfied, 25.6% were relatively dissatisfied, and 28.9% were slightly dissatisfied. Only 12.8% were satisfied. Jiang’s survey of primary school teachers in a poor county showed that the average salary satisfaction was 3.24 points [68]. Lai found that the average salary satisfaction of middle school teachers was 2.74 points [69]. The salary satisfaction of urban primary and secondary school teachers is at an unsatisfactory level. However, the satisfaction in remote and poor areas is close to 3, meaning that teachers in these areas are neither satisfied nor dissatisfied, which may be related to the low local actual consumption levels. Generally speaking, the salary satisfaction of primary and secondary school teachers needs to be further improved.

Turnover intention scored 2.78, generally showing a relatively low level of turnover intention. However, in terms of the ratio, 34.4% of teachers reported turnover intention, of which 12.2% had a very strong intention and 22.2% had a relatively strong intention. Government departments and education administrative departments should understand that such a high proportion of teachers with turnover intention, once transformed into actual turnover behavior, will inevitably affect the development of local basic education.

Low income and high pressure are the main reasons why teachers in public schools quit their jobs. This can be further analyzed via the following three points: First, teachers face great financial pressure. In terms of housing, teachers’ welfare housing has been canceled, house prices have risen sharply, and rent prices have risen rapidly. In large and medium-sized cities, newly recruited teachers who do not have housing or family support cannot afford to buy or rent a house on their own salary, which brings great financial pressure. In terms of salaries, since the implementation of performance-based salaries in 2008, teachers’ income mainly depends on professional titles and class hours. Schools cannot charge school selection fees or hold extracurricular training courses. Without a source of income, many benefits for teachers that could have been paid have now been canceled. Teachers’ nominal wages have increased, while real income has decreased. There is no regular adjustment mechanism for teachers’ salary income with economic development, and their salary income has not increased in over 10 years. Even worse, due to the impact of inflation, teachers’ income has actually declined. Second, teachers’ exit costs are low. Since teachers’ endowment insurance has been integrated into social endowment insurance, the great attraction of “officially budgeted posts” is fading. The retirement welfare guarantee, which used to be an “iron rice bowl”, has also been completely socialized. After this integration with the social pension, the costs of teacher turnover have been cut, and the barriers to exit have been greatly reduced. Third, teachers’ workloads are heavy. Teachers in primary and secondary schools invest much time in their work, shoulder heavy responsibilities, and face great work stress. In addition to preparing lessons, giving classes, and correcting homework and examination papers, many nonteaching “invisible work” tasks consume a great amount of time and energy. These include helping to maintain students’ mental health; organizing students’ participation in competitions; giving extracurricular guidance; writing papers; completing teaching and scientific research; competing for professional titles; making home visits; taking part in political learning, meetings, training, and teaching and research activities; contacting parents daily by phone and WeChat; and dealing with a wide variety of inspection and acceptance tasks. With the implementation of the two-child policy, a new problem has arisen, namely the “shortage of teachers for maternity”. Many teachers are now leaving their posts to give birth to a second child or take care of their children, resulting in a shortage of teachers. In some schools, there is a centralized or dramatic shortage of teachers in some subjects. Since the officially budgeted posts are limited, class hours can only be allocated to other in-service teachers, leading to an increase in these teachers’ workloads.

4.3. Theoretical Contribution

4.3.1. The Influence of Organizational Justice and Salary Satisfaction on the Turnover Intention of Primary and Secondary School Teachers

In this study, it was found that the influences of organizational justice and salary satisfaction on variation in the turnover intention of primary and secondary school teachers were 20% and 1%, respectively. Compared with salary satisfaction, organizational justice was a more important influencing factor. Regarding research on the relationship between organizational justice, salary satisfaction, and organizational outcome variables, Song studied the relationship between salary satisfaction and organizational justice and teacher engagement and concluded that salary satisfaction played a mediating role between distributive justice, procedural justice, and teacher engagement [70]. This study analyzed the mediating role salary satisfaction plays between organizational justice and turnover intention, and it found that salary satisfaction played a partial mediating role among the three factors of organizational justice and turnover intention. Previous studies have not analyzed the mediating role salary satisfaction plays between organizational justice and turnover intention; therefore, this study has expanded and enriched the research theory on organizational justice.

4.3.2. The Influence of Organizational Justice on Negative Outcome Variables

According to Konovsky, procedural justice and distributive justice have a significant influence on turnover intention, and distributive justice has a greater influence on turnover intention than procedural justice [71]. This study discovered that procedural justice, distributive justice, and interactive justice had significant influences on the turnover intention of primary and secondary school teachers, and their influences could be arranged from strong to weak as follows: procedural justice, distributive justice, and interactive justice. According to research on the influence of organizational justice on turnover intention within China, procedural justice has been found to have a greater impact than distributive justice, which is basically consistent with the conclusion of this study [72,73]. This may be because in the context of Chinese culture, institutional power imbalances are greater than those in the West [74,75], so procedural justice has a greater impact on employees. Another reason may be that the pay distribution methods and amounts in China’s schools are fixed, and there is a lack of flexible distribution mechanisms, resulting in less impact on distributive justice.

4.4. Practical Implications

From the perspective of economic growth, the international standard can be referred to. According to the report of the Organization for Economic Cooperation and Development (OECD), the wage index of primary and secondary school teachers with 15 years of teaching experience in developed countries is between 1.2 and 1.3 compared with the GDP growth rate. In the Chinese context, teachers’ income should be slightly higher than the average wage levels of people with the same educational backgrounds in society, and the proportion of teachers’ income in education recurrent expenditure should be increased. In addition, the proportion of senior professional titles of primary and secondary school teachers should be increased, and promotion prospects for professional titles should be improved.

In terms of procedural justice, pay distribution standards should be developed by soliciting teachers’ opinions, and the distribution process should be open and transparent. As for distributive justice, the front-line teachers should be assured that the distribution results can reflect their ability and efforts. In terms of interactive justice, teachers’ work performance should be recognized and commended by school leaders.

From the perspective of individual teachers, different references can produce a completely different sense of justice. By rationalizing the reference objects for income comparisons, a relatively stable sense of justice can be fostered. As a sense of justice is a subjective feeling, teachers can adjust themselves cognitively to obtain such a sense. From the nature of the work, teachers should be aware that education is a kind of service labor and that some types of work input and performance are tangible, while many are intangible. It is difficult to achieve absolute fairness through the quantitative measurement of distribution standards. In a nutshell, distribution standards cannot be tailored for every teacher but should be relatively fair for the majority of teachers.