Karizmatik Liderlik, Lidere Güven ve Duygusal Emeğe Yönelik Öğretmen Görüşleri: Trabzon İli Örneği

Article Details This research aimed to examine the relationship between charismatic leadership, emotional labor and trust in the leader, according to teacher views. The sample of the research consists of 522 teachers working in public primary, secondary and high schools in all districts of Trabzon province in the 2020-2021 academic year. The data were collected using the Emotional Labor Scale, Conger-Kanungo


Introduction
Administrators, teachers, students and parents, the significant stakeholders of education, are in constant communication and interaction in the school environment.Teachers are at the center of this interaction process, as well.Teachers, that play a key role in the successful conduct of education and training, are expected to exhibit behaviors with controlled emotions throughout the process.While teachers exhibit these behaviors, which are referred to as emotional labor, they are in need of self-emotion management in the context of the school's goals.Hargreaves (1998) compared this self-emotion management to the heart of teaching by personalizing it.He stated that it is important to understand children's emotions and exhibit appropriate emotional labor behavior during teaching process.
Emotional labor is the employee's management and regulation of his or her emotions in order to obtain a certain response (Hochschild, 1983).This emotional management becomes more prominent in professional groups where interpersonal interaction is intense.The teaching profession is one of these professional groups.In this profession, which requires emotional labor behavior at every stage of the educational process, positive emotions are expected to be used to ensure effective teaching with students (Hargreaves, 1998).
That teachers are happy to see the achievements they offered to students were realized, and on the contrary, that teachers feel disappointment in cases where the gains were not achieved can be presented as an indicator of the emotional dimension of teaching.
On the other hand, it has been discovered in various studies that some personal and organizational factors predict teacher emotional labor.In studies dealing with personal factors, it has been seen that the variables of emotional intelligence (Savaş, 2012;Çarıkçı, 2018), wellbeing (Bakar, 2018;Ertürk et al., 2016), job satisfaction (Cheung & Tang, 2011), job satisfaction (Zhang & Zhu, 2008), trust (Tuti, 2021) and burnout (Lee & Chelladurai, 2018;Zhang & Zhu, 2008) have been studied.In studies addressing organizational factors, variables such as organizational support (Kaçamak, 2019), organizational policy perceptions (Akbaş & Bostancı, 2019), managerial support (Ertürk, et al., 2016) and leader-member interaction (Cankal, 2020) have been studied.Apart from these, it is known that leadership has a significant impact on emotional labor (Gürer & Çiftçi, 2019;Özdemir & Koçak, 2018;Şenel & Aydoğan, 2019;Tuti, 2021;Uzun, 2021).During a possible crisis, they allow the problem to be solved by taking teachers' opinions and reflecting them to the management.As Adıgüzel (2016) states, listening to the employee's thoughts and feeling that the employee has contributed to the solution process positively affects their trust in the management.Leader school administrators ensuring organizational synergy should emerge in schools and that this is reflected positively in the organization.In addition, this creates an environment for supporting teachers' motivations by paying attention to their wishes and expectations during the education process.In this way, high performance can be achieved in the organization, and the individual and the school can achieve academic success.One of the types of effective leaders who achieve this by establishing trust within the organization is charismatic leaders.
Charismatic leader is the person who is distinguished from other people due to the reason the individual's quality and value have rare characteristics, and who has the power to make people follow him wholeheartedly by giving them a sense of faith and trust with his approach to society (Weber, 1995).Charismatic leaders have characteristics that build confidence, inspire, guide, create a positive perception of the future, and by broadening the horizons of their followers, enable them to realize the truly important things in their own lives (Uyguç et al., 2000).These characteristics of charismatic leaders have an impact on many variables in organizational processes.This effect can be explained by the positive correlation of correlational studies implemented between charismatic leadership and organizational commitment, organizational citizenship, organizational culture, team integrity, employee performance and motivation variables (Aykanat, 2010;Çelik, 2011;Işık, 2016).
It is possible to say that another variable that has a significant impact on emotional labor with charismatic leadership features is "trust".The increase in studies on the relationship between organizational trust and leadership in the last decade is evidence that it is an area that has gained importance.If an employee does not feel safe in his own organization and cannot establish adequate relationships with other employees, this will cause negativity in organizational processes.(Braun et al., 2013;Tuti & Ada, 2019).The task of eliminating these negativities falls on the leading school principals in educational organizations.In this context, the unique aspect of the research is that it focuses on the predictive power of charismatic leadership and trust 167 in the leader on emotional labor.Examining this focal point has several important contributions to the literature.First of all, the charismatic leadership behaviors exhibited by principals, especially in crises, are very important for the creation of a solution-oriented school culture and the continuity of organizational motivation.Understanding the factors that negatively impact school culture can help identify ways to support them in this role and promote positive outcomes for their school because the trust in school principals who are leaders is an important predictive variable of well-being of employees (Ceyanes, 2004;Kelloway et al. 2012;Liu et al., 2022;Taşpınar & Eryeni, 2021;Uslu & Şimşek, 2020;Yin et al., 2013), motivation (Bektaş et al., 2013;Huggins et al., 2016;İrge, 2016;Karabekir et al., 2016;Okçu et al., 2020;Pranitasari, 2020) leadership practices (Bass et al., 2003;Bulut, 2012;Cemaloğlu & Kılınç, 2012;Vineburgh, 2010).Therefore, understanding to what extent school principals' charismatic leadership behaviors affect their emotional labor in the context of the trust they create in teachers can provide insights into how to promote more effective leadership and decision-making in schools.Another important contribution is that emotional labor is an under-researched construct in educational leadership.Examining the predictive levels of trust in the leader and charismatic leadership in the emotional labor process according to teachers' opinions can provide new perspectives by defining the connectional structure of this relationship.
Overall, this research by contributing to the charismatic leadership and leader trust literature has the potential to provide a vision of how principals can be supported in their impact on teachers' emotional labor and thereby to promote positive outcomes for schools, teachers, and students.In the literature review carried out in this context, it was seen that the relationship between emotional labor and charismatic leadership has been the subject of limited study.Moreover, there is a gap in the literature regarding the relationship between emotional labor and the variable of trust in the leader.However, the relationship between charismatic leadership, trust in the leader and emotional labor has not been studied.Therefore, the aim of this study is to examine the multifaceted relationships between charismatic leadership, trust in the leader and emotional labor based on teacher data.In this context, answers to the following questions were sought in the research; 1. What are the participants' views on charismatic leadership, emotional labor and trust in the leader?2. Are charismatic leadership and trust in the leader a significant predictor of emotional labor?

Research Design
The purpose of this study is to examine the relationship between teachers' charismatic leadership, emotional labor and trust in the leader.The study was designed with a correlational survey model using quantitative research methods.The survey is the description of a current or past situation (Büyüköztürk et al., 2014).The relational research model was preferred to determine the existence, scope and possible consequences of the differences between charismatic leadership, emotional labor and trust in leaders (Fraenkel et al., 2012).

Participants
The population of the research consists of 8336 teachers working in public primary, secondary and high schools in Trabzon in the 2020-2021 academic year.It was assumed that 522 teachers would represent the sample as a result of calculations made at 95% confidence level and .05significance level (Özdemir, 2018).In determining the sample, the number of teachers working in 18 districts of Trabzon (Akçaabat, Araklı, Arsin, Beşikdüzü, Çarşıbaşı, Çaykara, Dernekpazarı, Düzköy, Hayrat, Köprübaşı, Maçka, Of, Ortahisar, Sürmene, Şalpazarı, Tonya, Vakfıkebir, Yomra) was taken as a criterion and from which public schools to be selected was determined by the stratified sampling method.
Participation was achieved from all districts.Indicators regarding the demographic information of the participating teachers are presented in Table 1.When the study sample is examined from a demographic perspective, it is seen that there is a homogeneous participation in terms of gender and school level.When examined according to professional seniority, it can be said that the participation rate of those who have just started working is lower and the majority are bachelors.

Measurement (Data Collection) Tools of the Study
Confirmatory factor analysis (CFA) was performed for the validity analysis of the data collection tools of the study.Model-fit fit criteria were used to evaluate CFA results (Şimşek, 2007).These model-fit criteria are presented in Table 2.The evaluations were carried out within the framework of these criterion values (Schermelleh et al., 2003).

Trust in Leader Scale (TLS)
In order to measure the participants' trust in the leader, the "Trust in Leader Scale" developed by Tuti & Özdemir (2020) was used.The scale, designed as a five-point Likert type, is one-dimensional and consists of a total of 20 items (sample item: Respects different opinions).A rating between 1 (Strongly Disagree) and 5 (Strongly Agree) was used in the scoring of the scale.With 522 samples in the current study, the Cronbach's Alpha coefficient of the TLS was calculated as 0.98.As a result of the analysis, the scale explained 73% of the variance.The goodness of fit values obtained as a result of the CFA findings regarding the single-factor model; χ2/df=2.12,RMSEA=0.072,GFI=0.90 and AGFI=0.85 were found to show acceptable fit; NFI=0.97,CFI=0.99 and IFI=0.99 values were found to show perfect fit (Schermelleh et al., 2003).Based on these values, it was concluded that CK-CLS was suitable for this study.

Conger-Kanungo Charismatic Leadership Scale (CK-CLS)
In this study, CK-CLS, developed by Conger & Kanungo (1994) and adapted into Turkish by Özdemir & Pektaş (2020), was used to measure teachers' charismatic leadership views.CK-CLS consists of 25 items and six dimensions rated on a five-point Likert type.These dimensions are (i) determining and explaining the vision (six items) (sample item: He is an enthusiastic speaker), (ii) showing environmental sensitivity (seven items) (sample item: The organization is aware of the limits of its employees), (iii) exhibiting extraordinary behaviors(three items) (sample item: Exhibits unusual attitudes in achieving organizational goals), (iv) taking personal risks (four items) (sample item: Takes great personal risks for the sake of the organization), (v) showing sensitivity to member needs (three items) (sample item: Is sensitive to the needs and feelings of employees) and (vi) not maintaining the current situation (two items) (sample item: Is in favor of a risk-free and solid road map to achieve organizational goals).A rating between 1 (I Completely Disagree) and 5 (I Completely Agree) was used in the scoring of the scale.As a result of the analyzes conducted with 522 samples in the current study, the Cronbach Alpha coefficient of the CK-CLS was calculated as 0.95, which shows that the CK-CLS is reliable.The fit values of CK-CLS were obtained as χ2/df=4.41,RMSEA= 0.094, AGFI=0.78,NFI=0.89,CFI= 0.90, IFI=0.84 and GFI= 0.82.
Based on these acceptable range values (Schermelleh et al., 2003), it was concluded that the CK-CLS was suitable for this study.

Emotional Labor Scale (ELS)
In this research, ELS developed by Diefendorff et al.(2005) and adapted into Turkish by Basım & Beğenirbaş (2012) was used.ELS consists of a total of 13 items and three sub dimensions rated on a five-point Likert type.These dimensions are (i) superficial acting (six items) (sample item: I pretend to be interested in students), (ii) deep acting (four items) (sample item: I make an effort to actually feel the emotions I need to show) and (iii) natural emotions (three items) (sample item: The emotions I show to students are sincere).A rating between 1 (Never) and 5 (Always) was used in the scoring of ELS.Considering the internal consistency coefficients calculated with 522 samples in the current study, it was calculated as 0.92 for the superficial acting dimension, 0.85 for the deep acting dimension and 0.83 for natural emotions.These coefficients support the reliability of the scale.The total Cronbach's Alpha coefficient of the scale was found to be 0.86.

Data Analysis and Procedures
During the data analysis preparation phase, ethical commission and legal permission were obtained first.Within the data collection process, data was collected online due to the effects of the pandemic.The online Google form was shared with all schools in the sample, and participation was achieved on a voluntary basis.Among the 534 data returned, 12 forms that did not respond to all responses, were incomplete, and were not filled out properly were identified and sorted out.Thus, missing data and outlier controls were provided.Additionally, different biases may arise when a researcher collects and analyzes data using the same method.Such bias can lead to erroneous conclusions when comparing results obtained using different data sources or research methods (MacKenzie & Podsakoff, 2012).These results are referred to as the common method bias problem.Harman's single factor test was applied to determine this problem.After the analysis, a structure that did not consist of a single factor was revealed, according to the EFA findings.The variance rate of the first factor was determined as 41%.When this rate is below 50%, it can be stated that there is no common method bias problem (Podsakoff et al., 2003).In addition to the applied analyses, assumptions of linearity, homoscedasticity and normality were also examined.When the graphs created for the linearity assumption were examined, it was seen that the scatter diagram showing the standardized residual values and the standardized predicted values showed a linear relationship and the points were gathered around an axis.Normality test was applied to determine the analysis method.
According to the findings, the skewness and kurtosis values of the total and sub-dimension scores of the three scales are within ±3 limits, except for the natural emotions sub-dimension and the total score of the trust in leader scale.Based on these findings, it can be said that the scale data show a normal distribution.Based on the statement that small deviations from normality do not affect the results much in large samples (Field, 2009) by keeping the upper limit of these values high in data consisting of more than 200 samples, it was accepted that the data met the univariate normal distribution assumption (Çokluk et al., 2012).As a result of the homoscedasticity examined when the scale total scores are the dependent variable and demographic characteristics are the independent variable, it was determined that the variance and covariance matrix of the scales was not homogeneous according to Box's M test (p <0.05).Levene's test was applied to determine the homogeneity of group variances.When the homogeneity of the scales was examined separately, it was found that the natural emotions, charismatic leadership (total), determining and explaining the vision, showing environmental sensitivity, taking personal risks, showing sensitivity to member needs and trust in the leader scale provided homoscedasticity, that is, the variance and covariance matrix was homogeneous (p >0.05).It was found that it was not homogeneous in the dimensions of emotional labor (total), superficial acting, deep acting, exhibiting unusual behavior and not maintaining the current situation (p <0.05).Dependent and independent variables used in regression analysis should generally be continuous variables that are expected to be normally distributed, measured on the least equal interval scale.While continuous variables were included in the regression analysis with their original values, discontinuous variables were included as "dummy variables" in this study.The gender categorical variable in the study by being converted into dummy variables, male was coded as 0 and female was coded as 1, and was included in the analysis.The correlation values of the scale total scores were also examined and the scale total scores were found to be statistically significant (p<0.001).Since the correlation coefficients between the scale scores are r<0.90, it can be stated that there is no multicollinearity between the scores.Data analysis was performed with IBM SPSS v23.Correlational findings of non-normally distributed data were interpreted with Spearman's rho correlation coefficient.Although multiple linear regression analysis is mostly used in the literature in regression models with one independent and two dependent variables, this analysis is preferred in cases where it is tried to explain the change caused by independent variables on the dependent variable at the same time.In this study, hierarchical regression analysis was used with the influence of the research context of the independent variables of trust in leader and charismatic leadership, which affect the explanation percentage of the dependent variables composed of the sub-dimensions of emotional labor, because it aimed to show the effects separately by adding them in blocks (Howell, 1992).Finally, the classification made by Cohen (1988) was used to evaluate the correlation coefficient between variables.
According to this classification, when (r); is between 0 -0.09, there is no relationship, when it is between 0.10 -0.29 there is low relationship, when it is between 0.30 -0.49there is medium relationship, and when it is between 0.50 -1.00 there is high relationship.

Findings Findings Regarding Measurement Model
Before examining the structural relationships between the independent and dependent variables of the research, a measurement model was developed for the research variables.The results of the confirmatory factor analysis performed with the AMOS program are presented in Table 3.As seen in Table 3, the goodness of fit values of the measurement model are within acceptable reference ranges (Hu & Bentler, 1999).

Findings Regarding Charismatic Leadership, Emotional Labor and Trust in Leader
The descriptive and relational findings of the participants in the study in the context of variables are presented in Table 4.When Table 4 is examined, it is seen that Spearman's rho correlation values are significant across the variables and their sub-dimensions.
According to the correlational analysis between the total scale scores obtained, there is a positive, high level relationship between trust in the leader and charismatic leadership (r = .64,p <.05).According to the relational analysis between the sub-dimensions, the highest values (r = .68,p <.05) were found between not continuing the current situation and natural emotions; the lowest values (r = .11,p<.05) are observed between trust in the leader and superficial acting.

Findings Regarding the Predictivity Power of Charismatic Leadership and Trust in Leader on Emotional Labor
In the study, analyzes of charismatic leadership and trust in the leader's predictive power of emotional labor were presented as three separate findings under the headings of sub-dimensions of emotional labor.Hierarchical regression analysis showing the predictive levels of trust in the leader and charismatic leadership on the superficial acting dimension of the participants is presented in Table 5. Hierarchical regression analysis showing the predictive levels of trust in the leader and charismatic leadership on the deep acting dimension of the participants is presented in Table 6.Hierarchical regression analysis showing the predictive levels of trust in the leader and charismatic leadership on the natural emotions dimension of the participants is presented in Table 7.

Discussion and Conclusion
This study aims to examine teachers' views on charismatic leadership, trust in the leader and emotional labor.In this regard, the results obtained were discussed by comparing them with the results in the literature.
As the first question of the research, the relationships between the variables were examined descriptively.First of all, the correlational between the sub-dimensions of charismatic leadership and emotional labor was handled.According to the results, the strongest relationship appears to be between the sub-dimension of charismatic leadership, showing sensitivity to member needs and not maintaining the current situation, and the natural emotions sub-dimension of emotional labor.
Based on these results, it can be stated that there is a strong linear relationship between school principals' sensitivity to member needs and a decrease in their behavior of not maintaining the current situation, and teachers' levels of displaying natural emotions.When looking at the literature, no research was found on these two main variables.However, there are researchers working with other leadership areas of emotional labor.The finding of the relationship between instructional leadership and natural emotions in Tombak's (2017) study and the finding of the relationship between ethical leadership and natural emotions in Kafadar and Kaygın's (2017) study are similar to the research results of the study.When looking at the correlational relationship between trust in the leader and the subdimensions of emotional labor, it was concluded that the strongest relationship was between trust in the leader and the natural emotions sub-dimension.Based on this result, it can be expressed that the more teachers trust their school principals, the more their level of displaying their natural emotions will increase.The results of studies detecting the positive relationship between emotional labor and trust in the manager variables support this statement (Beğenbirbaş & Turgut, 2014;Tuti, 2021;Ünal, 2011).
Based on the total scores of the scale, it was concluded that there was a statistically significant, positive and high level correlation between charismatic leadership and trust in the leader.Based on this result, it can be said that as the charismatic leadership behaviors of school principals increase, there may be an increase in teachers' trust in school principals who are leaders.In other words, it can be said that the charismatic leadership behaviors of school principals have a high positive effect on teachers' trust in the leader behaviors.When the outcomes of charismatic leadership towards the leader is looked at, these outcomes are expressed as (i) respect for the leader, (ii) trust in the leader, (iii) satisfaction with the leader (Özdemir & Pektaş, 2020).
These results regarding the leader can be stated to establish the basis that charismatic leadership explains a significant proportion of the total variance of trust in the leader in the study.
When looking at the literature, very few studies were found that examined the variables of charismatic leadership and trust in the leader relationally.The research conducted has shown that there are studies on trust in the manager and different types of leadership.In these studies, in Kavak's (2019) study, relationship between trust in the manager and fatherly leadership; in Doğan (2019) study, the relationship between trust in the manager and transformational leadership and in Paşa's (2015) study, the relationship between trust in the manager and ethical leadership are positive and significant, which is similar to this result of the study.According to the results of the hierarchical regression performed to determine the predictive level of trust in the leader and charismatic leadership on the superficial acting, deep acting and natural emotions dimensions of emotional labor, it was seen that all three models predicted all dimensions of emotional labor.When the hierarchical regression results were evaluated in general, it was seen that the predictive percentage of charismatic leadership and trust in leader variables on the sub-dimensions of emotional labor was the most in the superficial acting dimension and was the least in the deep acting dimension.In addition, according to the research results, charismatic leadership and trust in the leader negatively predict only the superficial acting dimension, unlike other dimensions of emotional labor.So, it can be said that teachers' seeing school principals as charismatic leaders and trusting them reduces their superficial role-playing behaviors and increases their deep roleplaying and natural emotion behaviors.When looking at the literature, not any research examining the prediction of charismatic leadership and leader trust variables on the sub-dimensions of emotional labor is found.However, studies examining the predictive power of different leadership types on the sub-dimensions of emotional labor were seen.
In the research conducted by Beğenirbaş and Yalçın (2020), it was determined that the perception of servant leadership had a positive correlation with employees' deep acting and natural emotion behaviors, while it had a negative correlation with the superficial acting dimension.This finding is similar to the research results, as well.
In the research conducted by Şenel and Aydoğan (2019), the effect of ethical leadership on the sub-dimensions of emotional labor was examined and it was concluded that ethical leadership had a positive significant effect on all dimensions of emotional labor.This finding does not coincide with the results of the research.As for the reason, it can be stated to stem from sample characteristics of the research consisting of different regions and school types.

Limitations and Further Research
This research also has some limitations.The first limitation is that this research was conducted with quantitative research methods based on cross-sectional data.Based on this limitation, conducting a study in which temporal change can be monitored based on longitudinal data might be recommended.In the proposed research, the reasons for the quantitative data can be examined in more depth by using qualitative or mixed research methods.Another limitation is that the research was conducted based on teachers' opinions.For this reason, it may be recommended that future research should be conducted by consulting the opinions of school administrators.

Conclusion and Suggestions
When the results of the research are evaluated in general, some suggestions can be made to policy makers and practitioners.In the research, it is seen that school principals exhibit charismatic leadership behaviors at a moderate level.Based on the characteristic of charismatic leadership to emerge in crisis situations, it can be said that there is an expectation that school administrators should have a more adequate level of crisis management skills during the epidemic.
In this context, it may be recommended that the Ministry of National Education (MoNE) prepare crisis management training modules for school administrators that will strengthen charismatic leadership behaviors and conduct administrator trainings.In addition, based on the result of the study that the charismatic leadership behaviors of school principals have a high positive effect on the trust in the leader behaviors of teachers, it is thought that the charismatic leadership and trust behaviors of school administrators are of significance.Therefore, in the selection of school administrators by the MoNE, especially at the interview stage, it may be recommended to establish special commissions, with reference to leadership, trust and charismatic leadership traits of school administrator candidates, and appoint individuals who can show sensitivity to the needs of school stakeholders, have a vision, have high persuasive abilities and can take personal risks.
Ethics Committee Approval: This research was conducted with the permission obtained by the Hacettepe University Scientific Research and Publication Ethics Board's decision dated 20/05/2020 and numbered 35853172-300.

Conflict of Interest:
The authors declare that they have no conflict of interest.
Author's Contribution: Both authors have contributed fairly through division of labor and solidarity.

Table 2
Criteria for Fit Indices

Table 5
Predictive Level of Trust in Leader and Charismatic Leadership on Superficial Acting Dimension

Table 6
Predictive Level of Trust in Leader and Charismatic Leadership on Deep Acting Dimension

Table 7
Predictive Level of Trust in Leader and Charismatic Leadership on Natural Emotions Dimension p<.05) do not significantly predict natural emotions.It was observed that the predictive effect was greater in favor of female participants in the gender variable and in favor of the participants between the ages of 41-50 in the age variable.When examined in Model 2, all variables except gender (β=-.191,t(522)= -3.735, p<.05) and age (β=.174, t(522)= 2.046, p<.05) do not significantly predict natural emotions.In Model 3, it appears that all variables except for gender (β=-.184,t(522)= -3.614, p<.05) and charismatic leadership (β=.150*, t(522)= 2.243, p<.05 do not significantly predict natural emotions.