“Positive leadership and organizational identification: Mediating roles of positive emotion and compassion”

This study examined the effect of positive leadership experienced by employees on their organizational identification and the mechanisms through how positive leadership enhances organizational identification. Despite the importance of positive leadership at work, such as increasing the followers’ positive self-concepts at work, our understanding of the mechanisms through how positive leadership leads to employees’ organizational identification is still in a nascent state. Thus, this study addresses the research gap by investigating the mediating roles of compassion and positive emotions in the relationship between positive leadership and organizational identification. This study used path analysis and bootstrapping to analyze the survey data collected from 312 full-time employees in 10 large-sized companies located in South Korea. The path analysis results confirmed that positive leadership increases employees’ organizational identification. Furthermore, positive leadership enhances compassion and positive emotions at work, which, in turn, leads to higher organizational identification. Most importantly, the bootstrapping results revealed that both compassion and positive emotion mediate the positive relationship between positive leadership and organizational identification. The empirical results broaden the understanding of the mechanisms through how positive leadership could bring positive effects at work, such as employee organizational identification. Thus, from the managerial perspective, organizations could emphasize the importance of positive leadership at work, and provide the managers with opportunities to develop their positive leadership.


INTRODUCTION
Employees interact with a wide range of organizational insiders. Among the insiders, leaders' behaviors (i.e., leadership) play important roles in shaping employees' (i.e., followers') work behaviors and attitudes positive leadership behaviors can be applied through the positive organizational scholarship lens within work environments, family, and personal lives. However, empirical results of the relationships between positive leadership and other positive organizational constructs and the mechanisms through how positive leadership leads to desirable outcomes at work are still limited. Hence, this study attempts to investigate the positive effects of positive leadership on employees' organizational identification. Besides, to enhance the knowledge about the mechanisms through how positive leadership enhances employees' organizational identification, this study aims at exploring the mediating roles of compassion and positive emotions in the relationship between positive leadership and organizational identification.
This study provides positive leadership literature with theoretical implications. Firstly, by exploring the positive effects of positive leadership on organizational identification, this study extends the knowledge on positive leadership. It emphasizes its importance, especially for leaders dealing with employees under a great deal of work stress, exhaustion, and burnout in the organizations. Besides, by examining the mechanisms through how positive leadership energizes organizational identification, this study provides fruitful ways to further investigate the effects of positive leadership at work. This study also provides practical implications to the organizations by highlighting positive leadership in organizations. Specifically, when employees experience high levels of stress at work, positive leadership behaviors can facilitate the quality relationships with leaders (e.g., development of gratitude, forgiveness, consideration, and care) and development of positive emotion at work, which helps them perceive high levels of organizational identification. Affective event theory (Weiss & Cropanzano, 1996) suggests that the experiences of work events trigger employees' affective reactions, which, in turn, emotion their work attitudes and behaviors. According to this theory, it is likely that employees who experience positive leadership behaviors interpret the interactions with their leaders positively, leading to their positive emotions. It is consistent with Cameron (2012) and Lilius et al. (2008) arguing the positive relationships between positive leadership and employee's positive emotions. It is also congruent with Kelloway et al. (2013), emphasizing the roles of positive leadership behaviors in leading to employees' positive emotions. Thus, based on the affective events theory and previous literature, the following hypothesis has been established.

LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
H1: Positive leadership is positively related to positive emotion. Employees can create a positive organizational culture by being considerate, forgiving, and expressing gratitude toward one another through the act of compassionate care. Positive leadership can likely facilitate employees' compassionate behaviors because it leads to a high-quality connection beyond simply being close to one another or having a one-time relationship among the employees (Kahn, 1993). Positive leadership can serve as the driving force for employees to spread narratives about care and forgiveness, thanksgiving, and compassion within organizations, and create a compassionate climate at the organizational level (Miller, 2007).

Positive leadership and compassion
Social exchange theory (Blau, 1964) also supports the above prediction. It provides a conceptual basis for understanding the relationship between individuals and organizations and the formation and development of interactions between individuals (Eisenberger et al., 1997). Social exchange argues that if an individual receives a positive response from another, such as appreciation and trust, he or she must give back. As such, the social exchange theory em-phasizes the inner psychological exchange between human beings or between organizations and workers. Thus, when employees experience positive leadership from their leaders, they develop feelings of gratitude and trust, and the feeling of appreciation and trust is, in turn, passed on to other employees through compassion.
H2: Positive leadership is positively related to compassion.

Positive emotion and organizational identification
The positive effects of positive emotion at work are well established in the literature. For example, Staw and Barsade (1993) found that employees who experienced more positive emotions are more active in their organizations compared to those who did not. A study by Brief and Roberson (1989) also found that the positive emotion experienced by employees over the previous few weeks had a significant impact on overall performance variables, and Fisher (2000)  It is predicted that employees who experience positive emotions at work are likely to perceive high organizational identification levels. Organizational identification refers to the "perception of oneness with or belongingness to [the organization]" (Ashforth & Mael, 1989, p. 34). In other words, it captures the extent to which employees perceive the congruence of their own and their organizational values. When employees experience positive emotions at work, they are likely to form positive attitudes toward the organizations, leading to emotional attachment to and identification with the organization (Meyer & Allen, 1991). Supporting this, Fredrickson (2001) argues that employees' positive affective responses lead to positive outcomes, such as organizational identification and emotional attachment, and reduce dysfunction, such as negative attitudes toward the organization.
H3: Positive emotion is positively related to organizational identification.

Compassion and organizational identification
When employees experience compassion in an organization, they develop a greater emotional attachment to the organization (Meyer & Allen, 1991). According to the affective event theory (Weiss & Cropanzano, 1996), the experience of compassion is likely to be perceived as positive events within an organization. Specifically, employees who experience compassionate acts by others are likely to perceive they are genuinely cared for by their organizations. As a result, employees are likely to perceive their organizations as good places to work with and have a sense of pride toward their organization (Cialdini et al., 1976). Thus, based on the social identity theory (Tajfel & Turner, 1986), employees are more likely to be attached to their organizations (Rhoades et al., 2001), which, in turn, identify themselves with the organizations. Especially, because compassionate acts within an organization often arise in response to employees struggling with their tasks, the experience of compassion could assist in balancing their lives and tasks and reduce conflict and further strengthen organizational identification (Allen, 2001).
Dutton and his colleagues argued that identifying with an organization is determined by two organization images (Dutton & Dukerich, 1991;Dutton et al., 1994). First is the organization's identity as recognized by members of the organization. This is the belief in how different a particular organization is from other organizations and how long the differentiated image can last. Second, organizational identification is the belief of members of an organization about how people outside the organization perceive the organization. Dutton et al. (1994) stated that these two beliefs determine how attractive an organization is to its members. They also argued that an organization's attractiveness determines how its members identify with it (Dutton et al., 1994). Members who experience positive leadership within the organization will experience compassion through the caring behavior of the leader who responds to suffering and will develop an organizational identification in the sense that an organization with a compassionate organizational culture is different from other organizations. Besides, members of the organization who have experienced compassion through posi-tive leadership develop an organizational identification from the perspective that external members will have a good image of their organization. Therefore, based on the social identity theory, positive leadership induces compassion within an organization, and eventually, it becomes the driving force for improving organizational identification.
H4: Compassion is positively related to organizational identification. Besides, based on the social identity theory (Tajfel & Truner, 1986), employees are likely to identify themselves with the organization. They have the passion and attachment to achieve the organization's common goals with their colleagues, subordinates, and supervisors (Ashforth & Mael, 1989). Furthermore, positive leadership experience allows healthy and productive interactions with others facilitating employees' sense of pride of belonging to their organizations (Bartel, 2001).
H5: Positive leadership is positively related to organizational identification.
Based on the prior hypotheses, it is predicted that positive emotion and compassion mediate the relationship between positive leadership and organizational identification. First, following the affective events theory (Weiss & Cropanzano, 1996), as previously mentioned, the experience of positive leadership triggers employee's positive emotions, which, in turn, leads to the perception of organizational identification. In other words, positive leadership, as a work environment, can serve as a contextual factor inducing employees' positive emotions and attitudinal responses (Ashkanasy & Daus, 2002;Weiss & Cropanzano, 1996). Specifically, employees who experience positive leadership at work are likely to perceive higher organizational identification, recognizing that their organization has a positive organizational culture compared to other organizations (Dutton & Dukerich, 1991). Second, it is predicted that compassion could mediate the relationship between positive leadership and organizational identification. Specifically, positive leadership creates positive relationships and communication among employees, such as compassion (i.e., the experiences of consideration, gratitude, care, and forgiveness by others in the organization). Therefore, based on the social identity theory, such employees are more likely to identify themselves with their organizations (Allen, 2001;Ashforth & Mael, 1989).
H6: Positive emotion mediates the relationship between positive leadership and organizational identification.
H7: Compassion mediates the relationship between positive leadership and organizational identification.

AIMS
This current study aims at underlining the mechanisms thorough how positive leadership enhances employees' organizational identification in large-sized Korean companies. Specifically, this study seeks to examine the mediating roles of compassion and positive emotions at work as a missing linkage in the relationship between positive leadership and organizational identification. Given that not much empirical studies have been conducted on compassion at work (Ko & Choi, 2019), this study seeks to shed light on the positive roles of compassion in the workplace.

Participants and procedures
In this study, surveys were conducted with employees working at ten large corporations in South Korea. Among the fourteen companies the research team contacted, ten companies decided to voluntarily participate in this study. Upon their approval, the research team visited the ten companies and gave a short presentation encouraging employees' voluntary participation. A total of 330 employees responded to the online surveys; excluding 18 surveys with insincere responses or central tendency, 312 surveys were used for data analysis. The majority of the participants were in their thirties (N = 130; 41.7%). Among the participants, 130 (41.7%) were male. Many of the participants had worked for their current organizations for 1 to 4 years (N = 144; 46.2%), followed by those who had worked for their organizations for between 5 and 9 years (N = 68; 21.8%).

Positive emotion
Positive emotion was defined as "the joyful, proud, and inspiring emotions experienced by organizational members" (Lilius et al., 2008). Specifically, four items by Lilius et al. (2008) were used. Sample items include "I am proud of everything" and "I am happy about everything." The items were measured on a 5-point Likert scale, and the scale's reliability was acceptable (Cronbach's alpha = 0.725).

Organizational identification
Organizational identification was defined as "to identify with the organization through the formation of organizational identity and to recognize the organization and its members as a common fate by sharing the goals of the organization to which they belong" (Mael & Ashforth, 1992

Compassion
Compassion was defined as "an act of responding to the suffering of others" (Lilius et al., 2011). Three items developed by Lilius and his colleagues (2008) were used. Sample items are "I often experience compassion at my workplace," "I often experience compassion from my supervisor," and "I often experience compassion from my colleagues." The items were measured on a 5-point Likert scale, and the scale's reliability was acceptable at a Cronbach's alpha value of 0.894.

Control variables
Demographic variables (gender, age, position, the term of service, term of continuous service), which might affect the constructs of positive leadership, positive emotion, compassion, and organizational identification, were controlled. Furthermore, other compounding effects that individuals' propen-sity might have on the current study constructs were also controlled.

Common method bias
A cross-sectional study that simultaneously measures independent and dependent variables with self-reported responses may be biased (Podsakoff et al., 2003). Thus, Harman's one-factor test was conducted (Podsakoff & Organ, 1986) to determine if the measurement of this study suffered common method bias issues. The results confirmed the presence of four distinct factors with an eigenvalue greater than 1.0, rather than a single factor, and the covariance explained by one factor was 17%, indicating that the common method bias is not a serious concern (Podsakoff & Organ, 1986; Podsakoff et al., 2003).

Results of confirmatory factor analysis
The discriminant validity was verified by Average Variance Extracted (AVE), and the internal reliability was examined with Cronbach's alpha coefficients. Cronbach's alpha coefficients of all the study variables were greater than 0.7, and AVE values were greater than 0.6 for all latent variables, meeting the traditionally recommended criteria. The χ 2 estimate was significant (χ 2 (314) = 583.257; p < .001). The confirmatory factor analysis results are presented in Table 1.

Correlation analysis and validity test
Multicollinearity was checked before testing the hypotheses. Pearson's coefficients were used to examine the correlations between latent variables. Regression analysis was conducted to check for multicollinearity, and the results showed that the variance inflation factors (VIFs) were between 1.127 and 1.262, indicating no serious multicollin-earity issues. Table 2 shows the correlation analysis results.

Hypotheses testing
After the confirmatory factor analysis, items with a low factor loading were removed from the model, resulting in three items for the construct of compassion, three items for compassion, four items for positive emotion, and five items for organizational identification. The hypothesis test was based on this final set of measurement items.
The fit of the research model was verified through a confirmatory factor analysis. The confirmatory factor analysis results for the measurement model are presented in Table 1, and the results of hypothesis testing are presented in Table 3. First, positive leadership experienced by employees turned out to be positively and significantly associated with positive emotion (β = .605, p < .001), confirming H1. Positive leadership was positively and significantly associated with compassion (β = .415, p < .001), so H2 was supported. Third, positive emotion was positively and significantly associated with organizational identification (β = .830, p < .001), supporting H3. Fourth, compassion was positively and significantly associated with organizational identification (β = .477, p < .001), confirm-ing H4. Finally, organizational members' positive leadership turned out to be positively and significantly associated with organizational identification (β = .605, p < .001), confirming H5.
Although regression has been commonly used to test the mediation effect, it failed to adequately reflect measurement errors and have been known to have limitations in examining complex and sophisticated mediation models such as a multi-media-

Implications
The results of this current study provide several important implications. First, in contrast to the study by Lilius et al. (2008), which demonstrates the causal relationship between compassion, positive emotion, and affective commitment, this study demonstrates the mediating relationship between positive leadership and organizational identification through the formation of positive emotion and the experiences of compassion at work. Thus, the results add knowledge on the mechanisms through how positive leadership could lead to positive outcomes, such as organizational identification.
Second, this study emphasizes the benefits of virtuous behaviors such as gratitude, forgiveness, and positive emotion through positive leadership for employees struggling with excessive work burden, job stress, emotional exhaustion, and turnover intention. Because it is inevitable for employees to experience suffering at work, leaders' positive leadership behaviors could be integral to help employees form organizational identification, which, in turn, would lead to higher performance.
Third, this study emphasizes the development of relationships among the members of a particular company to high-quality connections when such members experience positive leadership, compassion as care, and compassionate acts (Dutton & Heaphy, 2003). The results emphasize the roles of positive leadership in facilitating the formation of positive relationships among employees.
Fourth, this study demonstrates that positive leadership significantly influences organizational identification, unlike an empirical study on positive leadership regarding positive psychology or the Positive Organizational Scholarship (POS). Therefore, this current study provides a practical implication that when employees experience positive leadership, they develop compassion as an act of positive emotion and care. Ultimately, this reduces the turnover of excellent employees and facilitates the members' identification with the organization.

Limitations and future research directions
First, this study collected samples from employees in the metropolitan area, which limits generalization. In future studies, samples from various companies in the nation should be collected, including samples from Korea's non-metropolitan areas.
Second, this study used three items from the compassion scale based on the research by Lilius et al.
(2008) who have conducted qualitative and quantitative research centered on the positive organizational scholarship since the late 1990s. However, in future studies, it is necessary to derive variables related to compassion, that is, core categories, through a qualitative research based on grounded theory in terms of the validity of the scale (Strauss & Corbin, 1997). Therefore, in future studies, it is necessary to derive a research model for empirical study based on the causal relationships of these core categories to further prove the validity of the compassion scale.
Third, this study demonstrated the causal relationship between organizational identification and positive leadership as the dependent variable. However, in the future, studies should be conducted to demonstrate the causal relationship between positive leadership and job enthusiasm, positive leadership and incivility, and positive leadership and other performance variables.
Fourth, this study shows the limitation of conducting a cross-sectional study, and the problem of the directionality of the causal relationship can be raised as the samples were collected at some time. For example, the hypothesis implies a causal assumption that members of an organization who have experienced positive leadership will positively affect compassion. However, in the actual results, it is impossible to exclude the possibility that compassion, the outcome variable of this current study, affects positive leadership. Therefore, in future research, it is necessary to collect samples through longitudinal studies to make up for the limitations of such a cross-sectional analysis.

CONCLUSION
This current study provided empirical evidence supporting the mediating effects of compassion and positive emotion in the relationships between positive leadership and organizational identification, given the increasing importance of positive leadership in organizations, especially in the realms of positive organizational scholarship. This current study expands the understanding of the mechanisms through how positive leadership enhances followers' organizational identification. Future research employing a longitudinal research design and samples in different cultures could further validate and strengthen the results. It is integral for organizations to recognize the importance of positive leadership in practice. Especially, given that organizational identification is critical for organizational performance and talent management, organizations should provide leaders with adequate leadership training so that they can develop their positive leadership skills.