Employees, the major human resource constituting an organization, contribute to organization development and success by fulfilling their duties and work. To achieve their duties, employees continuously collaborate with other organizational members such as co-workers and their leaders and participate in organizational activities. Through collaboration and communication with the organizational members, employees share organizational values, and develop and shape their organizational attitudes that are directly or indirectly related to organizational performance.

However, in the current working environment where there is high competition among employees, and in turn, a negative mindset and increased stress, employees have trouble in building a quality relationship with their co-workers based on trust and in engaging in their work (Chen & Spector, 1992; Harris, Harvey, & Booth, 2010). To overcome these problems, organizational efforts and supports such as supportive leadership and positive organizational environment are needed to facilitate employees’ cooperative activity and to encourage employees’ motivations. Especially needed are supportive organizational environments that put emphasis on humans and leaders who encourage and deliver hope to employees.

In response to the concerns, several studies have called for supportive and authentic leadership to help employees develop high self-confidence and interpersonal trust, which promote employees to be more engaged in their work and performance improvement (Avolio, Gardner, Walumbwa, Luthans, & May, 2004).

In work environments that are supportive in developing employees’ capabilities and in creating interpersonal trust among organizational members, employees are able to engage more actively in their work and to devote their efforts to achieving better performance. In this regard, the presence of sincere and supportive leadership is the required element for improving employees’ performance and organizational success by positively changing employees’ attitudes in the current business environment.

Since the 1970s, a considerable number of studies have been conducted on leadership, including transformational leadership, ethical leadership, transactional leadership, charismatic leadership, and authentic leadership (e.g., Bass & Riggio, 2010; Dionne, Yammarino, Atwater, & Spangler, 2004; Ghafoor, Qureshi, Azzemi, & Hijazi, 2011; Hmieleski, Cole, & Baron, 2011). These studies have focused on the critical role of leaders within an organization for organizational success. Authentic leadership, however, has only recently emerged in the literature (Gardner, Cogliser, Davis, & Dickens, 2011). Moreover, few studies have focused on the relationship between authentic leadership and employees (Khan, 2010).

Even though some studies have been conducted to study the critical role of authentic leadership and the difference between authentic leadership and other leadership styles (e.g., Champy, 2009; Clapp-Smith, Vogelgesang, & Avey, 2009), most of the research on authentic leadership has been conceptual studies, not statistical studies (Gardner et al., 2011). In their content analysis study, Gardner et al. (2011) found that 91 publications focused on authentic leadership and 59 of those were classified as conceptual studies.

Interest in leadership in Korea has gradually increased since the late 1990s, and studies have recently begun to focus on authentic leadership (e.g. Kang, 2013; Koo, 2013); however, none of these studies has focused on the linkage of authentic leadership and followers’ performance through their attitude and behavior change within a Korean business context.

Thus, the main purpose of this study was to investigate the critical role of authentic leadership and its effects on employees’ attitudes and behavior. More specifically, this study described the influence of authentic leadership on employees’ performance through employee attitudes (self-efficacy and interpersonal trust) and organizational behavior (work engagement). To this end, this study reviewed the literature and developed research hypotheses based on the literature review. The hypotheses were tested using statistical methods.

As many researchers (e.g., Lester, Vogelgesang, Hannah, & Kimmey, 2010; Walker & Henning, 2004) have emphasized the important role of leaders within the organization, leaders’ behavior and value are important because leaders supervise the organization and the followers, and serve as a role model for their followers. Indeed, a leader influences employees’ attitudes and behaviors such as developing confidence in themselves and having increased work engagement through working together (Dirks & Ferrin, 2002; Gardner & Schermerhorn Jr., 2004).

Authentic leaders are “deeply aware of how they think and behave and are perceived by others as being aware of their own and others’ values/moral perspectives, knowledge, and strengths aware of the context in which they operate; and who are confident, hopeful, optimistic, resilient, and of high moral character” (Avolio, Luthans, & Walumbwa, 2004, p. 4). In other words, authentic leaders do not focus on developing their image as leaders and do not engage in their role for honor and personal rewards (Shamir & Eilam, 2005), but rather they act based on their conviction and value-based cause that are internalized by their own personal experiences and reflections on those experiences.

Authentic leaders continuously endeavor to fully understand themselves and to be ready for the future. As a result of their efforts, authentic leaders develop self-awareness of not only their values and beliefs, but also their strengths and weaknesses, and this knowledge becomes the base of their standards for personal conduct. Authentic leaders are also hopeful, optimistic, and confident, and they continuously promote a positive state of confidence in themselves and their followers. In turn, they become ethical role models for their followers (Luthans & Avolio, 2003). By modeling and providing professional development, authentic leaders try to help followers do the same, i.e., developing a better understanding of themselves and being positive and optimistic. Observing and emulating their authentic leaders, followers will also be authentic followers who have high self-confidence and trust in others and positive organizational behaviors resulting in performance improvement (Dirks & Ferrin, 2002; Hassan & Ahmed, 2011).

Several studies have shown how leaders’ authenticity is contagious to their followers (e.g., Avolio et al., 2004; George, 2003). For example, Gardner et al. (2005) suggested that followers who work with authentic leaders also develop authentic followership resulting in workplace well-being and increased work engagement. Moreover, Ilies, Morgeson, and Nahrgang (2005) found that authentic leadership positively affects followers’ behaviors through providing support for self-determination.

In the current study, authentic leaders’ behaviors, values, and beliefs are viewed as factors that result in changes in followers’ behaviors and attitudes such as increased self-confidence and trust. Authentic leaders motivate followers to be more engaged in and aware of their duties by building optimism and hope, by fostering a positive environment, by helping followers find the meaning of work, and by showing consistency in their behaviors according to their values and beliefs that build trust and commitment among followers so that followers can best contribute their efforts for fulfilling their duties(Avolio & Gardner, 2005).

Leadership is an interactive process between leaders and followers (Graen & Scandura, 1987), and this interactive process influences followers’ perceptions, attitudes, and behaviors in organizations. According to Bandura (1997), trustworthiness and credibility of the person being modeled are to be highly valued by followers, and then the followers have intentions to learn and to emulate. Authentic leadership can be positive modeling, which allows authentic leaders and followers to build an authentic relationship, resulting in positive work attitudes such as commitment to work, job satisfaction, and employee engagement (Luthans & Avolio 2003).

In daily life, people continuously face situations in which they must make a decision such as what method they should use to solve problems, and the decisions are usually made based on their judgment ability and information they have. The judgment is called self-efficacy, and affects one’s behavior and attitude toward the given situations or work (Bandura, 1982). Self-efficacy is the “beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments” (Bandura, 1997, p. 3). Depending on the level of self-efficacy, they make decisions about what activity they will participate in, how much effort they will exert, and whether they will embrace adventure and take risks (Bandura, 1977). Self-efficacy is affected by a person’s direct or indirect experiences with success/failure and emotional status, and can also be boosted by others’ efforts. In the workplace, leaders’ confidence in employees’ successful performance improvement helps employees to have high self-efficacy (Eden, 2003). Gardner and Schermerhorn (2004) claimed that authentic leaders who truly know themselves are confident of themselves, and exemplify high moral standards to help their followers find their abilities (efficacy).

The presence of self-efficacy determines whether employees engage in the work, whereas the absence of self-efficacy limits employees’activities and engagement in the work. In their empirical study to investigate the role of occupational self-efficacy, organizational support, and supervisor support in increasing employees’ engagement, Pati and Kumar (2010) found that occupational self-efficacy positively predicts employee engagement. This shows that the differences in the level of self-efficacy of employees indicate the differences in work engagement (Prakash & Kumar, 2010). Moreover, Stajkovic and Luthans (1998) showed that the level of occupational self-efficacy is positively related to organizational performance. They found that self-efficacy has the strongest effect on work performance among organizational behavior modification, goal setting, and feedback intervention.

Meanwhile, in the workplace, working together involves interdependence, and interpersonal trust enables employees to work together more effectively (Mayer, Davis, & Schoorman, 1995). That is, employees’ trust in co-workers and leadership promotes employees’ active participation in communication, information sharing, and their work.

People’s trust is determined by their personal traits, propensity to trust others based on their past experiences, and expectations of others, and trust is also affected by the other party’s attributes (Mayer et al., 1995). Trustees’ characteristics and behaviors such as expertise and trustworthiness are critically related to the level of trust (Mayer et al., 1995). Especially the trustworthiness of the trustees plays an important role in increasing the trustors’ trust in the trustees. Authentic leadership can build trust by showing respect for their followers and providing for their concerns (Avolio et al., 2004). Moreover, authentic leaders’ authenticity and integrity based on their internalized values also increase the level of their followers’ interpersonal trust.

Interpersonal trust is also positively related to employees’ work engagement and performance improvement. When trust in co-workers and management exist within an organization, then that trust results in felt support from management, increased collaboration, and effective communication with the co-workers (Ferres, Connell, & Travaglione, 2004; Tan & Tan, 2000). When employees have a low level of trust in leadership, they are more likely to be psychologically distressed and this distress causes negative results such as low performance and low work engagement with high intention to leave (Dirks & Ferrin, 2002). Previous research has provided much evidence that trust is positively related to employees’ behaviors in terms of work engagement (e.g., Dirks & Ferrin, 2002; Korsgaard, Brodt, & Whitener, 2002). And other studies have shown that interpersonal trust is an enabling factor that facilitates and promotes performance improvement (e.g., Mayer et al., 1995; McAllister, 1995). As a result, employees' positive attitude (high selfefficacy and interpersonal trust) is tied to changes in work engagement, which result in higher performance achievement (Salanova, Lorente, Chambel, & Martinez, 2011).

Kahn (1990) defined engagement as “the harnessing of organization members' selves to their work roles; in engagement, people employ and express themselves physically, cognitively, and emotionally during role performances.” Thus, engaged employees are more likely to be physically, cognitively, and emotionally present when they are participating in their work activities.

The level of employees’ work engagement determines the quality of work performance, and the level of employees’work engagement is determined by many factors in the workplace. Bakker (2011) explained that the level of work engagement is mainly driven by external and internal resources such as job resources and personal resources. Job resources include social support from co-workers and leaders, autonomy, interpersonal trust, and feedback, and these resources perform motivational roles that are directly related to increased work engagement. Authentic leaders act according to their inner value and try to be truthful in their relationship with their employees, which can increase employees’ identification with their work and help employees to feel more psychologically empowered to do their work (George, 2003). This authentic leaders’ behavior and employees’ feeling make employees have ownership for their work and help them to be more engaged in their duties. Leaders’ influence on the degree of individuals’ engagement in their work was also reported by Kahn (1990).

Meanwhile, authentic leadership could affect work engagement indirectly, through occupational self-efficacy and interpersonal trust. According to social learning theory, leaders' behavior serves as a guide to followers' behavior by role modeling. When leaders are more attractive and credible, the effect of their role modeling becomes more critical on changes of followers' behavior (Bandura, 1997). As discussed earlier, authentic leadership affects followers' behavior (e.g., level of work engagement) because leaders serve as role models through whom followers can build interpersonal trust and expand their knowledge and ability. Avolio et al. (2004) asserted that authentic leadership can improve followers' engagement through employees' trust and positive emotions. Therefore, this study expected employees' selfefficacy and interpersonal trust to act as mediators for the relationship between authentic leadership and work engagement.

Positively changed employees’ behaviors that are directly and indirectly affected by authentic leadership are also related to performance improvement. Employees who are engaged in their work and duties with positive work-related experiences are expected to achieve better performance with lower intentions to leave the organization (Bakker & Bal, 2010; Schaufeli & Bakker, 2004). With regard to the mediation role of work engagement in the relationship between employees' attitude (occupational self-efficacy and interpersonal trust) and role-based performance, this study used work engagement as an indicator of intrinsic motivation. Unlike extrinsic motivation, which comes from external factors such as money and compensation, intrinsic motivation comes from inside an individual, such as the sense of satisfaction in completing or even working on a job, and is linked to high quality work performance and job satisfaction. Since work engagement refers to a high level of vigor, dedication, and absorption to work, it can be expected that a high level of work engagement is positively associated with job performance, thus playing a mediating role between employees' attitude variables (occupational self-efficacy and interpersonal trust) and role-based performance.

These aforementioned studies emphasize the important influence of leadership on employee attitudes and behaviors that eventually affect employees’ organizational performance. Based on the discussions above, the following hypotheses were developed:

To obtain sample cases, this study first considered contacting Korean Industrial Complex Corp. (KICOX), which has six institutions across the country. Among the six branches, one branch was selected to obtain company and employee contact information. Consequently, the sample for this study was employees who worked in one of the industrial complexes in Korea. The potential participants of the survey were asked to respond to each of the questions that measured their perceptions of the following items: authentic leadership, occupational self-efficacy, interpersonal trust, work engagement, and role-based performance.

Approximately 2,500 Korean employees were selected as potential survey participants, and 365 participated in the survey (a return rate of 14.6%). Among the questionnaires obtained, 29 were excluded—28 had missing data and one was an outlier (case number 9). The cases with missing data and the outlier were not included in the final sample based on statistical procedures described below. The final sample consisted of 336 employees.

Of the 336 employees, 59.5% were male, with the largest group between 30 and 39 years of age (25.0%), followed closely by those 40-49 years of age (23.2%). In terms of job tenure, 24.7% had 3-5 years with their current job; 19.9% had 1-3 years on the job. With regard to size of company, 58.9% worked in conglomerates, whereas 41.1% worked in a small-to-medium sized enterprise (SME).

All constructs were measured using a 5-point Likert-type scale ranging from “strongly disagree”(1) to “strongly agree” (5). All instruments used were open to researchers, except the role-based performance questionnaire, whose developers required permission. This study obtained permission from the first author via email. First, to measure authentic leadership, the Authentic Leadership Inventory (ALI), which was recently developed and validated by Neider and Schriesheim (2011), was used. The ALI contains 14 items that can be categorized into 4 sub-factors: self-awareness (S), relational transparency (R), balanced processing (B), and internalized moral perspective (M). Neider and Schriesheim (2011) tested the ALI scales for internal consistency reliability and empirical factor structure with the data from MBA students. The MBA students provided their perceptions of two presidential candidates (McCain and Obama in 2008) regarding leadership styles. The results showed that Cronbach’s alpha for both datasets (McCain and Obama) ranged from .74 to .85, indicating acceptable internal consistency reliabilities (≥.70; Nunnally & Bernstein, 1994). This fourdimensional ALI was empirically validated in various studies in terms of internal consistency reliability (e.g., Cerne et al., 2013). In this study, the ALI scales had an internal consistency reliability of α=.93. Second, a short version of the occupational self-efficacy scale (Rigotti, Schyns, & Mohr., 2008) was used to measure employees’ occupational self-efficacy. In their study to validate the short version of the occupational self-efficacy scale containing 6 items across five countries (Germany, Sweden, Belgium, Britain, and Spain), Rigotti, Schyns, and Mohr (2008) showed that reliability of the occupational selfefficacy scale was between .85 (Belgium) and .90 (Britain), indicating a high reliability of construct. In this study, the occupational selfefficacy scale showed good reliability (α=.87). Third, the Interpersonal Trust at Work Scale (ITWC), which was developed by Cook and Wall (1980), was used to measure interpersonal trust in the workplace. The ITWC is comprised of 6 items and has two dimensions, trust in peers (3 items) and trust in management (3 items). In this study, ITWC had an internal consistency reliability of α=.87 for whole items. Fourth, the Utrecht Work Engagement Scale-9 (UWES-9), developed by Schaufeli, Bakker, and Salanova, (2006), was employed to measure work engagement. The scale is comprised of 9items with three dimensions: vigor, dedication, and absorption. Schaufeli et al. (2006) showed that Cronbach’s alpha of the total 9-item scale exceeded the value of .70, indicating reliability of construct. In this study, UWES-9 had an internal consistency reliability of α=.91. Finally, to measure role-based performance, this study employed the Role-based Performance Scale (RBPS), developed and validated by Welbourne, Johnson, and Erez (1998). This scale was designed to measure multidimensional aspects of employees’ performance that include job and non-job dimensions. This scale is composed of 20 items classified into 5 sub-categories: job (doing specifically required work), career (increasing their value by participating career development program and obtaining new skills), innovator (being creative and innovative in their job), team member (working with team members and co-workers), and organization citizenship behavior (doing work and helping others even though it is not required). According to Welbourne et al. (1998), Cronbach’s alpha for the instrument, which indicates internal consistency ranging from .86 to .96, and measurement construct validity were both satisfied. In this study, RBPS had an internal consistency reliability of α=.94.

Before conducting the main survey, it is important to conduct a pilot test to ensure the clarity of the instrument with the representative population. The pilot test is especially important in this kind of study that relies on questionnaires originally developed in one language and then translated into another. In this study, the researcher conducted a pilot test to make sure that all translated instruments were understandable for Korean workers. A total of 46 paper-based questionnaires were distributed to Korean workers, who were then asked to complete the questionnaires and also to provide their feedback regarding clarity of the instruments. Based on the feedback, a few minor changes were made: two language changes, the replacement of a word, and the deletion of an unnecessary word. Following these minor instrument changes, the decision was made to proceed with the main round of the survey.

Correlation analysis indicated acceptable intercorrelations among the latent variables at the p = 0.01 level, as shown in Table 1. However, the higher level of correlation coefficients among the latent variables could result in multicollinearity issues, which could threaten effective interpretations in further data analyses and mislead the investigators. The variance inflation factors (VIF), which reflect large VIF scores indicating the presence of a high degree of multicollinearity among the latent variables, and tolerance were used to detect multicollinearity. To avoid the issue of multicollinearity, tolerance should be greater than .20 (or .10) (O’Brien, 2007) and VIF should be less than 4 (Miles & Shevlin, 2001). In this study, VIF scores ranged between 1.74 and 2.03, and tolerance values ranged between .49 and .58. Thus, it can be concluded that multicollinearity was not a problem in this study.