The Mediator Role of Organizational Support in the Relationship between Organizational Identity and Organizational Stress

The purpose of this research is to test the theoretical model developed for the mediator role of perceived organizational support in the relationship between organizational identity and organizational stress. The research is conducted with 320 teachers who work in preschools, primary schools, elementary schools, and high schools. The data were collected using organizational stress scale, perceived organizational support scale, organizational identity scale, and personal information form. The direct and indirect relationships between perceived organizational support, organizational stress, and level of organizational identity were analysed using the Structural Equation Model. The proposed structural model was verified by the analyses. It is observed that the relationships between all variables in the research have significant values and their goodness of fit indices were within the acceptable level. The results of the analyses showed that, organizational identity significantly predicts organizational stress and perceived organizational support. It was also found that the impact of organizational identity on organizational stress was fully mediated by perceived organizational support. According to these results, organizational support plays a mediating role in the relationship between organizational identity and organizational stress.

Strong organizational identity can prevent or reduce the occurrence of organizational stress. Identity at the individual level is the answer given to the question of "who am I" which reveals the basic characteristics of individuals (Gioia, 1998). The notion of organizational identity is depicted with distinctive attributes and is based on the organization's character (Gioia et al., 2000). Organizational identity is defined as a set of individual evaluations that define the general behavior of the organization and reveals the different perceptions of individuals about the organization which they are a member of (Duttonet al., 1994). Organizational identity affect teamwork (Mesmer-Magnus, et al., 2018) and success or failure of the teams has important implications for organizational success (Bankins & Waterhouse, 2019;Hurst &Pattath, 2019). Employees who adopt the values, goals and objectives of their organizations, see themselves as part of the organization, and exert extra effort towards organizational success (Tore, 2019). Every organization needs to have a strong identity unique to itself (Sanli & Arabaci, 2016). In addition, internal factors such as organizational culture (Hatch & Schultz, 1997) and external factors such as environment (Echeverri, 2018) of an organization are effective in the construction of the identities of the organizations (Dhalla, 2007).
The identity created in an organization affects the organization and its employees. Employees committed to the organization and working for the good of the organization with an outward mindset is called organizational support (Eisenberger et al., 1986). According to another definition, organizational support means employees' not only being appreciated through financial rewards but also being valued for their happiness, success, and work ambition (Tanriverdi & Kilic, 2016). Perceived organizational support contributes to employees' physical and emotional attachment to the organization, which in turn affects the performance of the employees and the organization, and contribute to the organizational success (Eisenberger et al., 1990;Chang & Bangsri, 2020). For instance, if the perception of organizational support is high, the workers might not be prone to organizational cynicism (Kanbur & Canbek, 2018). Additionally, their likelihood to leave the organization might stay low (Wayne et al., 1997). Also, organizational support factors might contribute to the work-life balance of the employees (Clark et al., 2017).
When the literature is examined, it has been determined that the concepts of organizational identity, organizational image and organizational culture are related (Hatch & Schultz, 2002). In this context, it is emphasized that a consistent and strong organizational identity is of great importance for organizational success (Brown & Starkey, 2000). In addition, while organizational support positively affects organizational identity, cooperation and emotional commitment of employees (Lam et al., 2016), it affects organizational stress negatively (Korkmaz & Koseoglu, 2018). It can be presupposed that the workers who perceive the value given to them as high and believe that their organization has a strong organizational identity will be successful in overcoming organizational stress. However, in literature, there are no studies about whether organizational support has mediator in the effect that organizational identity has on organizational stress. Therefore, in this study the mediating role of organizational support in the relationship between organizational identity and organizational stress was investigated, a theoretical model was proposed and the model was tested with the data collected.

Research Design
The purpose of this study is whether organizational support has a mediator role in the effect that organizational identity has on organizational stress. In line with this main purpose, answers were searched for the following questions: (i) Is there a correlation between organizational support, organizational identity, and organizational stress?
(ii) Does organizational support hold a mediator role in the relationship between organizational stress and organizational identity?
In order to determine the mediator role of perceived organizational support in the relationship between organizational stress and organizational identity, one of the quantitative research methods, correlational research design, was used. In this model, a statistical approach was used to determine the relationship between two or more variables (Robson & McCartan, 2011). The Structural Equation Model was used to identify the mediator role of organizational support in the relationship between organizational stress and organizational identity. The mediator model used in this research is shown in Figure 1. . The sample of the research was selected using one of the probability sampling methods: Stratified random sampling. 320 teachers volunteered to participate in the study.112 (35%) of the teachers were male while 208 (65%) were female.

Data Collection Tools
Three distinct scales were used to develop the scale in this study to determine the mediator role of perceived organizational support in the relationship between organizational stress and organizational identity. These three scales were developed by the researchers abroad and were translated into Turkish. The first part of the scale in this study asked questions about the genders, professional seniorities, marital statuses, levels of education, and incumbencies of the participants. In the second part of the scale, Balci's (1993) validity and reliability approved 24 items organizational stress scale was used. The Cronbach's Alpha coefficient for the scale was .876. In this study, the Cronbach's Alpha coefficients for the scale have been found to be .944. In the third part, an organizational support perception scale developed by Eisenberger (1986) was used. The original scale of Eisenberger contained 36 items and after the validity and reliability studies it was shortened to 12 items by Giray and Sahin (2012). The reliability of the scale was .93. In this study, the Cronbach's Alpha coefficient for the scale has been found to be .906. In the last part of the data collection tool, an organizational identity scale which was developed by Gioia and Thomas (1996) was used. The scale was translated in to Turkish by Sahin (2014). The Cronbach's Alpha value for the scale was .69. In this study, the Cronbach's Alpha coefficients for the scale have been found to be .850.

Data Analysis
In this study which aims to determine the mediator role of perceived organizational support in the relation of organizational stress and organizational identity, correlation analysis and the Structural Equation Model were used. The scales used in the research were validated and approved beforehand. The collected data were analyzed using SPSS 21 and AMOS 25 packages. The mean scores of the variables, organizational stress, perceived organizational support, and organizational identity, were calculated. The Structural Equation Model was used to test out the proposed model. In the Structural Equation Model, measure and goodness of fit index are crucial for analysis. At the end of the analysis, these values contribute to reach a verdict considering every constructed model in order to make evaluations (Karagoz, 2017;Meyers et al., 2006). Data entry has been checked to prevent possible errors. In order to ensure that the data are compatible with statistical methods, outlier analysis and missing value analysis were performed. To find the outliers, z scores have been calculated for each variable. Afterwards, the responses of the participants exceeding ± 3 have been excluded from sample. The skewness and kurtosis coefficients were examined. The skewness and kurtosis values for the organizational stress scale were 1.183 and 1.031 respectively. The skewness and kurtosis values for the perceived organizational support scale were -.269, and-.842 respectively. For the perceived organizational identity scale, the skewness value was measured as -.407, and the kurtosis value as -.283. Skewness and kurtosis values between -1.5 and 1.5 are normality indicators (Tabachnick & Fidell, 2013).
In this study, Harman's one-factor analysis was used to examine the common method bias that may occur due to the fact that multiple scales are included in the same questionnaire and all scales are evaluated at the same time by the participants (Podsakoff et al., 2003). If the variance value of the first occurring factor of the scale is less than 50%, there is no common method bias in the scale (Fuller et al., 2016). Since a single factor in the scale does not explain most of the variability with 31.751% variance value, there is no common method bias in the scale.

Organizational support
Organizational stress Organizational identity The mediator roles can be analyzed using various methods in the structural equation model. In order to test the organizational model to determine the effects between organizational identity, organizational stress, and organizational support, the Structural Equation Model was used. Path analyses were conducted to determine the mediator role that perceived organizational support has in the relationship between organizational stress and organizational identity. If the three variables have causal consecutiveness, a mediator variable can be discussed. There are certain rules for a variable to be a mediator.
1. The independent variable (organizational identity) should have an effect on the dependent variable (organizational stress).
2. The independent variable (organizational identity) should have an effect on the mediator variable (organizational support).
3. The mediator variable (organizational support) should have an effect on the dependent variable (organizational stress).
4. When the mediator variable (organizational support) and independent variable (organizational identity) are added to the model together, the effect of the independent variable (organizational identity) on the dependent variable (organizational stress) should decrease; the effect of the mediator variable (organizational support) on the dependent variable (organizational stress) should be significant (Baron & Kenny, 1986;Karagoz, 2017).

Organizational identity Organizational stress
Organizational identity Organizational support

Organizational support Organizational stress
Organizational support

Organizational stress a Organizational identity
While the model is being tested, it is analyzed according to the model shown in Figure 2. The effect organizational identity has on organizational stress (path a), the effect organizational identity has on organizational support (path b), and the effect organizational support has on organizational stress (path c) were calculated. Finally, the effect that organizational identity and organizational support together have on organizational stress was tested. In the effect that organizational identity has on organizational stress, when the organizational support is added to the model as a mediator, the path coefficient between organizational identity and organizational stress (path a) was in significant, so it is deleted. At the end of the analysis, in the model in Figure 1, whether organizational support has a mediator role was evaluated according to fit indices. Findings In this research which investigated the mediator role of perceived organizational support in the relationship between organizational stress and organizational identity, in order to determine the relationship between the variables, Pearson Product-Moment Correlation Coefficient was used. The data collected at the end of the correlation is shown in Table 2.  Table 2, a negative and moderate significant relationship was found between organizational identity and organizational stress (r=-.367; p<.000).Similarly, between organizational identity and organizational support, it has been found that there is a positive medium correlation, which is statistically significant (r=.591; p<0.01). Additionally, between organizational stress and organizational support, it has been found that there is a negative medium correlation, which is statistically significant (r=-.476; p<0.01). The analysis results for the research's sub-problems of the study are shown in Figure 3.

Figure 3. Regression Values between Variables
The effect of the independent variable (organizational identity) on the dependent variable (organizational stress) was examined, and it was found to be significantly and negatively effective (-. 414; p <0.01). The effect of the independent variable (organizational identity) on the mediator variable (organizational support) was examined, and it was found to be significantly and positively effective (.652; p <0.01). The effect of the mediator variable (organizational support) on the dependent variable (organizational stress) was examined and it was found to be significantly and negatively effective (-.524; p <0.01).
After determining the effect levels between dependent, independent and mediator variables, all variables were added to the model, and the theoretically proposed model was tested with the collected data. The model is given in Figure 5.

Organizational identity Organizational stress
Organizational identity Organizational support Organizational support Organizational stress According to the model given in Figure 4, organizational identity affects organizational stress negatively (-.128) and organizational support positively (r = 652). In addition, organizational support negatively affects organizational stress (-.441).In order to evaluate the model, it was first checked whether the regression coefficients were significant.
Regression values of the model are given in Table 3.  Table 3, the path coefficients of the model were analyzed and the path coefficient between organizational identity and organizational stress was found to be bigger than p<.05 (p<.082), was eliminated from the model and analysis has been done again. Regression values of the model are given in Table 4.  Table 4, it is seen that the regression coefficients between variables are significant. After deleting the path coefficient showing the direct effect of organizational identity on organizational stress, the resulting model is given in Figure 5. According to Figure 5, when the organizational support variable is added to the model, the direct effect of organizational identity on organizational stress disappears. In this sense, in the theoretical model in which the effect that organizational identity has on organizational stress, organizational support has full mediation. According to Baron and Kenny (1986), when a mediator model is being tested, the effect that the independent variable (organizational identity) has on the dependent variable (organizational stress) should decrease or get close to zero when the mediator variable (organizational support) is added to the model. Correspondingly, it was found that with the mediator effect of the perceived organizational support, the direct negative effect between organizational stress and organizational identity regressed to -.13. This shows us that our model is compatible and that perceived organizational support has

Organizational support
Organizational stress Organizational identity Organizational support

Organizational stress Organizational identity
International Journal of Educational Methodology 649 full mediation. Again, as a result of the analysis, in the ratio of the Ki-square value calculated for the measuring model to the degree of freedom (x 2 /sd=2.42), a value under 5 shows good compatibility. The model is found to have good compatibility values when goodness of fit indices of the measuring model are analyzed (NFI=.912, CFI=.946, GFI=.886, RMSEA=.067, RMR=.028).Direct, indirect and total effects between variables in the study are presented in Table 5. When the total effect between the variables given in Table 5 is analyzed, organizational identity positively affects organizational support (.655, p <.01), while organizational identity affects organizational stress negatively (-.347, p<.01). In addition, organizational support negatively affects organizational stress (-.529, p<.01). When the direct effect is examined, organizational identity positively affects organizational support directly (.655, p<.01), while organizational support directly affects organizational stress negatively (-.529, p<.01). Finally, it has been determined that organizational identity has a negative and indirect effect on organizational stress through organizational support (-.347; p<.01). In other words, organizational support plays a mediating role in the relationship between organizational identity and organizational stress.

Discussion
In this study, the mediating role of perceived organizational support in the relationship between organizational identity and organizational stress was examined. Based on the findings, while a negative and significant relationship was found between organizational identity and organizational stress a positive relationship was found between organizational identity and organizational support. In addition, a negative relationship was determined between organizational support and organizational stress.
As a result of the analysis conducted for the main purpose of the study, it was determined that organizational identity has a positive effect on organizational support and a negative effect on organizational stress. In addition, organizational support negatively affects organizational stress. Finally, when these three variables are analyzed as models, it is seen that the relationship between organizational identity and organizational stress is the full tool of my organizational support. In other words, organizational identity negatively affects organizational stress through organizational support. When the related literature is examined, there is no study on the relationship between organizational identity and organizational stress, and organizational support and organizational stress. However, Dhalla (2007) determined that the perceptions of the external and internal audiences through leadership skills, management practices and strategic decision making will help form a strong organizational identity; Lam et al. (2016) also determined that organizational support positively affects organizational identification, collectivism, extra-role behavior and affective commitment. Korkmaz and Koseoglu (2018) found that positive perceptions of support among employees have an effect on the stress arising from policies and wage systems, and negative support perceptions have an effect on the stress arising from the "organizational structure". From these findings, it can be said that organizational support is a mediating variable between organizational identity and organizational stress. In this sense, it can be assumed that the workers who perceive the support given by their organization as high and believes that their organization has a strong organizational identity will be successful overcoming organizational stress. Additionally, it is observed that there is a significant and negative relationship between organizational stress and organizational support. In this case, perceived organizational support offers a strong way to decrease the negative effects of certain stress directly and indirectly. For this reason, strengthening the organizational identity and increasing the organizational support will help reduce organizational stress. Organizational stress can be reduced by increasing organizational support. For example, organizational support can be increased by contributing to the professional development of employees. Therefore, in accordance with professional competence and requirements, planning and implementation of in-service training activities will be beneficial. Including employees in planning and applications to the extent of their competence will facilitate acceptance and understanding of training by employees (Tore, 2017). Organizations that aim to acquire employees' new knowledge, skills and competencies, update their existing knowledge, skills and competences will be perceived as more supportive (Tore, 2018b). To strengthen the organizational identity, an effective organizational leader and a strong organizational culture can contribute.

Conclusion
When the findings obtained from the research are evaluated together, employees should be supported by the organization in order to increase the organizational identity awareness that contributes to the development of the organization. A strong organizational identity perceived by the employees will contribute to the reduction of organizational stress by increasing the employees' perception of organizational support. The organization will be more successful by ensuring that employees who perceive the organizational identity strongly experience less stress. To combat stress in organizations, building a strong organizational identity and increasing organizational support will help.
The findings of the study show that organizational identity and organizational support negatively affect employees' perception of organizational stress. For this reason, organizations should structure their activities in a way that supports their employees and makes them feel that they are always with them. Thus, the employee's trust in the organization will increase and contribute to the efficiency and effectiveness of the organization. In addition to these results, the management styles in an organization have serious effects on the organization. The importance of managerial behavior or the characteristics of managers that can increase the performance of the members are increasing (Mert & Ozgenel, 2020). Administrators should increase their competencies in topics, such as human affairs, communication, enhancing employees' morale and motivation and making objective performance assessment (Koc, 2018). Administrators can increase employee compliance to the organization. The participation of the employees in the organization to the decisions and taking initiative will provide a healthier working environment and a common identity. In this way, the values adopted by the organization will spread to all members of the organization and it will be possible to create a common working environment.

Suggestions
It is thought that executives and leaders considering these points in the process of school administration will benefit the organization's identity. Alongside this, the research will reach more explanatory results with more research done on different sectors and different regions. For the following research, it might be advised to the researchers to analyze the mediator role of organizational support with variables such as organizational justice and organizational trust using the Structural Equation Model.

Limitations
The study was conducted with335 teachers. Data were collected from teachers working in kindergarten, primary, secondary and high schools. The participants were selected from the public schools where the researcher works. Therefore, the conclusions of this study cannot be generalized.