European Journal of Educational Management The Role of Perceived Servant Leadership in Schools in Predicting Loyalty to Supervisor

: Focusing on servant leadership and its outcomes are considered fairly essential in terms of its contribution to the literature. This study seeks to determine the relationships between the servant leadership of school principals and the loyalty of teachers to their principals and the constructs of loyalty to supervisors. Secondary school teachers working in Diyarbakır constitute the population of the study. 26 secondary schools were randomly selected from Diyarbakır province and the scales were applied to the teachers working in these schools. The sample consists of 202 teachers. The data were obtained with servant leadership and loyalty to supervisor scales. Validity and reliability analyzes were performed on the scales to determine whether the scales were valid and reliable or not. While analysing the data, descriptive statistics, correlation, and regression analyzes were used, respectively. It was found that the variables examined in the study had a positive and significant relationship. Regression analysis findings revealed that servant leadership predicted loyalty to supervisors and the constructs of loyalty to supervisors (dedication, extra effort, attachment, identification, and internalization) in a statistically significant way. School principals wishing to establish a sense of loyalty in schools are recommended to be aware of the positive role of servant leadership on psychological mechanisms and to have speeches and actions compatible with this type of leadership.


Introduction
It is known that leadership behaviors affect the climate of an organization. Effective leadership behaviors help exhibit positive attitudes and behaviors such as trust, satisfaction, cooperation, and solidarity. Servant leadership, one of the current leadership approaches, has a significant influence on the atmosphere and subordinates. Greenleaf (1977), who conducted research on "Servant Leadership" on a large scale, described true leadership as being willing to support and help others. This shows that school principals, who value their teachers' priorities, keep their motivation alive, support their professional development, and assume the role of servant leadership. It is also essential to exhibit servant leadership behaviors to develop a great number of positive attitudes and behaviors in organizations. Therefore, it is of utmost importance to intensify research on servant leadership and the attitudes and behaviors of subordinates who serve (Gandolfi & Stone, 2018;Northouse, 2013), expanding the theoretical framework for servant leadership. In addition, the practitioners' use of the influence of this type of leadership in organizations reveals how important such influence is for the sense of loyalty. Indeed, Koçel (2018) argues that loyal employees in organizations, apart from working for long terms in the organization, also play a crucial role in keeping the organization alive and gaining competitive power. Loyalty is an attitude that helps solve many problems that may occur in an organization.

Literature Review
The concept of servant leadership was coined and explained by Greenleaf (1970). Servant leadership is an approach in which a leader empathizes with followers, is aware of their needs, and makes extra efforts to meet these needs (Greenleaf, 1977). Servant leaders put the interests of their followers above their own and stay away from selfishness (Hale & Fields, 2007). Such leaders do not use their power to dominate others but share their power with others and help them take on responsibilities. Thus, competition in an organization is reduced while perceived justice increases (Northouse, 2013). Patterson (2003) proposes a model of servant leadership. The constructs of servant leadership within this model can be explained as follows (Aslan, 2013;Bakan & Doğan, 2013;Erkutlu, 2014;Patterson, 2003): • Agapao love: The leader is aware of each follower's desires and needs. The leader leading with Agapao love focuses on followers' basic needs, potentials, and the benefit of the organization, respectively. • Humility: The leader values followers rather than him/her and acts modestly. Humility means having a healthy personality structure and viewing oneself as no better or worse than others do. • Altruism: The leader possesses strong motivation to help followers gratuitously.
• Vision: Vision means having long-term foresight. The leader builds up the vision of the organization in line with the visions of followers. • Trust: The leader is open and honest to followers and exhibits proper behaviours. Besides, trust is the core concept of servant leadership. • Empowerment: Empowerment means empowering followers and helping them become independent.
• Service: Service means the leader thinking of the interests of his followers instead of his own and exhibiting behaviors in this direction. The servant is described as the heart of leadership.
The servant leader values the desires and needs of his followers, helps them reveal their potential and improve themselves, empowers their followers with responsibilities, and allows them to be independent (Bakan & Doğan, 2013). Servant leaders empower and motivate followers and do not prioritize their own personal interests (Aslan, 2013). Therefore, the focus of the servant leadership process is the principles based on care for others. In other words, this type of leadership offers a philosophical approach that advises leaders to be more humane and selfless. In this respect, it differs from other leadership theories and becomes unique (Northouse, 2013).
Servant leaders are concerned about their followers while also being careful not to hurt them. They attempt to meet the followers' desires and needs, value their interests, help them develop personally, and treat them fairly (Greenleaf, 1977).
What is expected from a manager is to be fair, loyal to ethical principles, honest, reliable, accessible, and respectful, interact with his followers in high quality, and deal with them individually (Erkutlu, 2015). In other words, these traits are met by servant leadership.
Loyalty expresses commitment (Koçel, 2018). Commitment is defined as internalizing the organization's goals and values and being loyal to these goals and values, being willing to work selflessly for its benefit, and having a strong internal motivation to maintain membership in the organization (Becker et al., 1995;Şenturan, 2014). In addition, commitment is explained as loyalty to an idea or person and responsibilities related to it (Celep, 2014).
Loyal employees act shoulder to shoulder with their organization even in difficult times. These employees do not flee or leave their organizations in tougher times. It is known that the sense of loyalty is an important factor in creating trust and value among people (Vuong et al., 2021).
A high level of organizational commitment is what bespeaks mutual loyalty between the employee and the supervisor (Balay, 2014). In return, high level of organizational commitment increases job satisfaction, happiness, and performance at work (Saruhan & Yıldız, 2014) and decreases the intention to leave school (Demir, 2019). Promotion and empowerment are crucial in increasing employees' loyalty (Saruhan & Yıldız, 2014). Servant leaders support the development of their followers both individually and in terms of specialization in their work. They provide opportunities for their development in organizations. They show interest in and even dedication to empowerment (Spears, 2002).
Teachers naturally encounter various administrative practices of their principals in schools. If principals are not fair in their behaviour and practices, even teachers with high levels of loyalty are likely to develop negative attitudes and behaviors towards work. In this context, Balay (2014) emphasizes that an unfair management style alienates teachers from the school. Another important issue is to trust supervisor. A servant leader is a leader who is trusted by subordinates in an organization. For, a servant leader treats individuals fairly and sincerely and establishes open communication with them (Barutçugil, 2014). Therefore, a servant leader creates an environment of trust. In an environment where trust is perceived, subordinates help each other and have no sense of doubt. This also contributes to an increase in their performance. On the contrary, in an environment where distrust is perceived, group consciousness is damaged, subordinates are suspicious of each other and tend to consider only their interests and show self-protective behaviour (Robbins & Judge, 2011). From this point of view, it is clear that servant leaders create group consciousness and do not allow distrust, thus strengthening their subordinates' loyalty.
The purpose of this paper is to determine the role of school principals' servant leadership on teachers' loyalty to their principals and the constructs of loyalty according to teachers' perceptions. Within the scope of this main purpose, the following research questions were addressed: Q1: Is there a significant relationship between the servant leadership of school principals and loyalty to supervisors and its constructs?
Q2: Does servant leadership significantly predict loyalty to supervisors? Q3: Does servant leadership significantly predict the constructs of loyalty to supervisors (dedication, extra effort, identification, internalization, and attachment)?

Research Design
In line with the main purpose of the paper, the researcher examined whether there were significant relationships between the variables, and where the relationships were statistically significant, they were shaped with the correlational model that revealed the degree and direction (Büyüköztürk, 2012). First of all, teachers' perceptions of their principals' servant leadership and loyalty to their principals were determined. Then, the relationship between dependent and independent variables was examined to observe that the variables in the study had a significant relationship. Finally, predictive analyzes were made in the study.

Sampling and Data Collection
The population of the study consists of teachers working in secondary schools located in Diyarbakır in the 2021-2022 academic year. By drawing lots from the population of the study, 26 secondary schools were chosen completely randomly. The scales were applied to the teachers working in these secondary schools. Therefore, the cluster sampling method, in which each school was evaluated as a cluster, was used as a sampling method. 207 teachers participated in the study, however; the data obtained from 5 participants revealed that most of the items were left blank or filled carelessly. As a result, statistical analyses were carried out using the data obtained from 202 participants.
In the study, the servant leadership of school principals is measured with 7 items, and the scale of loyalty to supervisor is measured with 17 items. It is seen that the total number of items on the scales is 24. In this context, Bryman and Cramer (2001) underline that the sample size should be at least 5 times the number of items, which means the number should be at least 120 and thus, a sample including 202 participants is considered sufficient enough to represent the population.
The ethical approval of this study was obtained from Şırnak University with the ethics committee decision dated 17/03/2022 and numbered 43. Two scales were applied to the participants in a form. Scores indicating the level of participation of the participants in the five-point Likert scales ranged between "1: I totally disagree" and "5: I totally agree".
The Servant Leadership Scale: This scale was first developed by Liden et al. (2008) in the form of 7 sub-scales and 28 items. Then, Liden et al. (2013) arranged the short form consisting of 7 items and one sub-scale. The short form of the scale was adapted into Turkish by Kılıç and Aydın (2016). In the current study, the fit indices produced as a result of Confirmatory Factor Analysis are as follows: X2 = 14.11, X2/df = 1.28, p = .22, GFI = .98, IFI = .99, CFI = .99, RMSEA = .03. It is seen that the fit indices are at a good level. In addition, as a result of the reliability analysis made on the data in this study, it was determined that the Cronbach-Alpha Reliability Coefficient was .90.
Loyalty to Supervisor: The scale developed by Chen et al. (2002) was adapted into Turkish by Ceylan and Doğanyılmaz (2007). In the scale, servant leadership was measured with 17 items and 5 sub-scales. The confirmatory factor analysis results applied to the data of the current study are as follows: X2 = 243.91, X2/ df = 2.25, p = .00, GFI = .88, IFI = .94, CFI = .94, RMSEA = .07. It is seen that the fit indices consist of good and acceptable values. In addition, the Cronbach's alpha reliability values calculated as a result of the reliability analyzes performed using the data of this study were .77 in the sub-scale of dedication to supervisor, .88 in the sub-scale of extra effort for supervisor, .89 in the sub-scale of attachment to supervisor, .87 in the sub-scale of identification with supervisor, .81 in the sub-scale of internalization of the values of supervisor, and .93 in the total scale.

Data Analysis
The data of this study were collected through two different scales. Therefore, obtaining data with 2 or more scales may cause common method bias. Whether this occurs or not has been examined with the "Harman's Single Factor Test", which many researchers frequently use (Aulakh & Gencturk, 2000). According to this test, exploratory factor analysis (EFA) is applied to all items in the scales at once. If a single-factor or multi-dimensional structure occurs or if the variance explained by the first dimension is over 50%, a common method bias happens (Podsakoff et al., 2003). When EFA was applied to all the items in the scales at once, a 6-factor structure occurred. The amount of variance explained by the first factor was 21.49. Considering the above explanations, there was no common method bias.
There was no reverse coded item in the scales. Research data were analyzed in terms of the assumption of normality. It was found that the data were normally distributed by looking at the skewness-kurtosis coefficients. The skewnesskurtosis coefficients of the data ranged between ±1. In the servant leadership scale, the skewness value was -.70 and the kurtosis value was .26 while in the loyalty to supervisor scale, the skewness value was calculated as -.52 and the kurtosis value as .32. In this context, Tabachnick and Fidell (2014) report that the skewness-kurtosis coefficients ranging between ±1.5 indicate that the data provides the assumption of normality.
In order to reach stronger evidence for the assumption of normality, the z-values calculated by dividing the skewness and kurtosis coefficients by the standard error are taken into account. For medium-sized samples (50 < n < 300), the zvalue calculated greater than 3.29 indicates that the data group is not normally distributed (Kim, 2013). The z-value obtained by dividing the skewness coefficient by the standard error in the servant leadership scale was -4.11 and the zvalue obtained by dividing the kurtosis value by the standard error of the kurtosis was found to be .76. In the loyalty to supervisor scale, the z-value calculated by dividing the skewness coefficient by the standard error was -3.05 and the zvalue calculated by dividing the kurtosis coefficient by the standard error was found to be 0.94. As the sample was limited to 202 and z-values were less than 3.29, it was proved that the data were normally distributed.
Calculated Tolerance (1.00 > .20), VIF (1.00 < 10) and CI (10.51 < 30) values showed that there was no problem. As Büyüköztürk (2012) reported, the correlation values of the independent variables remaining below .90 indicate that there is no multicollinearity problem between the variables. In this study, the calculated correlation values were found to be below .90.
This study investigated teachers' perceptions of school principals' servant leadership and their loyalty to their principals. It was found that the servant leadership scale and loyalty to supervisors and constructs of loyalty to supervisors had significant relationships and following the finding, the predictive analyzes were conducted. In each regression analysis, servant leadership was included in the analysis as the independent variable, while loyalty to supervisors and its constructs were included as dependent variables.

Findings
This section includes information about the findings. Thus, the findings obtained as a result of the analyses applied to the data are presented respectively.

Findings Related to Descriptive Analysis
Descriptive analysis findings for servant leadership and loyalty to supervisors and its constructs are presented in Table  1.  Table 1 show that teachers perceive school principals' servant leadership at a partially high level ( X = 3.97). Teachers also perceive loyalty to supervisors ( X = 3.83), dedication to supervisors ( X = 3.85), attachment to supervisors ( X = 3.86), identification with supervisor ( X = 3.56) and internalization of the supervisor's values ( X = 3.45) at a partially high level while teachers' perception of extra effort for supervisors is high ( X = 4.39).

Findings Related to Correlation
The correlation values obtained in the study are given in Table 2. The correlation coefficients in Table 2 show that there is a high, positive, and significant relationship between servant leadership and loyalty to supervisors (r = .77, **p < .01) and dedication to supervisors (r = .71, **p < .01). Besides, servant leadership has a moderate, positive, and significant relationship with extra effort for supervisors (r = .58, **p < .01), attachment to supervisors (r = .69, *p < .01), identification with supervisor (r = .52, **p < .01) and internalization of the supervisor's values (r = .63, **p < .01). All of the variables examined in the study have a statistically significant relationship.

Findings Related to Regression Analyses
The role of servant leadership on loyalty to supervisors is given in Table 3. Loyalty to supervisors as the dependent variable **p < .01, *p < .05. Table 3 reveal that, according to teachers' opinions, school principals' servant leadership significantly predicts teachers' loyalty to their principals (β = .77**, p < .01). 59% of the variance in the variable of loyalty to supervisors is explained by the servant behaviour of the school principals (ΔR 2 = .59; p < .01). In addition, it can be stated that 1-unit increase in teachers' servant leadership perception strengthens their loyalty to their principals by .77-unit.

The values obtained in the regression analysis in
The role of servant leadership in predicting loyalty to supervisors and dedication to supervisors has also been analysed. And the findings are presented in Table 3. Dedication to supervisors as the dependent variable **p < .01, *p < .05.
According to the values reached in the analysis in Table 4, servant leadership significantly predicts the variable of dedication to supervisors (β = .71**, p < .01). It is seen that the perceived servant leadership of school principals explains 50% of the variance in teachers' dedication to supervisors (ΔR 2 = .50; p < .01). It can be said that an increase of 1-unit in perceived servant leadership contributes .71-unit to dedication to supervisors. Extra effort as the dependent variable **p < .01, *p < .05.
According to the values reached in the regression analysis (Table 5), the perception of servant leadership significantly predicts the variable of extra effort for supervisors (β = .58**, p < .01). Teachers' perception of school principals' servant leadership explains 34% of the variance in extra effort for their principals (ΔR 2 = .34; p < .01). It can be stated that a 1unit increase in servant leadership perception reveals a .34-unit increase in extra effort for supervisors.
The role of servant leadership in predicting attachment to supervisors has been scrutinized. And related findings are given in Table 6. Attachment to supervisors as the dependent variable **p < .01, *p < .05.
Considering the analysis results in Table 6, it is seen that school principals' servant leadership significantly predicts teachers' attachment to principals (β = .69**, p < .01). 48% of the variance in attachment to supervisors is explained by servant leadership (ΔR 2 = .48; p < .01). It can be said that 1-unit increase in teachers' servant leadership perception increases attachment to supervisors by .69-unit.
The findings regarding the role of servant leadership in predicting identification with supervisors are given in Table 7. And Table 7 is interpreted based on the findings. Identification with supervisors as the dependent variable **p < .01, *p < .05.
As seen in Table 7, servant leadership predicts identification with supervisors (β = .52**, p < .01). 27% of the variance in identification with supervisors is explained by servant leadership (ΔR 2 = .27; p < .01). It can be stated that 1-unit increase in perceived servant leadership of school principals contributes to the level of identification of teachers with their principals by .52-unit.
The analysis results for the role of perceived servant leadership in predicting teachers' internalization of the supervisor's values are given in Table 8. These findings are interpreted accordingly. Internalization of the supervisor's values as the dependent variable **p < .01, *p < .05.
The analysis results (Table 8) reveal that servant leadership significantly predicts the internalization of the supervisor's values (β = .63**, p < .01). Servant leadership explains 40% of the variance in the internalization of the supervisor's values (ΔR 2 = .40; p < .01). It can be stated that 1-unit increase in the perceived servant leadership level increases the level of internalization of the supervisor's values by .63-unit.

Discussion
Servant leadership behaviors help subordinates realize their potential and become better at fulfilling their duties (Greenleaf, 1977). From this point of view, it is clear that school principals helping teachers reach their individual goals and professional development will increase their work performance too. Previous studies (Biçer, 2020;Rachel et al., 2009;Sungu et al., 2019;Vuong et al., 2021) revealed that organizational loyalty and job performance are interrelated. The activities and behaviors of school principals play an important role in teachers' loyalty to their organizations. Therefore, it is clear that servant leadership behaviors and loyalty to supervisors are related to each other. Similarly, one of the findings of this study is that servant leadership increases loyalty to supervisors.
The servant leader is sincere and hearty in his relationships (Patterson, 2003). The sincerity and openness of school principals who exhibit servant leadership behaviors naturally play a positive role in teachers. If teachers think that their interactions with school principals are of high quality, then this thought is likely to contribute to their development of a high level of commitment (Balay, 2014). In this study, it was found that servant leadership increased dedication and attachment to supervisors. Teachers are convinced that their principals are honest. Being positive about honesty helps strengthen dedication and attachment to supervisors. In this context, Robbins and Judge (2011) emphasize that subordinates do not follow leaders who they believe are not trustworthy. In addition, subordinates are of the opinion that trust in their leaders ensures that leaders will not abuse their rights and interests (Schoorman et al., 2007). Therefore, subordinates feel commitment and dedication to a leader who cares about their interests, respects their rights, and acts fairly. It is also important that the leader is altruistic in the formation of commitment and dedication. In this context, Barutçugil (2014) asserts that servant leaders think less of their interests and strive to increase their subordinates' performance and make them feel good and happy.
If the leader creates a trustworthy and fair environment in the organization, the followers become more helpful to each other. The effort exhibited by the followers is increasing (Robbins & Judge, 2011). The environment created by the leader affects the followers and contributes to the increase in their efforts. Therefore, the followers work beyond their official roles for managers who exhibit servant leadership behaviors. Coetzer et al. (2017) found that servant leadership increases employee commitment. In this study, it was found that servant leadership increased teachers' display of extra behaviors outside of their formally defined roles.
In his research to determine the most positive and negative personality traits, and to what extent these personality traits are liked or disliked, Anderson found that the most valued traits are sincerity, honesty, loyalty, realism, reliability, and connectivity, and the least valued traits are lying, dishonesty, and fraudulence (Freedman et al., 1993). The behaviour of leaders such as giving value and priority to their subordinates, and being equal and honest is at the core of servant leadership. In fact, it is known that servant leadership begins when leaders dedicate themselves to these concepts (Northouse, 2013). Considering the personality traits and behaviors of the servant leader, it can be argued that the favourite personality traits are met with servant leadership. For, as stated by Barutçugil (2014), the servant serves to be a figure of mastery that matures with high character, love for people, and benevolence. It is clear that subordinates like their behaviour and are satisfied with their behaviour and thus, they will naturally identify with their leaders. In so doing, it is possible for subordinates to adopt supervisors' values.
Identification is a type of commitment that includes being content with co-workers and trying to maintain this relationship (O'reilly & Chatman, 1986). Identification with supervisors, on the other hand, occurs when the employee has a satisfactory relationship with his supervisor and feels loyal. Liden et al. (2008) emphasize that the servant leader puts effort into recognizing and empowering his subordinates and that his relationships with his subordinates are sincere and long-lasting. In this study, it was found that the servant leadership behaviors of school principals strengthened the identification of teachers with their principals.
Internalization is a commitment to achieving harmony between individual and organizational values. It is pointed out that the best form of loyalty occurs when the organization and the employee have equal benefits (Vuong et al., 2021). An employee who has internalized the values of the manager perceives a harmony between his own values and the values of his manager (Demir, 2021). Servant leadership is a type of leadership that has positive effects on society. In addition, compliance with ethical sensitivities takes an important place in the behaviors of a servant leader (Gandolfi & Stone, 2018). Therefore, it is clear that their values will be highly compatible with people feeling the same sensitivity. Similarly, in this study, it was determined that school principals' servant leadership behaviors contributed to teachers' internalization of their principals' values.

Conclusion
The results obtained based on the findings of this study are summarized as follows: (1) there are significant and positive relationships between the servant leadership behaviors of school principals and loyalty to supervisors and its constructs.
(2) School principals' servant leadership behaviors strengthen teachers' loyalty to their principals. (3) Servant leadership behaviors contribute to teachers' attitudes and behaviors of dedication and attachment, extra effort, identification, and internalization of values. (4) The relative order of importance regarding the contribution of servant leadership to the constructs of loyalty to supervisors is as follows: dedication to supervisors, attachment to supervisors, internalization of the values of supervisors, extra effort for supervisors, and identification with supervisors. Therefore, according to teachers' views, servant leadership has the most important role in dedication to supervisors while it has the least role in identification with supervisors.

Recommendations
It is considered important that principals, who are apt to increase effectiveness and efficiency in their organization, exhibit behaviors that improve teachers' loyalty, and for this reason, they should be servant leaders at school. In addition, considering the outcomes of loyalty, it has an important role in preventing and solving problems in the organization. It is also necessary for subsequent researchers to focus on loyalty in terms of creating a broad framework for this concept, especially in educational organizations.

Limitations
There are some limitations of this research. First, the research is only based on the quantitative research method. Second, the data is obtained with the support of teachers in a province. Third, the relationships between these variables are not examined causally.