1
Ethical leadership and follower’s
OCB/responsibility: the mediating and moderating
role of followers’ interdependent self-construal
Name Student: Keetie Varkevisser Student number: 10681930
Submission date:
Qualification: Executive Programme in Management Studies – Leadership track Institution: Amsterdam Business School
First supervisor: Dr. Annebel de Hoogh Second supervisor:
3
Table of contents
Abstract ... 4
1 Interdependent self-construal as a moderator and mediator in the relationship between ethical leadership and follower behavior and attitude ... 5
2 Theoretical framework ... 9
2.1 Ethical leadership ... 9
2.2 Ethical leadership and organizational citizenship behavior ... 11
2.3 Ethical leadership and responsibility ... 13
2.4 Interdependent self-construal ... 14
2.5 Ethical leadership and interdependent self-construal ... 15
2.6 Interdependent self construal and organizational citizenship behavior/responsibility ... 16
2.7 The mediating role of interdependent self-construal on the relationships between ethical leadership and organizational citizenship behavior/responsibility ... 17
2.8 The moderating role of interdependent self-construals on the relationships between ethical leadership and organizational citizenship behavior/responsibility ... 18
3 Method ... 20
3.1 Research method ... 20
3.2 Sample and procedures ... 20
3.3 Variables ... 21 3.4 Control variables ... 23 3.5 Statistical analyses ... 23 4 Results ... 26 4.1 Correlation analysis ... 26 4.2 Analyses ... 26
5 Discussion and conclusion ... 34
5.1 Research findings ... 34
5.2 Limitations and strengths ... 37
5.3 Theoretical and practical implications ... 39
5.4 Future research recommendations ... 40
5.4 Conclusion ... 42
6 References ... 43
4
Abstract
Ethical aspects of leadership gained a great deal of interest in research due to recent business scandals. Ethical leadership is linked to prosocial work behavior and attitudes; however, relatively little is known about these relationships. Previous theories and research findings suggest that follower’s interdependent self-construal may act as a mediator and moderator in the relationship between ethical leadership and follower behavior and attitude.
This study examines whether follower’s interdependent self-construal mediates and moderates the relationship between ethical leadership and organizational citizenship behavior and responsibility. A quantitative, cross-sectional research was conducted among personnel of the Dutch Ministry of Defence, which resulted in a sample of 145 respondents. Results
indicate, in line with expectations, that the relationships between ethical leadership and follower’s citizenship behavior is mediated and moderated by follower’s interdependent self-construal. Furthermore, the correlation between ethical leadership and responsibility is mediated by follower’s interdependent self-construal.
5
1 Interdependent self-construal as a moderator and mediator in the relationship between ethical leadership and follower behavior and attitude
Recent ethical scandals in organizations have raised questions about the role of leaders in shaping ethical conduct (Brown, Treviño, & Harrison, Ethical leadership: A social learning perspective for construct development and testing, 2005). Examples of unethical behavior in top management are illustrated by the scandals in the governing body of international football, the Fédération Internationale de Football Association (FIFA), involving bribery, racketeering, wire fraud, and money laundering over many years (Silverman, 2015). Sepp Blatter, former president of FIFA, did not set the proper tone and tolerated a culture of corruption within FIFA. As a result, his unethical leadership harmed the reputation of FIFA.
The above-mentioned example illustrates that ethical scandals in organizations can be detrimental, as they decrease trust in leaders and harm leader’s credibility and their potential to meaningfully influence followers at all levels in an organization (Brown et al., 2005; Piccolo, Greenbaum, Den Hartog, & Folger, 2010). Also, unethical behavior reduces the financial success of organizations (Orlitzky, Schmidt, & Rynes, 2003) and harms the organization’s reputation.
As a response to these developments, ethical behavior and morality have become important topics in organizations and in line with this, ethical issues gained much interest in the field of leadership research. Ethical leadership is thought to be essential because of the impact leaders may have on the behavior of others in the organization and on organizational performance (Aronson, 2001; Kanungo, 2001; Trevino, Brown, & Hartman, 2003). Business ethics literature acknowledged the importance of leaders for setting an ethical tone at the top of organizations (Murphy & Enderle, 1995) and their formal and informal ethical cultures (Treviño, 1990; Treviño & Nelson, 2004). By communicating ethical standards and the use of rewards and punishments, leaders can reinforce ethical conduct among employees (Treviño, Hartman, & Brown, Moral person and moral manager: how executives develop a reputation for ethical leadership, 2000). This leadership style is labelled as ‘ethical leadership’ (Brown et al., 2005).
Ethical leadership is expected to have positive effects on ethical conduct of employees and their attitudes (Aronson, 2001; Brown, Treviño, & Harrison, 2005; Kanungo, 2001; Treviño, Brown, & Hartman, 2003). For instance, evidence suggests that ethical leadership is positively associated with organizational citizenship behaviors (OCBs) (Brown & Treviño, 2006; Piccolo, Greenbaum, Den Hartog, & Folger, 2010; Kalshoven, Den Hartog, & Den
6
Hoog, 2011; Walumba & Schaubroek, 2009), such as helping (Mayer, Keunzi, Greenbaum, Bardes, & Salvador, 2009; Piccolo et al., 2010; Kalshoven et al., 2011), follower voice
(Walumbwa & Schaubroek, Leader personality traits and employee voice behavior: Mediating roles of ethical leadership and workgroup psychological safety, 2009), initiative, courtesy, and altruism (Kalshoven, 2012). Moreover, positive relationships are found between ethical
leadership and performance, effort and effectiveness of both individuals and groups (Piccolo et al., 2010; Walumba et al., 2011, 2012).
As previously mentioned, ethical leadership also enhances positive work-related attitudes. An example of such an attitude is follower’s responsibility. Research has revealed that responsibility is imperative in the ethical leadership process (Kalshoven, 2010; De Hoogh & Den Hartog, 2008). Although research on the relationship between ethical leadership and follower responsibility is scarce, it is likely that ethical leaders enhance follower
responsibility because such leaders demonstrate a sense of responsibility at work, which followers may copy (Kalshoven, 2010).
Thus far, previous research showed demonstrated positive correlations between ethical leadership and follower citizenship behaviors, however, relatively little is known about
possible mediators or moderators of this relationship. In other words, research is needed to investigate how ethical leadership is positively related to follower citizenship behaviors and in which context ethical leaders are likely to positively influence these behaviors. For instance, current work does not yet fully explain why some people react stronger than others to the influence of an ethical leader. Nevertheless, it is no surprise that several authors have called for research on this subject (Brown & Treviño, 2006; De Hoogh & Den Hartog, 2009).
Possible mediators and moderators between ethical leadership and work behavior can be found in concepts about the ‘self’ (Van Knippenberg, Van Knippenberg, Cremer, & Hogg, 2004). Concepts about the ‘self’ are often operationalized in terms of self-construals. Self-construals refer to ´constellations of perceptions, feelings and actions concerning the extent in which the self is defined independently of others or interdependently with others’(Markus & Kitayama, 1991). Previous research indicated that self-construals may be affected by
leadership and may mediate the effects of leadership on follower behavior ((Van
Knippenberg, Van Knippenberg, Cremer, & Hogg, 2004; De Cremer & Van Knippenberg, 2002; Kark, Shamir, & Chen, 2003; Conger, Kanungo, & Menon, 2000). To illustrate this, Lord et al. (1999) suggested that, through a certain style of leadership, leaders activate a corresponding level of self-construal in the follower, which will also affect the follower’s
7
goals, self-views, and perceptions of the self in the future (Van Gils, Van Quaquebeke, & Van Knippenberg, 2010).
Self-construals may not only be affected by leadership, it may also affect perceptions of and responses to situational stimuli (Leary & Tangney, Handbook of self and identity, 2003). In line with this, research findings proposed that follower self-construal can act as a moderator of leadership processes (Hogg, 2001; Hogg & Van Knippenberg, 2003; Lord & Brown, 2004; Lord, Brown, & Freiberg, 1999; Van Knippenberg & Hogg, 2003; Martin & Epitropaki, 2001). Evidence for this finding can be found in research on transformational leadership. To illustrate, transformational leaders articulate an attractive and realistic vision of the future that can be shared, stimulate followers to observe old problems in new ways, support
subordinates, and pay attention to the differences among them (e.g. Bass, 1985; 1997; 2008). Studies on transformational leadership revealed that follower’s self-construal affects the outcomes of leadership on follower behavior. Furthermore, Martin and Epitropaki (2001) suggest that followers with an interdependent self-construal react stronger to the influence of a transformational leader than individuals with an interdependent self-construal, because this type of leader appeals to the collective self.
Based on these research findings, Van Knippenberg et al. (2004) suggest to develop theory about the interplay of the various aspects of follower self-conception in leadership processes and specify an agenda for future research in this area. For this reason, this study aims to contribute to current research in leadership processes and follower outcomes, by studying the mediating and moderating effects of follower’s interdependent self-construal on the relationship between ethical leadership and follower’s citizenship behavior and
responsibility. Moreover, this research also answers the call for research on mediators and moderators to shed more light on how ethical leadership relates to follower citizenship behaviors and attitudes (Brown & Treviño, 2006; De Hoogh & Den Hartog, 2009; Van Knippenberg et al., 2004).
The research question of this study is as follows:
To what extent does follower’s interdependent self-construal mediate and moderate the relationship between ethical leadership and follower’s citizenship behavior and
9
The following research models describe the relationships between ethical leadership and organizational citizenship behavior and responsibility. It also illustrates the mediating (research model I) and moderating (research model II) role of interdependent self-construal.
Figure 1. Research model I (mediation)
The subsequent chapter describes the theoretical background and provides hypotheses. The third chapter explains the used research methods. In addition, results of this study are presented in the fourth chapter. Finally, this research paper ends with a discussion and conclusion. Ethical leadership Organizational Citizenship Behavior Responsibility Interdependent Self-construal
Figure 2. Research model II (Moderation)
Organizational Citizenship Behavior Responsibility Ethical leadership Interdependent Self-construal
10
2 Theoretical framework
2.1 Ethical leadership
Ethical scandals in organizations triggered the reconsideration of the leadership role in shaping ethical behavior. Therefore, the concept of ethical leadership gained much interest in scientific literature. Brown & Treviño (2006) developed an ‘ethical leadership’ construct and defined it as ‘the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct to followers through two-way communication, reinforcement and decision-making’.
Treviño et al. (2000;2003) found a number of personal characteristics related to ethical leadership. For instance, ethical leaders are considered trustworthy and honest. Furthermore, ethical leaders are regarded as fair and principled decision-makers who care about people and the broader society. Researchers, moreover, labelled these characteristics as the moral person dimension of ethical leadership, representing observers’ perceptions of the leader’s personal traits, character, and altruistic motivation (Brown & Trevino, 2006). In addition, ethical leaders make proactive efforts to influence follower’s ethical conduct by communicating values, by role-modelling, and by using a reward system (Brown & Treviño, 2006). These actions represent the moral manager dimension of ethical leadership (Brown & Trevino, 2006), which entails visible behaviors that do not necessarily flow only from personal traits (Brown, Trevino, & Harrison, 2005). Kalshoven, Den Hartog, and De Hoogh (2011) build on the research by Brown and Treviño and distinguished several ethical leader behaviors, based on theory, interviews, and a student sample. These behaviors include: fairness, integrity, ethical guidance, people orientation, power sharing, role clarification, and concern for
sustainability (Kalshoven, Den Hartog, & De Hoogh, 2011). Table 1 provides descriptions of these ethical leader behavior dimensions.
11
Ethical leadership can be explained by several behavioral approaches such as social learning (Bandura, 1977; 1986), social exchange (Blau, 1964), and social identity (Hogg, A social identity theory of leadership, 2001). First, social learning is a theory that highlights that people learn from role models, rewards, and punishments. For instance, role-modelling refers to observational learning, imitation, and identification (Brown, Trevino, & Harrison, 2005) and employees can learn what behavior is expected, rewarded, and punished (Brown, Trevino, & Harrison, 2005). Thus, ethical leaders influence followers by visibly and intentionally role-modelling of ethical conduct and by communicating ethics and values (Brown & Treviño, 2006).
Second, ethical leadership is expected to affect followers through social exchange processes (Blau, 1964). Social exchange is based on the norm of reciprocity (Gouldner, 1960), which indicates that if one exchange partner does something beneficial for the other, it
generates an obligation to reciprocate good behavior (Cropanzano & Mitchell, 2005). In other words, followers are willing to reciprocate when treated fairly and are, in response, willing to make an effort (Kalshoven, Den Hartog, & De Hoogh, 2011).
Third, social identity theory explains effects of ethical leaders on followers. This theory states that individuals prefer to belong to a group, because the group membership offers information on the appropriateness of one’s attitudes and values (Hogg, A social identity theory of leadership, 2001) and elucidates how ethical leaders influence followers via identity effects. Ethical leaders highlight ethical values and this communicates identity
relevant information to followers. This information is likely to influence the way followers feel about themselves and the way they act, and it will enhance the distinctiveness of group values (Kalshoven & Den Hartog, 2009). Ethical leaders can also enhance greater
identification with the organization, as they represent the organization’s ethical standards and values (Van Knippenberg, Van Knippenberg, Cremer, & Hogg, Leadership, self, and identity: A review and research agenda, 2004).
In addition to identity information, group prototype is another central element of social identity theory. Leaders are usually most representative of the in-group, attract most attention, and therefore have most influence in a work group (Hogg, A social identity theory of
leadership, 2001). Research proved that followers identify with their leaders, because these leaders are regarded as group prototypical, and this will enhance the followers’ trust in their leaders (Kalshoven & Den Hartog, 2009). Several researchers found other perspectives on ethical leadership. For instance, De Hoogh and Den Hartog (2009) stressed that ethical leaders are socially responsible for the use of power. They stated that ethical leaders influence
12
people’s activities toward goal achievement in a social responsible way (De Hoogh & Den Hartog, Ethical leadership: The positive and responsible use of power, 2009). Similarly, Dickson, Smith, Grojean, and Ehrhart (2001) found that ethical leaders can shape an ethical climate. Previous studies have linked ethical leadership with positive effects on followers such as a higher perceived effectiveness of the leader and more job satisfaction (Brown, Trevino, & Harrison, 2005), more trust and less cynicism (Kalshoven, Den Hartog, & De Hoogh, 2011), more optimism about the future (De Hoogh & Den Hartog, 2008), and more meaningfulness of the work and motivation for the work (Piccolo, Greenbaum, Den Hartog, & Folger, 2010). The following section explains this further in detail.
2.2 Ethical leadership and organizational citizenship behavior
Recent research demonstrated that ethical leadership has positive effects on the behaviors of others in organizations (Aronson, 2001; Brown, Trevino & Harrison, 2005; Kanungo, 2002). For example, followers of ethical leaders display more work engagement and initiative (Den Hartog & Belschak, 2012), more helpfulness (Kalshoven, Den Hartog, & De Hoogh, 2011), and a higher willingness to report problems (Brown, Trevino, & Harrison, 2005). These behaviors are often operationalized in terms of OCB (Kalshoven, 2010).
Organ (1990) described OCB as ‘behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization’. These behaviors improve the social and
psychological context within which work tasks are performed (Borman & Motowidlo, 1993). Examples of OCBs are helping, keeping up with internal organizational information, taking the consequences for others of own actions into account, and being loyal (Organ, 1988; Podsakoff, MacKenzie, Paine, & Bachrach, 2000).
Moreover, Coleman and Borman (2000) defined three dimensions of OCB, based on twenty-seven citizenship performance behaviors generated from previous research. The first dimension, interpersonal citizenship performance, refers to behaviors that assist, support, and develop organization members through cooperative and facilitative efforts that go beyond expectations (Coleman & Borman, 2000).
The second dimension, organizational citizenship performance, denotes citizenship behaviors that demonstrate commitment to the organization through allegiance and loyalty to the organization and its objectives, and compliance with organizational rules, policies, and
13
procedures (Coleman & Borman, 2000). Promoting the organization and suggesting organizational improvements are examples of these behaviors.
The final dimension, job/task citizenship performance, describes extra efforts that go beyond role requirements; demonstrating dedication to the occupation, persistence, and the desire to maximise own employment performance. Hence, employees exerting extra effort to one’s their own occupation and engaging in self-development to improve their own effectiveness.
Literature on ethical leadership indicated that ethical leadership stimulates prosocial or OCB through social learning (Bandura, Social foundations of thought and action, 1986) and social exchange processes (Brown & Trevino, 2006). As mentioned before, social learning theory indicates that individuals learn from role models, rewards, and punishments. Ethical leaders act as role models by showing appropriate conduct and, concern about others, and they stimulate helping behavior for the good of the group (Kalshoven, Den Hartog, & De Hoogh, 2011). Furthermore, ethical leaders exhibit desired behavior by fair treatment and two-way communication. As a result, followers are likely to copy such behavior and show OCB (Kalshoven, Den Hartog, & De Hoogh, 2011). In addition to role-modelling, ethical leaders also make use of transactional elements, such as rewards and punishments, to enhance subordinates’ ethical and appropriate behaviors, which may include OCB (Brown, Trevino, &
Harrison, 2005).
Based on social exchange theory, ethical leadership is expected to affect follower behaviors by means of socioemotional exchange since ethical leaders enhance trust and are perceived as fair (Mayer D. , Kuenzi, Greenbaum, Bardes, & Salvador, How low does ethical leadership flow? Test of a trickle-down model, 2009). When ethical leaders treat employees in a fair way, employees know that they can count on them, and are therefore more likely to reciprocate this treatment and refrain from behaviors which could be detrimental to the organization (Mayer D. , Kuenzi, Greenbaum, Bardes, & Salvador, How low does ethical leadership flow? Test of a trickle-down model, 2009). In other words, employees who are treated fair and with concern of their leaders, are likely to reciprocate such treatment by
showing OCBs. Although empirical evidence is
scarce (Kalshoven, Den Hartog, & De Hoogh, 2011), several studies discovered a positive relationship between ethical leadership and OCB (Brown & Trevino, 2006; Mayer, Kuenzi, Greenbaum, Bardes, & Salvador, 2009; Piccolo, Greenbaum, Den Hartog, & Folger, 2010; Kalshoven, Den Hartog, & De Hoogh, 2011; Walumba & Schaubroek, 2009). For instance, Mayer et al. (2009) suggested that ethical leadership is positively related to group-level helping. In agreement with this, Piccolo et al. (2010) established a positive correlation
14
between ethical leadership and helping, rated by co-workers. Based on the above-mentioned findings and evidence, we propose that ethical leadership is positively related to followers’ OCBs.
Accordingly, the following hypothesis is formulated:
Hypothesis 1: Ethical leadership is positively related to OCB.
2.3 Ethical leadership and responsibility
Responsibility can be defined as being concerned about one’s effect on others, taking
responsibility for one’s actions, and behaving with dependability (Blasi, 1983; Winter, 1991). This suggest that individuals who take more responsibility are able to control their behavior, which implies they can fairly be blamed for negative actions or appreciated for positive ones (Winter, 1991). Prior research showed that individuals differ in the amount of responsibility they take at work (Frese & Fay, 2001). Essential in developing a sense of responsibility are
learning experiences (Winter, 1992).
Although research on ethical leadership and responsibility is limited, research findings indicate that ethical leadership is related to follower’s demonstration of responsibility. To illustrate, Spreitzer (1995) indicates that followers of leaders who exhibit more power sharing and who feel empowered by their leaders tend to display more responsibility in their
employments. In addition, according to Kalshoven, Den Hartog, and De Hoogh (2011), power sharing and the empowerment of followers (people orientation) are ethical leadership
behaviors. Similar research outcomes are found by Kalshoven (2013), who suggests that ethical leaders stimulate helping and initiative behavior by building social relationships and
enhancing the followers’ level of responsibility. Building on this insight, it is probable that ethical leadership augments followers’ responsibility. Social
exchange theory, moreover, can be used to explain this relationship; it is anticipated that employees will reciprocate leaders who clarify responsibilities and share power by acting in a responsible manner themselves. Masterson and Stamper (2003) also suggested that
demonstration of responsibility is regarded as an exchange mechanism with which followers
can reciprocate ethical treatment. In addition
to social exchange theory, we propose that social learning theory also provides an explanation for the relationship between ethical leadership and responsibility. According to social learning
15
theory, we expect that subordinates learn by observing and emulating leaders who display responsibility at work. Ethical leaders model responsibility, emphasise responsibility in their communication, and reward subordinates who exhibit responsibility at work.
Based on earlier research findings and social exchange and social learning mechanisms, we thus expect ethical leadership to relate positively to followers’ responsibility. This assumption leads to the following hypothesis:
Hypothesis 2: Ethical leadership is positively related to followers’ responsibility.
2.4 Interdependent self-construal
As previously pointed out, ethical leadership is known because of its positive effects on follower behaviors and attitudes (Brown et al., 2005; Treviño et al., 2003). These positive effects are frequently attributed to the leader’s influence (Yukl, 1998); nonetheless, leadership research has much to gain by research on the psychological effect of leadership on followers (Hunt, 1999; Lord & Brown, 2004). In recent years, the interest in the followers’ role in these psychological processes has increased, with special attention for follower’s self-concept and identity (e.g. De Cremer & Tyler, in press; Hogg & Van Knippenberg, 2003; Lord & Brown, 2004; Lord, Brown, & Freiberg, 1999; Shamir, House, & Arthur, 1993, Van Knippenberg & Hogg, 2003). These follower’s concepts are often operationalized in terms of self-construals.
The concept of self-construals was first introduced by Markus and Kitayama (1991), who defined it as a ‘constellation of perceptions, feelings and actions concerning the extent in which the self is defined independently of others or interdependently with others’. Self-construals are personality traits which coexist in individuals and shape their conduct (Markus & Kitayama, 1991). Self-construals are highly dynamic, so that the specific content of the self-concept is dependent on the situation (Van Knippenberg, Van Knippenberg, Cremer, & Hogg, Leadership, self, and identity: A review and research agenda, 2004). Depending on these situational factors, some individuals tend to use self-construal more than others to guide their behavior (Markus & Kitayama, 1991; Kanagawa, Cross, & Markus, 2001).
The definition of Markus and
Kitayama (1991) differentiates construals in independent and interdependent self-construals. While independent self-construal emphasises distinctiveness and individual uniqueness, the interdependent self-construal stresses social relations and interconnectedness
16
with others (Hoyt & Price, 2013). In this research, the focus is on interdependent self-construals.
Individuals with highly developed interdependent self-construals are attentive to others’ feelings and unexpressed thoughts (Singelis, 1994). Furthermore, to enhance their
interdependent view of the self, individuals tend to think and behave in ways that underscores their connectedness with others and that strengthen existing relationships (Cross, Bacon, & Morris, The relational-interdependent self-construal and relationships, 2000). Their most important tasks are to fit-in with their in-groups, act in an appropriate fashion, promote their in-groups’ goals, occupy their proper places, be indirect, and read in-group’ minds (Markus & Kitayama, 1991). Individuals with an interdependent self-construal focus on duties,
obligations, and harmonious social responsibilities (Cristina-Corina, 2012). Interdependent self-construals can, moreover, have specific cognitive, affective and behavioral consequences (Lord, Brown, & Freiberg, 1999). As such, interdependent self-construals can be linked to OCB and responsibility. Hence, it is expected that individuals with an interdependent self-construal are likely to display more organizational work behavior and responsibility, because they feel connected to the organization. For this reason, they also desire to contribute to organizational goals. The subsequent section discusses this in more detail.
2.5 Ethical leadership and interdependent self-construal
As earlier mentioned, self-construals are dynamic (Van Knippenberg, Van Knippenberg, Cremer, & Hogg, Leadership, self, and identity: A review and research agenda, 2004), indicating that the strength or salience of different self-construals may vary across time, situations, group membership, and relationships (Aron, 2003; Brewer, 2003, Hogg, 2003; Sedikes & Brewer, 2001; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). In line with this description, Van Gils et al. (2010) pointed out that levels of self-construal can be activated by another person. Moreover, Van Knippenberg et al. (2004) argued that leaders may affect which aspects of the self are salient. Lord et al. (1999; 2001) also suggested that through certain styles of leadership, leaders activate a level of self-construal which corresponds to the displayed leadership style (Van Gils, Van Quaquebeke, & Van Knippenberg, 2010). This argument is supported by Paul et al. (2001) who found that communication inspired by
17
idealized influence, which is a part of charismatic leadership, made collective self-construal
more salient. Based on these research findings,
we suggest that ethical leadership is positively related to followers’ interdependent self-construal.
Therefore, we expect that ethical leaders will activate followers’ interdependent self-construal by showing behavior that refers to the ‘people orientation’ dimension of ethical leadership behavior (Kalshoven, Den Hartog, & De Hoogh, 2011). In other words, follower’s
interdependent self-construal will be positively affected by an ethical leader who emphasises connectedness with others by showing concern, respect, and support to followers. Thus, the third hypothesis is formulated as follows:
Hypothesis 3: Ethical leadership is positively related to follower’s interdependent
self-construal.
2.6 Interdependent self-construal and organizational citizenship behavior and responsibility
Morris (2001) stated that the ways in which people evaluate themselves may have noteworthy outcomes at their occupation. In addition, Organ (1990) argued that individual differences play an important role in predicting OCB. As such, it is conceivable that an interdependent self-construal might be an important construct in understanding behavior at work. This
argument is supported by Moorman and Blakely (1995) who found that collectivism, which is a construct similar to the interdependent self-construal, is a predictor of OCB. Therefore, we expect that individuals high on interdependent self-construal tend to engage in helping behavior and demonstrate loyalty with the goal to maintain interconnectedness with others. By exhibiting loyalty and helping behavior, individuals with an interdependent self-construal display higher levels of organizational citizenship behavior than those with an independent self-construal. This results in the following hypothesis:
Hypothesis 4: Follower’s interdependent self-construal is positively related to follower OCB.
In line with findings of Morris (2001) and Organ (1990), Cross and Madson (1997) emphasise the role of interdependent self-construal in shaping human thoughts, feelings, and behavior and argue that self-construals partially determines responsibility. For this reason, we expect that individuals high on interdependent self-construal display more responsibility. As
18
and behaving with dependability (Blasi, 1983). Hence, individuals with an interdependent self-construal will maintain stable relationships with others and are motivated to fit in with their organization. Moreover, fulfilling their duties and responsibilities is consistent with their need to fit in the organization and will aid them to foster stable relationships with others. Therefore, we expect that interdependent self-construal is positively related to responsibility, which leads to the following hypothesis:
Hypothesis 5: Follower’s interdependent self-construal is positively related to responsibility.
2.7 The mediating role of interdependent self-construal on the relationships between ethical leadership and organizational citizenship behavior and responsibility
Van Knippenberg et al. (2004) stated that to understand leadership, it is vital to understand leadership effects on followers. However, leadership research has generally focused on the leader itself than on the possible effects on followers. As a reaction to this, several authors indicated the need for research on how ethical leadership relates to follower behaviors and attitudes (Brown & Treviño, 2006, De Hoogh & Den Hartog, 2009; Van Knippenberg et al., 2004). To clarify this mechanism, Van Knippenberg et al. (2004) asserted that follower’s self-construal may mediate the effect of leadership on follower behavior. Despite this suggestion, studies that examined the way in which interdependent self-construals mediate the
relationship between ethical leadership and OCB and responsibility are scarce. Nevertheless, Lord et al. (2001;1999) found that self-construals can act as a mechanism through which ethical leaders affect follower behavior and attitudes. Their research indicates that there are different styles of leadership at the various levels of self-identity (Van Gils, Van Quaquebeke, & Van Knippenberg, 2010). To illustrate, through a certain style of leadership, leaders can activate a corresponding level of self-construal in the follower, which consequently influences the follower’s behavior and attitude (Van
Knippenberg, Van Knippenberg, Cremer, & Hogg, Leadership, self, and identity: A review and research agenda, 2004). Thus, leadership behavior that is focused on the collective will arouse corresponding levels of self-construal in followers, which in turn will lead to
corresponding behaviors on the side of the followers (Van Knippenberg, Van Knippenberg, Cremer, & Hogg, Leadership, self, and identity: A review and research agenda, 2004). Based on these research findings, it is anticipated that ethical leaders will make follower’s
19
interdependent self-construals more salient, which results in OCB and responsibility. Hence, this results in the following hypotheses:
Hypothesis 6: Follower’s interdependent self-construal mediates the positive relationship
between ethical leadership and follower’s OCB.
Hypothesis 7: Follower’s interdependent self-construal mediates the relationship between
ethical leadership and responsibility.
2.8 The moderating role of interdependent self-construal on the relationships between ethical leadership and organizational citizenship behavior and responsibility
Self-construal may not only act as a mediator in the relationship between leadership and follower behaviour, it can also affect perceptions of and responses to, situational stimuli (Leary & Tangney, Handbook of self and identity, 2003). Furthermore, Van Knippenberg et al. (2004) and Lord et al. (1999) indicate that follower’s self-construal can result in different behaviors and responses to leadership effectiveness. Several leadership studies, which examined the moderating role of self-construal, also found evidence for this argument (De Cremer & Van Vugt, 2002; Hains, Hogg, & Duck, 1997; Lord & Brown, 2004; Platow & Van Knippenberg, 2001). For example, studies on transformational leadership indicate that
interdependent self-construals have a moderating effect on leadership outcomes. Moreover, converting follower’s motivation from self-interest to collective interest has been proposed to be the essence of transformational leadership (Van Knippenberg, Van Knippenberg, Cremer, & Hogg, 2004; Bass, 1985; Burns, 1978; House, 1977; Shamir et al., 1993).
In addition, a moderating effect of interdependent self-construal on transformational leadership was found in studies of Lord and Brown (2004) and Martin and Epitropaki (2001), which designated that followers with an interdependent self-construal react stronger to the influence of a transformational leader than individuals with an independent self-construal, as this type of leader appeals to the collective self. In other words, transformational leaders are more effective for followers who value interdependent self-construal.
Explanations for the moderating influence of self-construals on leadership effectiveness can be found in leader group prototypicality and leader group-orientedness (Hogg & Van Knippenberg, 2003; Van Knippenberg & Hogg, 2003). First, group
20
representative of the group’s identity (Van Knippenberg et al., 2004). Individuals with an interdependent self-construal attach importance to social relationships and interconnectedness with others, and this is an important reference point for their attitudes, beliefs, norms, and values (Hoyt & Price, 2013). As a consequence, leaders who are group-prototypical are especially influential and attractive (Van Knippenberg et al., 2004).
Second, leader group-orientedness refers to the extent to which the leader has the group’s best interest at heart. Interdependent self-construals also lead people to experience the in-group’s interest as self-interest, and to endorse others who also take the in-group’s best interest to heart (Van Knippenberg et al., 2004). Based on this, it is probable that interdependent self-construals are enhancers or neutralisers for leadership outcomes. Here, we propose an
enhancer role for interdependent self-construals in the relationship between ethical leadership and OCB and responsibility. In short, it is expected that the effect of ethical leadership on OCB and responsibility is stronger on interdependent followers. Consequently, the following hypotheses are formulated:
Hypothesis 8: Followers’ interdependent self-construal moderates the relationship between
ethical leadership and followers’ OCB, so that the relationship between ethical leadership and followers’ OCB is stronger when followers’ interdependent self-construal is higher.
Hypothesis 9: Followers’ interdependent self-construal moderates the relationship between
ethical leadership and followers’ responsibility, so that the relationship between ethical leadership and followers’ responsibility is stronger when followers’ interdependent self-construal is higher.
21
3 Method
3.1 Research method
A quantitative, cross-sectional research method was used to answer the research question:
To what extent does follower’s interdependent self-construal mediate and moderate the relationship between ethical leadership and follower’s citizenship behavior and
responsibility?
Quantitative research infers evidence for a theory through measurement of variables that produce numeric outcomes (Field, 2013). Additionally, statistical techniques are used to describe the results and test expectations on the outcome (Verhoeven, 2007). This research can also be characterised as cross-sectional, which means that variables are measured at a single time point (Field, 2013). For this study, questionnaires were distributed to collect data to measure the following variables in the research model: ethical leadership, interdependent self-construal, OCB, and responsibility. Questionnaires are useful for measuring opinions, attitudes, and knowledge of a large population (Verhoeven, 2007).
3.2 Sample and procedures
Potential participants, working at the Dutch Ministry of Defence, were approached and asked to participate in this research by completing a questionnaire (Appendix I). Moreover, the researcher’s contacts were requested to participate in this research on a voluntary basis. The participants had the opportunity in to fill in the questionnaire on paper or online. Both the paper and the online questionnaire included a short description of the research, an assurance for the confidential treatment of information, the researcher’s contact information, and an instruction for completing the questionnaire.
When participants completed the questionnaire on paper, the questionnaires were returned anonymously in a sealed envelope. In total, 57 questionnaires were returned from a total of 81. The data of these questionnaires was processed in the SPSS data file. Furthermore, other participants received an email asking to complete the online questionnaire. This email contained a link to the questionnaire in Qualtrics, a web-based software program for
conducting online questionnaires (Unknown, 2015). From a total of 125, approximately 90 participants filled in the online questionnaire.
22
In total, 147 participants filled in the questionnaire. The final working sample consisted of 145 respondents, as two questionnaires were removed due to incompleteness. The mean age of the respondents was 42 years, where the youngest respondent was 18 years old and the oldest 62 years old. Moreover, 35.2% was female and 64.8% was male. Of the respondents, 43.4% have a degree of intermediate vocational education (MBO), 42.4% possess a degree of higher professional education (HBO), and 6.2% hold a university degree. Respondents worked in different departments of the Dutch Ministry of Defence, including Royal
Netherlands Airforce, Royal Netherlands Army, Royal Netherlands Navy, Royal Netherlands Marechaussee, and the Support Command. Most respondents have a contract with the Support Command (59%), followed by the Royal Netherlands Airforce (15.9%), the Royal
Netherlands Army (14.5%), the Royal Netherlands Navy (6.9%), and finally the Royal Netherlands Marechaussee (3.5%). Moreover, the majority of the respondents worked as a civilian employee (62.3%) and the minority as a serviceman or servicewoman (37.7%). The mean tenure with the Dutch Ministry of Defence is 4.24 years (SD = 2.14), while the mean tenure with the leader is 1 year and 11 months (SD = 21.83). Data collected from the online questionnaire were automatically processed in a SPSS data file, which saved time
(Verhoeven, Collecting date, 2007).
3.3 Variables
The questionnaires contain scales that measure ethical leadership, interdependent self-construal, OCB, and responsibility. All ratings were made on five-point scales (1 = strongly disagree; 5 = strongly agree).
Ethical leadership Scale
The Ethical Leadership Scale (ELS), which contains 10 items, was used to asses ethical leadership (Brown, Treviño, & Harrison, Ethical leadership: A social learning perspective for construct development and testing, 2005). Examples of question items were as follows: ‘My supervisor listens to what employees have to say’ and ‘My supervisor makes fair and
23
Self-construal scale
The self-construal scale of Gudykunst and Lee (2003) can be used to measure both
independent and interdependent self-construal. As previously mentioned, this research only focusses on the interdependent self-construal. Therefore, only the items that measure ‘interdependent self-construal’ were selected and added to the questionnaire. The scale contained 14 items that identify an interdependent self-construal. Questions items were, for example, ‘I consult others before making important decisions’ and ‘I respect decisions made by my group’. Cronbach’s alpha for this scale was 0,75.
Organizational Citizenship Behavior
Organizational citizenship behavior was measured using Coleman and Borman’s (2000) 27-item scale of OCB. The scale was divided into the following clusters: Interpersonal
Citizenship Performance, Organizational Citizenship Performance, and Job/Task Citizenship Performance.
First, Interpersonal Citizenship Performance consisted out of two subclusters: Interpersonal Altruism (4 items) and Interpersonal Conscientiousness (4 items). The
statements were, for instance, ‘I help other members in the organization’ and ‘I keep others in the organization informed about upcoming events, activities, actions, etc.’. Cronbach’s alpha for this cluster was 0.77.
Second, Organizational Citizenship Performance had two subclusters, which were Organizational Allegiance/Loyalty (6 items) and Organizational Compliance (6 items). Examples of items were: ‘I do not complain about organizational conditions’ and ‘I
demonstrate respect for organizational rules and policies’. Furthermore, Cronbach’s alpha for this cluster was 0.85.
The final cluster, Job/Task Citizenship, contains 7 items. Examples of these items were ‘I put extra effort in my own job’ and ‘I display dedication on the job’. Here, Cronbach’s alpha for this cluster was 0.86.
Responsibility
To assess responsibility, 4 items were adapted from Hackman and Oldham’s (1980) Job Diagnostic Survey. By means of this survey, the items measure the experienced responsibility by employees for outcomes of work (Hackman & Oldham, 1980). Examples of items were: ‘It
24
is hard, on this job, for me to care whether or not the work gets done right’ and ‘I feel a very high degree of personal responsibility for the work I do on this job’.
At first, Cronbach’s alpha of this scale was 0.47. Yet, after removing the item ‘It is hard, on this job, for me to care whether or not the work gets done right’, Cronbach’s alpha was 0.76.
3.4 Control variables
Several control variables were used to measure the characteristics of the working sample. To illustrate, the most common characteristics were taken into account such as age, gender, highest completed education, type of contract, department, tenure with organization, and tenure with leader.
Research findings suggested that some control variables are associated with ethical behavior and OCB. For example, Kidwell et al. (1987) pointed out that longer tenure with the leader is related to increased levels of ethical response and tenure with leader and
organization correlates with OCB (Morrison, 1997). In this study, age, gender, and education were also included as control variable, despite findings of Brown et al. (2005), who argued that employee reports of their supervisor’s ethical leadership is not tainted by personal characteristics.
3.5 Statistical analyses
SPSS 22.0 was used to analyse the data. The data was checked on linearity and normality to justify the use of linear regression models. Furthermore, linear regression analysis was conducted to test the first five hypotheses, while the last four hypotheses were tested with regression analyses, conducted in Process (version 2.13). Process is a tool for path analysis-based moderation and mediation analysis (Hayes, 2013) and includes a bootstrap technique for testing the statistical significance of indirect effects (Preacher & Hayes, 2004).
Bootstrapping is a non-parametric approach, which means that it makes no assumptions about the shape of the distributions of the variables or the sampling distribution of the statistics. This technique estimates the sampling distribution of a statistic by taking repeated samples the dataset (Field, 2013). Subsequently, the statistics of interest are calculated for each sample, from which the sampling distribution of the statistic is estimated (Field, 2013). Finally, the
25
standard error of the statistic is estimated as the standard deviation of the sampling distribution created from the bootstrap samples (Field, 2013).
An advantage of this technique is that it is suitable for small and moderate samples with more
confidence (Preacher & Hayes, 2004).
In this research, the sixth and seventh hypotheses were tested with the use of model 4 in Process. The following variables were fed into the dialogue box to test hypothesis 6: OCB as an outcome variable, ethical leadership an independent variable, and interdependent self-construal as a mediator variable. This is illustrated in Figure 3.
The seventh hypothesis was tested using responsibility as an outcome variable, ethical leadership as an independent variable, and independent self-construal as a mediator variable, which is presented in Figure 4.
Ethical leadership Interdependent self-construal Organizational citizenship behavior
Figure 3. Mediation model ELS - ISC - OCB
Ethical leadership
Interdependent self-construal
Responsibility
26
For both models, bias-corrected bootstrap confidence intervals of 1,000 samples were used to estimate the direct and indirect effects. Furthermore, model 1 in Process is used to test
hypotheses 8 and 9.
For hypothesis 8, OCB was the outcome variable, ethical leadership was the independent variable, and interdependent self-construal was the moderating variable; this can be seen in Figure 5.
For hypothesis 9, OCB was the outcome variable, ethical leadership was the independent variable and interdependent self-construal was the moderating variable (Figure 6).
Ethical leadership Interdependent self-construal Organisational citizenship behavior
Figure 5. Moderation model ELS - ISC- OCB
Ethical leadership
Interdependent self-construal
Responsibility
27
Again, bias-corrected bootstrap confidence intervals of 1,000 samples were used for these models to calculate the direct and indirect effects.
28
4 Results
4.1 Correlation analysis
Table 2 displays descriptives, correlations, and reliability coefficients for each scale. The scales for ethical leadership, interdependent self-construal, OCB, and responsibility have high reliabilities, as all Cronbach’s α are above 0.7. Moreover, none of the control variables
demonstrated significant relationships with ethical leadership, interdependent self-construal, OCB, and responsibility. Ethical leadership had a significant and positive relationship with interdependent self-construal (r=.485, p<0.01), OCB (r=.465, p<0.01), and responsibility (r=.223, p<0.01). Furthermore, interdependent self-construal is significantly related to organizational citizenship (r=.536, p<0.01) and responsibility (r= 0.272, p <0.01).
Table 2. Descriptives, correlations, and reliabilities of variables
4.2 Analyses
A linear regression analysis was performed to test if ethical leadership is positively related to OCB. As such, it was established that ethical leadership significantly predicts OCB (β = .465,
SE = .060, p < .001). The control variables, however, did not have a significant impact on the
model’s ability to predict OCB. Ethical leadership also indicated a significant proportion of variance of 19.3% in OCB scores (F=7.782, p < 0.001). Table 3 provides the results of the linear regression analysis for ethical leadership and OCB.
Variable s M SD 1 2 3 4 5 6 7 8
1. Age (years) 42.37 9.49
2. Gender 1.35 0.48 .157
3. Tenure with leader (in months) 22.65 21.83 .319** .076 4. Tenure with Ministry of Defence 4.24 2.14 .781** .280** .218**
5. Ethical leadership 3.96 .46 .036 .076 .001 -.007 (.86) 6. Interdependent Self-Construal 3.81 .33 .100 -.011 -.046 -.012 .485** (.76) 7. Organizational Citizenship Behavior 3.99 .36 .074 .018 .037 .058 .465** .536** (.92) 8. Responsibility 4.24 .53 .029 .019 .025 .010 .223** .272** .584** (.76) Notes. **Correlation is significant at the 0.05 level (2-tailed); Cronbach alphas are in parentheses
29
Table 3. Results linear regression for ethical leadership and organizational citizenship behavior
Additionally, the second hypothesis speculated that ethical leadership is positively related to responsibility. To test this hypothesis, a linear regression analysis was conducted. After this, it was found that control variables did not have a significant impact on the model’s ability to predict responsibility. Results, moreover, reveal a significant positive relationship between ethical leadership and responsibility (β= .222, SE= .096, p < 0.01), which can be observed in Table 4.
Table 4. Results linear regression analysis for ethical leadership and responsibility
Step 1 Step 2 Independent variable Ethical leadership .465** (.060) Control variables Age .064 (.005) .015 (.005) Gender .031 (.068) -.005 (.060)
Tenure with leader .011 (.001) .023 (.001)
Tenure with organization .015 (.024) .043 (.021)
R² .007 .221
Adjusted R² -.022 .193
F .230 7.872**
Note. Standardized coefficients are presented: **p < 0.001
Step 1 Step 2 Independent variable Ethical leadership .222** (.096) Control variables Age .046 (.008) .023 (.008) Gender .018 (.098) .001 (.096)
Tenure with leader .014 (.002) .020 (.002)
Tenure with organization -.024 (.034) -.010 (.034)
R² .002 .051
Adjusted R² -.027 .017
F .062 1.485
30
The third hypothesis assumed that ethical leadership is positively related to interdependent self-construal, and a linear regression analysis was used to test this. Step 1 illustrates that age is significantly positively related to interdependent self-construal (β= .320, SE= .005, p < .05). Furthermore, step 2 demonstrates that ethical leadership is significantly positively related to interdependent self-construal (β= .479, SE= .053, p < .001) and this is also true for age (β= .270, SE= .004, p < .05). Ethical leadership also explains a significant proportion of variance of 24% in interdependent self-construal scores (F=10.085, p < 0.001). To illustrate, Table 5 presents the results of the linear regression analysis.
Table 5. Results linear regression analysis for ethical leadership and interdependent self-construal
Similarly, a linear regression analysis was also carried out to test the fourth hypothesis, which hypothesised that interdependent self-construal is positively related to OCB. The hypothesis is confirmed as it was established that interdependent self-construal is significantly positively related to OCB (β= .553, SE= .080, p < .001). Interdependent self-construal also explains a significant proportion of variance of 27.5% (F = 11.927, p < .001). Results of this analysis are presented in Table 6. Step 1 Step 2 Independent variable Ethical leadership .479** (.053) Control variables Age .320* (.005) .270* (.004) Gender -.023 (.060) -.061 (0.053)
Tenure with leader -.092 (.001) -.080 (.001)
Tenure with organization -.248 (.021) -.219 (.018)
R² .040 .266
Adjusted R² .012 .240
F 1.440 10.085**
31
Table 6. Results linear regression analysis for interdependent self-construal and organizational citizenship behavior
Next, the fifth hypothesis predicted that interdependent self-construal is positively related to responsibility. After a linear regression analysis, it is revealed that there exists a significant positive correlation (β= .279, SE= .133, p =.001). Interdependent self-construal also explains a significant proportion of variance of 4.3% in responsibility scores. Table 7 provides an overview of the outcomes.
Table 7. Results linear regression analysis for interdependent self-construal and responsibility
Step 1 Step 2 Independent variable Interdependent self-construal .553** (.080) Control variables Age .064 (.005) -.113 (.005) Gender .031 (.057) .044 (.057)
Tenure with leader .011 (.001) .062 (.001)
Tenure with organization .015 (.024) .152 (.020)
R² .007 .300
Adjusted R² -0.022 .275
F .230 11.927**
Note. Standardized coefficients are presented: **p < 0.001
Step 1 Step 2 Independent variable Interdependent self-construal .279** (0.133) Control variables Age .046 (.008) -.043 (.008) Gender .018 (.098) .025 (.094)
Tenure with leader .014 (.002) .040 (.002)
Tenure with organization -.024 (.034) .046 (.133)
R² .002 .077
Adjusted R² -.027 .043
F .062 2.306*
32
According to the sixth hypothesis, follower’s interdependent self-construal mediates the positive relationship between ethical leadership and follower’s OCB. Hayes’s PROCESS tool was employed to perform a mediation analysis. Results indicate a significant indirect effect of followers’ interdependent self-construal on the relationship between ethical leadership and OCB, b = 0,158, BCa CI [0.072, 0.260]. This represents a relatively medium effect, κ2 = 0.197, 95% BCa CI [0.098, 0.307]. The results are illustrated in Figure 7.
b = 0.35, p <0.001 b = 0.45, p = 0.001
Direct effect, b = 0.21, p < 0.001
Indirect effect, b = 0.16, 95% CI [0.07, 0.26]
Figure 7. Interdependent self-construal as a mediator in the relationship between ethical leadership and organizational citizenship behavior
Subsequently, the seventh hypothesis predicted that follower’s interdependent self-construal mediates the positive relationship between ethical leadership and follower’s responsibility. Again, Hayes PROCESS tool was used to assess this hypothesis. The outcomes indicate that the direct effect of ethical leadership on responsibility was non-significant (b = 0.138, p = 0.196). However, results present a significant indirect effect of interdependent self-construal on the relationship between ethical leadership and responsibility, b = 0,120, BCa CI [0.035, 0.237]. This represents a relatively medium effect, κ2 = .140, 95% BCa CI [.030, .204]. For an overview of the results, see Figure 8.
Ethical leadership Interdependent self-construal Organizational citizenship behavior
33
b = 0.35, p <0.001 b = 0.34, p = 0.02
Direct effect, b = .138, p = .196
Indirect effect, b = 0.120, p =, 95% CI [.035, .237]
Moreover, the eighth hypothesis anticipated that followers’ interdependent self-construal moderates the relationship between ethical leadership and followers’ OCB, so that the relationship between ethical leadership and followers’ OCB is stronger when followers’ interdependent self-construal is higher. Results reveal a significant interaction effect of followers’ interdependent self-construal on the relationship between ethical leadership and OCB, b = 0,383, 95% CI [0.121, 0.646], t = 2.89, p < 0.01. Furthermore, when
interdependent self-construal is low, there is a non-significant positive effect between ethical leadership and OCB, b = 0,075, 95% CI [-0.083,0.231], t = 0.939, p = 0.350. Yet, at the mean value of interdependent self-construal, there is a significant positive effect between ethical leadership and OCB, b = 0,201, 95% CI [-0.065, 0.336], t = 2.926, p < 0.05. In contrast, when interdependent self-construal is high, there is a significant positive relationship between ethical leadership and OCB, b = 0.326, 95% CI [-0.163,0.490], t = 3.9412, p < 0.001. Figure 9 illustrates the effect of interdependent self-construal on the relationship between ethical leadership and OCB.
Ethical leadership
Interdependent self-construal
Responsibility
Figure 8. Interdependent self-construal as a mediator in the relationship between ethical leadership and responsibility
34
Figure 9. The moderation effects of interdependent self-construal on the relationship between ethical leadership and organizational citizenship behavior
The results of the moderation analysis are also presented in the table below.
Table 8. Results moderation analysis of interdependent self-construal on the relationship between ethical leadership and organizational citizenship behavior
According to the final hypothesis, followers’ interdependent self-construal moderates the relationship between ethical leadership and followers’ responsibility, so that the relationship between ethical leadership and followers’ responsibility is stronger when followers’
interdependent self-construal is higher. Nonetheless, the outcomes reveal a non-significant interaction effect of followers’ interdependent self-construal on the relationship between ethical leadership and responsibility, b = 0,340, BCa CI [-0.370, 1.050], t = 0.948, p = 0.345.
3 3.5 4 4.5 5 5.5
Low Ethical leadership High Ethical leadership
Organiz ational Citiz enship Behavior Low Interdependent self-construal High Interdependent self-construal b SE B t p Constant 3.933 0.0254 155.931 p < 0.001 [3.913, 4.014]
Interdependent self-construal (centred) 0.421 0.093 4.511 p < 0.001
[0.237, 0.606]
Ethical leadership (centred) 0.201 0.069 2.926 p = 0.004
[0.065, 0.336]
Interdependent self-construal x Ethical leadership 0.384 2.891 p = 0.005
35
In addition, when interdependent self-construal is low, there is a non-significant relationship between ethical leadership and responsibility, b = 0.014, BCa CI [-0.403, 0.431], t = 0.065, p
= 0.948. Furthermore, at the mean value of interdependent self-construal, there is a
non-significant relationship between ethical leadership and responsibility, b = 0,125, BCa CI [-0.130, 0.380], t = 0.972, p = 0.333. When interdependent self-construal is high, there also exists a non-significant relationship between ethical leadership and responsibility, b = 0,237, BCa CI [-0.017, 0.491], t = 1.845, p = 0.067. To illustrate, Table 9 presents the results of the moderation analysis.
Table 9. Results moderation analysis of interdependent self-construal on the relationship between ethical leadership and responsibility
36
5 Discussion and conclusion
5.1 Research findings
This research explores the following research question: ‘To what extent does follower’s interdependent self-construal mediate and moderate the relationship between ethical
leadership and follower’s citizenship behavior and responsibility?’ Two research models were employed to test the mediating and moderating effect of follower’s interdependent
self-construal on the relationships between ethical leadership and OCB and responsibility. The
hypotheses are discussed below.
The first hypothesis, ethical leadership is positively related to OCB, is confirmed. This finding is in line with earlier studies that suggested a positive relationship between ethical leadership and OCB (Brown, Treviño, 2006; Mayer, Kuenzi, Greenbaum, Bardes, &
Salvador, 2009; Piccolo, Greenbaum, Den Hartog, & Folger, 2010; Kalshoven, Den Hartog & De Hoogh, 2011; Walumba & Schaubroek, 2009). Additionally, these results provide support for the social learning theory (Bandura, 1986) and the social exchange theory (Brown & Treviño, 2006). An explanation for this is that ethical leaders care about their followers and empower them, which motivate followers to exhibit more OCB. As a result, followers learn from role models, such as ethical leaders, and they are willing to reciprocate with OCB when they are treated fairly and with concern.
The second hypothesis, which expected that ethical leadership is positively related to follower’s responsibility, is confirmed. Social learning theory and social exchange, moreover, can be used to explain this relationship. Similar to findings of Kalshoven et al. (2011) and Masterson and Stamper (2003), followers tend to copy responsible behavior from an ethical leader and they are willing to reciprocate this conduct by demonstrating responsibility.
The third hypothesis, ethical leadership is positively related to follower’s
interdependent self-construal, is confirmed. It was expected that the people orientation of ethical leadership will positively affect follower’s interdependent self-construal. Support for this was found by Lord et al (1991), as they observed that leaders activate a self-construal among followers which corresponds to the displayed style of leadership. This finding suggests that leaders are able to change the way in which followers perceive themselves. Hence, by activating a certain level of self-construal, ethical leaders can shape followers’ perceptions and behavior, which in turn will lead to positive work outcomes.
37
positively related to follower OCB and this is confirmed. In agreement with research findings of Moorman and Blakely (1995), this study noticed that individuals high on interdependent self-construal are likely to display OCB. These individuals have a focus on duties and
obligations and therefore tend to show OCBs, such as helping other members and suggesting organizational improvements.
Next, the fifth hypothesis, follower’s interdependent self-construal is positively related to responsibility, is confirmed. Thus, individuals with an interdependent self-construal are motivated to fulfil their duties and responsibilities to the organization and therefore are inclined to display responsible behavior. As these individuals have a focus on
interconnectedness with others, they are likely to behave with dependability. Also, Individuals with an interdependent self-construal value relationships with others, similar to their leader. Hence, followers with an interdependent self-construal are willing to meet the leader’s
expectations by fulfilling their responsibilities. This is consistent with prior research by Cross and Madson (1997) who found that self-construals can partially determine responsibility.
The sixth hypothesis anticipated that follower’s interdependent self-construal mediates the positive relationship between ethical leadership and follower OCB. The results from the analysis indicate significant support for an indirect positive effect of follower’s
interdependent self-construal on the relationship between ethical leadership and follower OCB. In other words, leaders who are ethical, make followers’ interdependent self-construal more salient, which, in turn, translates into higher levels of followers. According to Van Knippenberg et al. (2004), leaders can enhance a corresponding level of self-construal in the follower by using a certain style of leadership, which will also influence the follower
behavior. Thus, this finding implies that followers of ethical leaders who show behavior that are related to fairness, integrity, ethical guidance, people orientation, power sharing, role clarification, and concern for sustainability (Kalshoven, Den Hartog, & De Hoogh, 2011), identify themselves with these leaders and therefore are likely to exhibit positive work
behaviors and attitudes.
The seventh hypothesis, follower’s interdependent self-construal mediates the positive relationship between ethical leadership and responsibility, is partly confirmed. The results reveal a non-significant direct effect of ethical leadership on responsibility; however, the indirect effect of interdependent self-construal on the relationship between ethical leadership and responsibility is significant. By clarifying responsibilities, expectations, and performance goals, ethical leaders appeal to individuals with an interdependent self-construal.
38
For this reason, follower’s interdependent self-construal translates into higher levels of responsibility. This result indicates that follower’s level of self-construal plays an important role in the relation between ethical leadership and follower attitudes.
The eight hypothesis predicted that follower’s interdependent self-construal moderates the positive relationship between ethical leadership and OCB, so that the relationship between ethical leadership and follower OCB is stronger when followers’ interdependent self-construal is higher. This study demonstrates that follower’s interdependent self-construal moderates the relationship of ethical leadership and OCB so that it was positive when follower’s
interdependent self-construal was high, but not significant when follower’s interdependent self-construal was low; therefore, hypothesis eight is confirmed. This result is, moreover, in line with expectations based on earlier research outcomes that suggest that self-construals can act as moderators in the relationships between leadership and positive follower behaviour (De Cremer & Van Vugt, 2002; Hains, Hogg, & Duck, 1997; Lord & Brown, 2004; Platow & Van Knippenberg, 2001). Furthermore, the outcome suggests that self-concepts of followers, their self-construals specifically, may significantly influence how or if they are affected by ethical leaders. In addition, this results infers that followers’ individual differences may influence the effectiveness of ethical leadership. Building on this, Van Knippenberg et al. (2004) predicted that leadership is more effective if it is consistent with the level of self-construals that it targets. Therefore, in order to be effective, leaders need to tailor their leadership to their follower’s self-construal; nonetheless, research is required to better understand the
effectiveness of such measures.
According to the ninth hypothesis, follower’s interdependent self-construal moderates the positive relationship between ethical leadership and responsibility, so that the relationship between ethical leadership and followers’ responsibility is stronger when followers’
interdependent self-construal is higher. This hypothesis is rejected. In contrast to the expectations, results did not demonstrate a moderating effect of follower’s interdependent self-construal on the relationship between ethical leadership and responsibility. An
explanation for this can be found in the use of the items that measured responsibility. These items were adapted from the Job Diagnostic Survey (Hackman & Oldham, 1980) and were based on the Job Characteristics Theory, which implies that responsibility is a critical
psychological state reflecting the extent to which ‘the individual feels personally accountable and responsible for the results of the work he or she does’ (Fuller, Marler, & Hester, 2006).
