Senior team shared leadership and organizational ambidexterity : the influence of vertical transformational leadership and task interdependence

(1)

Senior team shared leadership

and organizational ambidexterity

The influence of vertical transformational

leadership and task interdependence

Solka van Meer 10651551 24th of June, 2016 Final draft MSc. in Business Administration – Strategy ABS, University of Amsterdam Supervisor: Michiel Tempelaar

(2)

Statement of originality

This document is written by Solka van Meer, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

List of Tables

Table 1. Means, Standard deviations, Correlations and Reliabilities Table 2. Results of hierarchical regression analyses: Transformational

leadership, ST Shared Leadership, Task Interdependence, and Ambidexterity

List of figures

Figure 1. Conceptual model

Figure 2. The outcomes of the process analysis visualized in the conceptual model

(6)

Acknowledgement

This thesis has been written as a part of my Master of Science in Business Administration – Strategy. In my opinion, the topic of this thesis is interesting and relevant within the field of strategy, as well from a practical and theoretical point of view. With this note, I would like to reflect to the people who have supported and helped me throughout this period.

First of all, I would like to thank my supervisor Michiel Tempelaar for his valuable guidance. His insightful comments and constructive criticisms at different stages of my research were thought-provoking and they helped me focus my ideas. Besides, I would also like to thank my second reader in advance for taking the time to read and evaluate this research. Furthermore, I am grateful to my family and in particular to my parents for their unconditional support and their help to overcome setbacks and to stay focused on my study. Finally there are my friends, who supported me not only by discussing my problems and findings, but also by talking about other things than my paper.

(7)

Abstract

Organizational ambidexterity, defined as simultaneously pursuing exploration and exploitation, has gained increasing interest in the recent years. Prior studies emphasized the role of senior teams in managing the strategic contractions associated with explorative and exploitative activities. The current study elaborates on that perspective by examining the role of senior team shared leadership (ST) in achieving organizational ambidexterity. In addition, we consider under what conditions ST shared leadership enhances the development of organizational ambidexterity. Namely, vertical transformational leadership is assumed to be an important antecedent of senior team shared leadership and, in addition, we discuss how task interdependence moderates the relations between transformational leadership and ST shared leadership, and between ST shared leadership and organizational ambidexterity.

The moderated mediation framework is tested using cross-sectional data from a sample of 186 firms. Overall, the findings indicate that the effect of transformational leadership on organizational ambidexterity indeed operates through ST shared leadership. Interestingly, the results revealed no significant relationship for the previously asserted influence of task interdependence on the relationship between transformational leadership and ST shared leadership, and between ST shared leadership and organizational ambidexterity. However, this provides us with important avenues for future research. In conclusion, through a richer explanation and empirical assessment, we provide a more nuanced understanding of how organizations may effectively pursue both exploration and exploitation and we contribute to the debate on the role of leadership in managing the paradox.

Key words: Organizational ambidexterity; exploration and exploitation; senior teams; shared leadership; transformational leadership; task interdependence

(8)

1. Introduction

Over the past decade, one of the emerging themes in organizational literature is that successful organizations need to be ambidextrous, by focusing on simultaneously pursuing exploratory and exploitative innovation (Benner & Tushman, 2003; He & Wong, 2004; Birkinshaw & Gibson, 2004). March (1991) articulated contradictory organizational and strategic demands on firms engaging in both exploration and exploitation. Where exploration is rooted in variance-increasing, experimental activities and learning by doing, exploitation is rooted in variance-decreasing activities and controlled problem solving (Smith & Tushman, 2005). Exploitation builds on a firm’s past and exploration creates a future that may be different than a firm’s past. However, organizations must outcompete their rivals both in the short run and in the long run to achieve sustained performance. In order to do so, they must deal simultaneously with the inconsistent demands of exploration and exploitation (Benner & Tushman, 2003). Because exploratory and exploitative activities require fundamentally different and inconsistent structures and demands, attaining organizational ambidexterity appears to be complex and difficult to achieve (Li, 2013). Therefore, an important issue in organizational literature is, understanding what drives organizations to be ambidextrous.

Prior studies have recognized the complementary roles of differentiation and integration as mechanisms for enabling organizations to deliver ambidexterity (Lawrence & Lorsch, 1967). Differentiation refers to the separation of explorative and exploitative activities into different organizational units, and integration addresses the mechanisms that enable organizations to pursue exploratory and exploitative activities within the same organizational unit (Raisch, Birkinshaw, Probst & Tushman, 2009). Gilbert (2005) pointed out that the mere coexistence of explorative and exploitative

(9)

activities in differentiated organizational units is an important, but insufficient condition for organizational ambidexterity. Moreover, Tushman and O’Reilly (1996) argued that senior teams need to integrate the distinct activities across differentiated units in order to achieve organizational ambidexterity. The need to combine both processes of differentiation and integration is difficult to resolve and a lot more complex than a simple either/or trade-off (Eisenhardt, 2000).

Research that views organizational ambidexterity as a desired outcome has highlighted the organizational structures (Benner & Tushman, 2003; Tushman & O’Reilly, 1996), the behavioural and organizational contexts (Birkinshaw & Gibson, 2004; Gibson & Birkinshaw, 2004) and the integrated role of senior teams as an important factor for managing the tension between differentiation and integration in order to achieve organizational ambidexterity (Smith & Tushman, 2005). Birkinshaw and Gibson (2004) argued that ambidexterity cannot just be attained through a formal structure or through a vision statement of a charismatic leader. Rather, ambidexterity is to most extent achieved through the creation of a supportive context, in which individuals can make their own decisions about how and where to focus their energies. Recently, scholars have started to investigate the leadership characteristics that help organizations to manage the contradictory demands they face (Raisch & Birkenshaw, 2008; Smith & Tushman, 2005; Jansen, Vera & Crossan, 2009; Nemanich & Vera, 2009; Keller & Weibler, 2015). However, the specific means through which senior leaders manage the differentiation-integration challenge is still under-developed and unclear.

Since the senior team is considered as an important locus for resolving the conflicting tensions, the leadership style of senior team members is of particular importance in understanding the ability of organizations to pursue both exploratory

(10)

and exploitative activities (Jansen, George, van den Bosch & Volberda, 2008). Research has merely adopted a top-down view on leadership, whereby the CEO is responsible for overcoming the tensions between exploration and exploitation and for influencing both senior team members and organizational members (Mihalache, Jansen, van den Bosch & Volberda, 2014). However, scholars have identified a new approach on the role of leadership and proposed that leadership responsibilities may be shared among senior team members. This perspective holds a great potential to better understand a firm’s ability to achieve organizational ambidexterity, because in most cases the CEO’s leadership style is not sufficiently effective to overcome the differentiation-integration tensions and the distinct interests of members in a multifunctional senior team (Beer et al., 2005). Senior team (ST) shared leadership is expected to positively influence organizational ambidexterity by encouraging members to develop a shared mission and a supporting climate and by enhancing cooperative decision making and conflict management (Mihalache et al., 2014). Therefore, to gain a better understanding of the underlying processes of attaining organizational ambidexterity, we focus on the role that ST shared leadership plays in achieving organizational ambidexterity.

Thus, as ST shared leadership is expected as a key factor for achieving organizational ambidexterity, the question arises how its development can best be facilitated (Hoch, 2013). Pearce, Manz and Sims (2009) emphasized that ST shared leadership is not a replacement for leadership from above. Organizations can simultaneously engage in shared leadership together with other leadership approaches, such as vertical leadership. Moreover, Pearce (2004) stated that ST shared leadership is probably going to fail without the support from a vertical leader. In addition, Smith and Tushman (2005) proposed that a preferred leadership style is

(11)

an important determinant of the effectiveness of teamcentric or leadercentric teams, whereby the authors suggest that teamcentric teams are more associated with a democratic leadership style. As the characteristics of a team-centric team are similar to those of ST shared leadership, the focus of this research is concentrated on the effect of a democratic leadership style on ST shared leadership. Prior studies have demonstrated that vertical transformational leadership positively influences organizational ambidexterity (Jansen et al., 2008; Keller & Weibler, 2015; Nemanich & Vera, 2009). In addition, transformational leadership is considered to predict shared leadership as it may lead to the development of shared vision, as well as the creation of collective goals (Bass, Avolio, Jung & Berson, 2003). Moreover, transformational leadership has been shown to promote self-management skills (Keller & Weibler, 2015), which is also a likely predictor of ST shared leadership. For this reason, the second objective of this study is to investigate how vertical transformational leadership relates to ST shared leadership (see figure 1).

Fig. 1. Conceptual model.

In the past few years, research has already investigated the relationship between transformational leadership and ST shared leadership. However, scholars have called for a better understanding of this relationship and for the investigation of

(12)

the important contextual variables that influence this relationship (Hoch, 2013). According to Smith and Tushman (2005) task interdependence is an important condition under which teamcentric teams versus leadercentric teams dominate in managing strategic contradictions. They propose that teamcentric teams may better able to deal with interdependence than leadercentric teams. On the contrary, leadercentric teams may have better capabilities to deal with more limited coordination requirements. As mentioned before, the characteristics of teamcentric teams are comparable with those of ST shared leadership, which leads us to suggest that task interdependence may play an important role on the relation between transformational leadership and ST shared leadership and on the relation between ST shared leadership and organizational ambidexterity.

Altogether, this leads to the following research question:

How does senior team shared leadership relate to organizational ambidexterity and what is the role of vertical transformational leadership and task interdependence?

In conclusion, this study makes several contributions to the current leadership and organizational literature. First of all, it is a response to the call for more research on different leadership behaviours affecting the establishment of organizational ambidexterity. Largely overlooked so far, ST shared leadership holds a great potential to gain better understanding of the ability of firms to attain organizational ambidexterity (Mihalache et al, 2014). Second, given that leadership behaviours play an important role in stimulating organizational ambidexterity (Jansen et al., 2008; Lin & McDonough, 2011; Smith & Tushman, 2005; Smith, Binns & Tushman, 2010; Tushman & O’Reilly, 1996), it is important to examine various antecedents and/or moderators directly influencing the particular leadership style. Although multiple

(13)

theoretical frameworks and scholars within the shared leadership literature stress the importance of exploring moderators and antecedents of shared leadership (Bligh, Pearce & Kohles, 2006; Carson, Tesluk & Marrone, 2007; Hoch, Pearce & Welzel, 2010), only a few studies have done so. The resulting moderated mediation framework contributes to prior research by considering under what conditions ST shared leadership holds the potential to stimulate an organization to be ambidextrous. So this study contributes to an overall understanding about the importance of ST shared leadership in the realization of organizational ambidexterity. In addition, the current study contributes to the leadership literature by providing a more nuanced view of the effects of transformational leadership. By bringing together different leadership perspectives and ambidexterity, we elaborate on Smith and Tushman’s (2005) notion that embracing rather than deciding between contradictory leadership styles may be an important direction for future research.

The next section will review the existing literature on organizational ambidexterity, focussing on the role that leadership plays in establishing ambidexterity, followed by a review of the literature on shared leadership and briefly the literature on vertical transformational leadership and task interdependence. The section that follows describes the methods and subsequently the empirical findings are presented. The study concludes with a discussion of the results, the theoretical and practical implications and suggestions for further research.

(14)

2. Literature review and hypotheses

This theoretical chapter provides an overview of the existing leadership and ambidexterity literature. We will begin with exploring the concept of organizational ambidexterity, followed by a discussion of previous research on the topics of shared leadership, vertical transformational leadership and finally task interdependence. Based on prior literature, the proposed relationships between those concepts will be explained. So the influence of ST shared leadership on organizational ambidexterity will be reviewed. Afterwards, the relation between vertical transformational leadership and ST shared leadership will be discussed. Finally, the moderating effect of task interdependence on the proposed relations will be addressed. In sum, this chapter will provide a framework for the hypothesized relationships that will be tested in this study.

Organizational ambidexterity

A persistent theme in organizational and managerial literature is the concept of ambidexterity. To be ambidextrous, firms need to simultaneously engage in contradictory knowledge management processes: exploiting current competencies and exploring new product/market domains (Andriopoulos & Lewis, 2009; March, 1991; Tushman & O’Reilly, 1996). Exploration and exploitation entail distinct skills, structures, processes and strategies and they may have a different impact on firm performance (He & Wong, 2004). On the one hand, exploration is associated with refining existing knowledge, efficiency, implementation and short-term benefits. On the other hand, exploration is associated with developing new knowledge, experimentation, flexibility, risk taking and long-term survival (March, 1991; He & Wong, 2004). March (1991) noted that these two activities compete for the same scarce resources, which has often resulted in favouring one activity at the expense of

(15)

the other. The challenge that firms and researchers face is to discover how to allocate firm’s resources and capabilities in a way that organizations are able to successfully pursue both types of activities simultaneously (Lin, McDonough, Lin & Lin, 2013).

Since Lawrence and Lorsch’s (1967) article, several studies have recognized differentiation and integration as mechanisms for enabling organizations to deliver organizational ambidexterity. Differentiation addresses mechanisms for dividing distinct tasks into separate organizational units that tend to develop appropriate contexts for both exploration and exploitation. This differentiation may help organizations to maintain different competencies with which to approach the contradictory demands associated with explorative and exploitative activities (Raisch et al., 2009). Integration addresses exploration and exploitation within the same organizational unit. According to Eisenhardt and Martin (2000), exploration and exploitation have to be combined in order to create value. On top of that, Gilbert (2005) states that approaching exploration and exploitation in separate organizational units is an important, but insufficient condition for organizational ambidexterity. In line with that, scholars pointed to the need for senior teams to assure integration across differentiated business units (Lubatkin, Simsek, Ling & Veiga 2006; Tushman & O’Reilly, 1996; Smith & Tushman, 2005). So research started to suggest that senior teams need to balance these paradoxical demands and overcome the differentiation-integration tensions in order to achieve long-term organizational effectiveness (Raisch, 2008; Raisch et al., 2009). Therefore, senior managers need to be able to host inconsistencies in order to maintain an appropriate balance between exploration and exploitation. Organizations can achieve a high performance when the senior team successfully balances strategic contradictions and uses integration mechanisms to stimulate knowledge flows across distinct business units.

(16)

However, structural, psychological and social psychological barriers prevent senior managers from making these balanced decisions (Levinthal & March, 1993; Kahneman & Teversky, 1979; Smith & Tushman, 2005; Watkins & Bazerman, 2003). For instance, Levinthal and March (1993) argued that senior executives are myopic: they tend to sacrifice the long run to the short run (temporal myopia), they tend to favour effects that occur near to the them (spatial myopia) and they oversample successes and undersample failures (failure myopia). In most cases, these three types of learning myopia lead to difficulties in sustaining adequate exploration. Furthermore, Kahneman and Tversky (1979) found that managers are risk averse and as such prefer to invest in less risky (existing) products instead of in more risky (explorative) innovations. In addition, individuals tend to preserve consistency: the feeling that there must be a contingency that mediates between inconsistent ideas (Leana & Barry, 2000; Smith & Tushman, 2005). However, engaging in both exploration and exploitation result in conflicts, and inconsistencies between the two activities cannot be eliminated. Moreover, Watkins and Bazerman (2003) observed that humans inhibit psychological-cognitive defects that leave them blind to approaching threats. They tend to see the world as they would like it to be, rather than as it truly is. This self-serving bias can be especially destructive when the interests are in conflict.

So achieving ambidexterity may create conflicts among senior team members as the short-term, efficiency focus is at odds with the long-term, experimental focus (Jansen et al., 2008). Balancing strategic decisions require senior teams to recognize and embrace those conflicts instead of trying to resolve them, and senior teams must enable the coexistence of inconsistent agendas (O’Reilly & Tushman, 2008; Smith & Tushman, 2005). To make balanced decisions senior teams must be able to confront

(17)

and overcome those structural, social psychological and psychological barriers in order to support innovation and achieve ambidexterity. The crucial role that senior teams play in mediating between forces for exploration and exploitation, has often been highlighted (Jansen et al., 2008; Lin & McDonough, 2011; Smith & Tushman, 2005; Smith et al., 2010; Tushman & O’Reilly, 1996), but the specific means through which leaders influence organizational ambidexterity and how they overcome the tensions between exploration and exploitation is rather limited and under-developed (Jansen et al., 2009). Research on the role of senior team characteristics, such as paradoxical cognition (Smith & Tushman, 2005) and leadership (Keller & Weibler, 2015; Jansen et al., 2009; Nemanich & Vera, 2009), on firm-level ambidexterity has only begun to emerge.

Smith and Tushman (2005) stated that the “design” and composition” of a senior team is crucial for reconciling the paradoxical challenges and for attaining organizational ambidexterity. In addition, the research of Birkinshaw and Gibson (2004) suggested that a supportive social context, characterized by a combination of performance management and social support, a culture and interpersonal relations are associated with higher levels of ambidexterity. Moreover, Lubatkin et al. (2006) proposed that the ability to jointly pursue both exploration and exploitation is directly rooted in the extent to which senior teams are behaviourally integrated. This view coincides with research on leadership characteristics, suggesting that the distribution of leadership among several members of a team encourages members to develop a shared vision and a supportive climate (Solansky, 2008). Thereby, Weick and Robert (1993) proposed that shared leadership provides more frequent team interactions and subsequently more opportunities for knowledge sharing, which promotes senior team interpersonal relations and behavioural integration. Moreover, prior studies have

(18)

indicated that shared leadership enhances team and organizational performance and team effectiveness (Ensley, Hmieleski & Pearce, 2006; Pearce, 2004; Pearce & Manz, 2005). Therefore, the purpose of this study is examining how ST shared leadership influences the ability of a senior team to achieve organizational ambidexterity, whereby the senior team is considered as the CEO and senior executives who are responsible for important decisions about the organization (Carmeli & Halevi, 2009). Senior team shared leadership and ambidexterity

In recent years, it is has been ever more difficult for any employee to have all of the knowledge and skills required for the job on hand. This is especially the case for knowledge work, which can be described as work that requires intellectual capital and skilled professionals (Pearce, 2004). For example, in a team of specialists, one person generally does not have all the skills and knowledge to understand all the aspects of the job. This has resulted in a shift to team-based work and new challenges for management teams (Pearce, 2004).

Traditional leadership has been conceived around a single individual (the leader). However, researchers have challenged this conception, stating that leadership involves roles that can and should be shared among members of a team (Pearce et al., 2009). Firms need to develop leaders at all levels, when it comes to knowledge work, and one approach to management has been ST shared leadership (Hoch, 2013). ST shared leadership can be described as internal, informal leadership, where the objective is to lead one another towards the achievement of collective goals, and requires the distribution of leadership across different individuals (Hoch, 2013; Pearce & Manz, 2005; Pearce et al., 2009). ST Shared leadership is characterized by cooperative decision-making and shared responsibility, collective efficacy and

(19)

collaboration. So the leadership is carried out by the team as a whole, rather than by single individual (Ensley et al., 2006).

It is expected that ST shared leadership will play a role in facilitating organizational ambidexterity in at least two different ways. First, ST shared leadership provides more frequent team interactions, which allows for more opportunities of information sharing (Weick & Robert, 1993). Sharing information is valuable for attaining ambidexterity for at least two reasons. First, as team members will share knowledge and ideas, the team as a whole will possess more ideas and they will achieve higher levels of creativity (Hoch, 2013). Second, as team members provide more information, other team members might also build upon those ideas, which will also lead to higher levels of creativity (Hoch, 2013). Eisenhardt (2000, p. 703) argued that managing the paradox requires “a creative way that captures both extremes” instead of a simple either/or decision. In addition, Andriopoulous and Lewis (2009) stated that organizations need managerial creativity in order to achieve organizational ambidexterity. Moreover, Lewis (2000) noted that organizing paradoxes, such as the differentiation-integration challenge, requires an ongoing process of balancing the contradictory demands that encourage commitment, trust and creativity. Thus, as managers’ creativity is an important condition for managing the paradox, this leads us to suggest that ST shared leadership, by enhancing creativity, encourages the ability of senior executives to attain organizational ambidexterity.

Second, ST shared leadership might also be helpful in developing a shared vision, shared mission and a supporting climate (Solansky, 2008). When a team is working under a shared vision and collective mission, they are more likely to work towards a common goal (Hoch, 2013). Lewis (2000) argued that overarching goals

(20)

are important for balancing opposing forces, because common goals and values provide a strategic direction, which reduces conflicts of interest and disagreements. In addition, ST shared leadership encourages senior team members to develop a shared vision, and as a result of that widely acknowledge vision, team members will be more willing to accept conflicting perspectives about strategic and tactical issues (Jansen et al., 2008). In line with that, Lewis (2000) stated that, amongst other things, acceptance – learning to live with the paradox – is an important means of managing the paradox, because it helps senior team members to avoid debates, which in turn may contribute to development of new ideas. So enhancing senior team members to create a shared vision will contribute to higher levels of acceptance, which is beneficial for managing the paradox.

In sum, by stimulating creativity, by encouraging members to develop a shared vision and superordinate goals, ST shared leadership may contribute to the development of organizational ambidexterity. For this reason, the subsequent hypothesis can be formulated:

Hypothesis 1. Senior team shared leadership is positively related to organizational ambidexterity.

Vertical transformational leadership and shared leadership

While ST shared leadership might enhance organizational ambidexterity, the question remains how it can best be developed (Hoch, 2013; Pearce, 2004). Moreover, only little research has empirically addressed the antecedents of ST shared leadership (Carson et al., 2007; Hoch, 2013). Pearce (2004) stated that shared leadership can reduce the need for on-going vertical leadership, but without support and maintenance

(21)

from a vertical leader it is likely going to fail. Therefore, vertical leadership behaviour is expected to affect the development of ST shared leadership.

Vertical leadership can be defined as the leadership by an external or internal designated leader. Vertical leadership takes place through a top-down process and is dependent upon the knowledge and skills of an individual person (Ensley et al., 2006). The type of leadership style considered in this research is a set of behaviours known as transformational leadership. The four components of transformational leadership include: 1) idealized influence, 2) inspirational motivation, 3) intellectual stimulation, and 4) individualized consideration (Bass et al., 2003).

Idealized influence means that leaders are admired, respected and trusted by their followers. The leader considers follower’s needs over his or her own needs, he shares risks with his subordinates and he is consistent in conduct with acknowledged values and principles (Bass et al., 2003). This results in an increase of involvement, commitment and cohesiveness among employees and it enhances team processes and collaboration (Hoch, 2013). This corresponds with one of the core values of shared leadership, mentioned by Morgeson, DeRue and Karam (2009): effective collaboration and coordination among team members.

Another characteristic of transformational leadership is inspirational motivation, which can be defined as a leader motivating his or her employees by providing meaning and challenge to their work (Bass et al., 2003). The transformational leader increases the shared vision among team members (Pillai & Williams, 2004), and by doing so self-centered goals are replaced with the focus on team goals (Vera & Crossan, 2004). As focus on team goals is an important antecedent of shared leadership, and as transformational leadership has been shown to

(22)

enhance a shared vision and team goals, it is likely that this leadership style will be related to the development of ST shared leadership (Hoch, 2013).

The third component of transformational leadership, intellectual stimulation, can be described as a leader stimulating his or her employees to be innovative and creative. Moreover, the transformational leader articulates that there is no ridicule or criticism of employees’ mistakes and new solutions and ideas are highly appreciated (Bass et al., 2003). In addition, team members are included in the process of addressing problems and finding solutions, which fosters the development of the follower’s efficacy and leadership skills (Manz, 1986). In turn, self-leadership has been argued to develop shared self-leadership skills (Pearce & Manz, 2005). Therefore, this third component of transformational leadership also contributes to the development of ST shared leadership.

Finally, individualized consideration means that leaders pay attention to each individual’s need for achievement and group learning. In addition, different needs and desires of individual group members are recognized and the transformational leader develops new learning opportunities and creates a supporting climate in which members can grow (Bass et al., 2003). Employees are encouraged to believe in themselves and in their (joint) mission, which is beneficial for the collective efficacy and team potency. Since collective efficacy and team potency are also important components of shared leadership, it is likely that transformational leadership will foster shared leadership.

In total, we expect that transformational leadership will be positively associated with ST shared leadership. Therefore, we propose following:

Hypothesis 2. Vertical transformational leadership is positively related to senior team shared leadership.

(23)

The mediating role of shared leadership

Previous research has shown a positive, direct relationship between vertical transformational leadership and organizational ambidexterity at individual and organizational level (Jansen et al., 2009; Keller, 2006; Keller & Weibler, 2015; Nemanich & Vera, 2009). For example, Keller (2006) stated that transformational leadership behaviours foster innovative performance of teams, in R&D organizations, in both exploratory and exploitative projects. Moreover, Keller and Weibler (2015) observed a positive influence of transformational leadership on manager’s ambidexterity at individual level. These findings offer a starting point to gain insights into the mechanisms that stimulate ambidexterity. Both Morgeson and Hofmann (1999) and Keller and Weibler (2015) ask for a better understanding of the relationship between transformational leadership and ambidexterity at organizational level.

Several studies have demonstrated that transformational leadership increases team performance through processes of increased goal commitment and self-efficacy (Bass et al., 2003), or via a shared vision and intellectual stimulation (Pearce, 2004). Thereby, Hunter and Cushenbery (2011) noticed that innovation and creativity could be facilitated and encouraged indirectly by a leader, such as by developing a supportive climate or by rewards and recognition. In addition, vertical leadership behaviours have been shown to impact collective team outcomes via team potency and a collective efficacy (Bass et al., 2003) and to increase organizational ambidexterity through a learning culture (Nemanich & Vera, 2009). Similarly, this indirect impact might also be true for the relationship between vertical transformational leadership and organizational ambidexterity, through the processes of ST shared leadership.

(24)

According to Smith and Tushman (2005), some senior teams integrate the strategic contradictions between exploration and exploitation at the leader’s level (leadercentric teams) and in other senior teams the team members share the responsibility for integrating the strategic contradictions (teamcentric teams). The authors argue that a leadership style may be an important condition under which leadercentric versus teamcentric teams dominate in managing the strategic contractions. The leadership style associated with leadercentric teams is much more autocratic, whereas the leadership style associated with teamcentric teams is more democratic, whereby the latter type of leader encourages his/her team members to attend to both their own products as well as organization-wide issues (Smith & Tushamn, 2005). In sum, this study assumes that a preferred leadership style has an impact on organizational ambidexterity, mediated by the locus of integration (teamcentric or leadercentric), whereby a more democratic leader has an impact on ambidexterity via the locus of integration at teamcentric level. Since a teamcentric team is characterised by team-members working together to make balanced decisions and by integrating the contrasting activities themselves, this team design is comparable with ST shared leadership. Thereby, the democratic leadership style described by Smith and Tushman (2005) corresponds with the transformational leadership style addressed in this research. Accordingly, we assume that transformational leadership has a positive influence on organizational ambidexterity, mediated through processes of ST shared leadership.

In conclusion, due to the impact of transformational leadership on ST shared leadership, we expect the relationship between transformational leadership and organizational ambidexterity to be indirect. Accordingly, the following hypothesis can be formulated:

(25)

Hypothesis 3. The positive relationship between transformational leadership and organizational ambidexterity is mediated by senior team shared leadership.

Interdependent tasks as a moderator

In the past years, the first steps have been set in exploring the impact of vertical transformational leadership on ST shared leadership. However, scholars have called for a better understanding of the contextual variables that may strengthen or weaken this relationship and recommended that future researchers should investigate potential moderators (Hoch, 2013; Pearce, 2004). Thereby, little research has investigated potentials moderators on the relationship between ST shared leadership and organizational ambidexterity. Therefore, another purpose of this paper is to investigate contextual moderators of the proposed relationships.

Prior research has investigated that task interdependence which can be defined as the degree to which goal accomplishment requires the interaction of group members (Liden, Wayne & Bradway, 1997; Bligh et al., 2006), is of particular importance to group performance (Manz & Sims, 1987). For example, Latané, Williams and Harkins (1979) found out that teams outperform individuals on tasks that involve higher levels of interconnectedness and integration. In addition, Pearce et al. (2009) emphasize that shared leadership should be considered for situations in which the tasks of employees have a certain level of interconnectedness and interdependence. Furthermore, Smith and Tushman (2005) argued that task interdependence is an important condition under which teamcentric teams versus leadercentric teams dominate in managing strategic contradictions. Collaboration and cooperative decision-making between senior team members may be necessary for exploratory and exploitative activities that are highly interdependent, because such

(26)

tasks may require increased interaction between senior team members, due to the uncertainty associated with such interdependent tasks (Smith & Tushman, 2005). Accordingly, Smith and Tushman (2005) suggested that teamcentric teams might be better able to deal with such independence. Since the characteristics of teamcentric teams are comparable with the characteristics of ST shared leadership, this may implicate that the influence of transformational leadership on ST shared leadership may depend on the degree of task interdependence, and that the influence of ST shared leadership on organizational ambidexterity may also depend on the degree of task interdependence.

Task interdependence, transformational leadership and shared leadership

As mentioned before, Bass (1985) argued that transformational leaders develop follower’s capabilities via individualized consideration by providing opportunities for learning, by acting as a coach or mentor or by delegating responsibility and authority. This enhances the collective efficacy and the team potency, which in turn will foster ST shared leadership. According to Lee, Lin, Huan and Huang (2015) and Kiggundu (1983) task interdependence could also arouse a sense of responsibility among team members, resulting in advanced self-leadership skills and cooperation. So both transformational leadership and task interdependence stimulate a sense of responsibility, which in turn fosters ST shared leadership.

Another dimension of transformational leadership, idealized influence, enhances commitment and cohesiveness among group members by highlighting shared characteristics and by stressing the importance of a collective identity (Wayne & Howell, 2010). The article of van der Vegt, Emans and van de Vliert (2000) has found that task interdependence is also positively related to job commitment and team commitment. Namely, considerable planning and coordination is required for tasks

(27)

that are highly interdependent, which makes them challenging and may therefore increase job commitment. Task interdependence might also result in more frequent and extended interactions with co-workers, which fosters friendship and feelings of attachment to the group and can thereby enhance team commitment. As mentioned before, team commitment and cohesiveness are also core values of shared leadership (Morgeson et al., 2009). Thus, both transformational leadership and task interdependence enhance job and team commitment, and job and team commitment are positively related to ST shared leadership.

Lastly, transformational leaders build personal and social identification among senior team members, by articulating a clear mission and goals (Bass et al., 2003). Furthermore, those leaders pay attention to both individual’s need and to the achievement of group learning. So transformational leaders are expected to enhance performance of their individual followers, but also build collective confidence and team potency (Bass et al., 2003). The focus on group performance rather than individual’s achievements contribute to the development of ST shared leadership. Task interdependence makes it more likely that individuals focus on the collective instead of on individual efforts, because it increases collaboration among senior team members and team commitment. Therefore, we assume that task interdependence may have a positive effect on the relationship between transformational leadership and ST shared leadership.

Given that both transformational leadership and task interdependence arouse a sense of responsibility and that task independence increases the likelihood that senior team members focus on collective goals, we propose that the relationship between transformational leadership and ST shared leadership is stronger, when task interdependence is high. Therefore, the following hypothesis can be formulated:

(28)

Hypothesis 4. The positive relationship between transformational leadership and senior team shared leadership is moderated by task interdependence, so that this relationship is stronger for higher values of task interdependence.

Task interdependence, shared leadership and ambidexterity

Prior research has shown that task interdependence has a potential moderating role on the relationship between team processes and team outcomes (Shea & Guzzo, 1987; Guzzo & Shea, 1992; Barrick, Bradlet, Kristof-Brown & Colbert, 2007; Liden et al., 1997). In particular, Johnson (1973) argued that task interdependence increases communication between team members. Moreover, Staples and Webster (2008) and Crawford and Haaland (1972) stated that task interdependence facilitates team processes such as knowledge sharing. Team members will interact and communicate more when they rely on each other and when they need each other for the completion of a task. In turn, interaction and communication will foster knowledge exchange. As mentioned before, shared leadership also provides more frequent team interactions, which allows for more opportunities of information sharing (Weick & Robert, 1993). As stated before, information sharing is valuable for attaining organizational ambidexterity because it leads to higher levels of creativity, and managers’ creativity is an important condition for managing the paradox (Eisenhardt & Martin, 2000; Hoch, 2013; Andriopoulos & Lewis, 2009).

Since both shared leadership and task interdependence lead to more frequent team interactions, which stimulates knowledge exchange, this leads us to suggest that task interdependence may strengthen the relationship between ST shared leadership and organizational ambidexterity, when the level of task interdependence is high rather than low. Formally stated:

(29)

Hypothesis 5: The positive relationship between senior team shared leadership and organizational ambidexterity is moderated by task interdependence, so that this relationship is stronger for higher values of task interdependence

(30)

3. Methods

The main purpose of this study is to investigate the relationship between ST shared leadership and organizational ambidexterity and the influence of vertical transformational leadership and task interdependence. For this research, a deductive approach has been used and the hypotheses were formulated based on theories from existing literature and, subsequently, the propositions have been tested. This chapter outlines by what means the study was conducted. We will start with the research setting and the method of data collection, and end up with a discussion of the measurement and validation of constructs.

Research setting and data collection

This research is of explanatory character and aims to provide insights into the influence of vertical transformational leadership on organizational ambidexterity, mediated by ST shared leadership and moderated by interdependent tasks. In order to subject our moderated mediation framework to empirical testing, a sample of 2502 email addresses was identified, using the database of Integrand Nederland. In addition, another data set of 3219 email addresses, using the database from a commercial provider (the Chamber of Commerce), was identified. Both databases provided us with contact information of small, medium and large sized companies throughout the Netherlands. Of the 5721 questionnaires mailed, a total of 186 complete responses was returned, which corresponds with a 3,25 per cent response rate. The sample covered a wide range of industries, including primary industry (6 per cent), manufacturing industry (16 per cent), financial industry (11 per cent), professional service industry (43 per cent), retail (14 per cent) and public sector (9 per cent).

(31)

Due to the cross-sectional nature of the study, a survey questionnaire was used in order to reach a large number of respondents in a short time frame. Given the focus on leadership styles, the data was gathered from members of the senior teams. The senior managers were addressed via personalized email, asking them to participate in our online survey. We ensured respondents’ interest, conscientiousness and commitment, to provide accurate data by assuring them of confidentiality and by offering them a report with the results. The survey, designed for “senior team” members, includes items on ST shared leadership, interdependent tasks, organizational ambidexterity and the team leaders transformational leadership style. Measurement and validation of constructs

All of the constructs were measured by using existing scales from the literature. The multi-item scales were verified through various analyses. The items of all the constructs are provided in the appendix.

Dependent variable - Organizational ambidexterity

Following prior studies, we assumed that exploration and exploitation are orthogonal (Birkinshaw & Gibson, 2004; He & Wong, 2004). Therefore we measured organizational ambidexterity as a multiplicative score of exploration and exploitation. In order to do so, a two-step analysis was performed. First, the levels of exploration and exploitation were assessed, using existing scales from Jansen et al. (2009). Exploratory innovation is measured, using four-item scale (α = .76) that captures the degree to which units depart from existing knowledge and pursue innovations for emerging customers or markets. Respondents were asked to assess the relevance of the four items on a 7-point scale, ranging from “strongly disagree” to “strongly agree”. All items of the scale can be found in the appendix. The measure for

(32)

exploitative innovation is also adapted from Jansen et al. (2009) and this four-item scale (α = .71) captures the degree to which units build upon existing knowledge and pursue incremental innovations that meets the needs of existing customers or markets. Respondents indicated their agreement on these statements on a seven-point scale, ranging from “strongly disagree” to “strongly agree”. All items of the scale can be found in the appendix. To provide evidence of convergent and discriminant validity for exploratory and exploitative innovation, various types of analysis were performed. An exploratory factor analysis replicated the intended two-factor structure. However, one item “Our organization frequently expands services for existing clients” did not clearly loaded on its intended factor, since the cross loading for this item was above .40. Therefore, this item was removed from the scale. After this item was eliminated all items clearly loaded on its intended factor and all factors have eigenvalues greater than one (all factor loadings are above .52 with cross loadings below .30).

Second, to develop a measure for the level of organizational ambidexterity we followed prior studies (Jansen et al., 2006; Jansen et al., 2008) and computed the multiplicative interaction between exploratory innovation and exploitative innovation. This computation is in line with the previous mentioned argumentation that explorative and exploitative activities are orthogonal (Birkinshaw & Gibson, 2004; He & Wong, 2004).

Independent variable - Transformational leadership

Transformational leadership, referring to the ability of a leader to create a shared vision and to motivate team members to go beyond self-interest and focus on group goals (Hoch, 2013), is measured using an existing five-item scale (α = .90), developed by Hoch et al. (2010a). The scale represents the following five behaviours of transformational leadership: 1) idealized influence, 2) inspirational motivation, 3)

(33)

individualized consideration, 4) intellectual stimulation, and 5) expecting exceptional performance. Respondents were asked to assess the relevance of the five items on a 7-point scale, ranging from “strongly disagree” to “strongly agree”. Exploratory factor analysis indicated that all factors loaded on a single factor with an eigenvalue higher than one (factor loadings were all above .80).

Mediator – Senior team shared leadership

Shared leadership refers to the degree to which senior team members jointly engage in decision-making processes, and motivating and encouraging group maintenance. For measuring ST shared leadership, we used an existing scale developed by Jordi Borst (2014). This five-item scale (α = .84), based on the article of Smith and Tushman (2005), was developed to measure the extent to which strategic decisions within senior teams are jointly made and implemented. For further refinement, five academic experts and two corporate members were consulted, which resulted in a five-item scale. Respondents were asked to assess the relevance of the five items on a 7-point scale, ranging from “strongly disagree” to “strongly agree”. Exploratory factor analysis replicated a one-factor structure, with each item loading on the intended factor and the factor has an eigenvalue higher than one (factor loadings were all above .66). All items can be found in the appendix.

Moderator - Task interdependence

For this study, task interdependence is measured using an existing three-item scale (α = .90), developed by Liden et al. (1997). Liden et al. (1997) modified a scale developed by Pearce and Gregersen (1991), which originally consisted of 15 items. The scale captures the degree to which goal accomplishment requires the interaction of group members (Liden et al., 1997; Bligh et al., 2006). Respondents were asked to indicate their agreement on the three statements on a seven-point scale, ranging from

(34)

“strongly disagree” to “strongly agree”. Exploratory factor analysis indicated that all items loaded on a single factor, with an eigenvalue higher than one (factor loadings were all above .83). All items of the scale can be found in the appendix.

Control variables

This study controlled for several alternative explanations by including relevant control variables. First, large organizations, compared with smaller organizations, may have more recourses but may not have the flexibility to achieve ambidexterity (Ahuja & Lampert, 2001). Therefore, the number of full-time employees within the organization was included to control for firm size. In addition, Gilbert (2005) argued that incumbent firms are more likely to engage in exploitative activities than young firms. For this reason, the number of years from the organization’s founding was included in order to account for firm age. Third, the senior team size could affect the heterogeneity of senior teams, which in turn may affect the extent to which leadership is shared and may impact realization of ambidexterity (Siegel & Hambrick, 2005). This control variable is measured through the number of senior executives who are responsible for important decisions within the organization. Finally, industry effect may impact the extent to which organizations engage in exploratory or exploitative activities (He & Wong, 2004). Therefore, six industry dummies are included, based on Standard Classification codes: primary industry, manufacturing industry, financial services, retail trade, professional services and public administration.

(35)

4. Results

This chapter will give a description of the results. First, the descriptive statistics and the correlations of the research variables will be addressed. Further, the multiple regressions will be elaborated on in order to check the assumptions for the moderated mediation model. The analysis in process will confirm or disconfirm the construed hypotheses. To further verify the findings, a post hoc analysis will address the conditional effects of both the mediator and the moderator. Lastly, an additional analysis is performed to ensure the robustness of the findings and to minimize the potential contaminating influence of common method bias.

Analyses

In order to conduct further statistical analysis, several assumptions have been checked. First of all, all incomplete questionnaires have been deleted from the data set. Second, visual inspection of scatterplots for ambidexterity and ST shared leadership shows that the residuals contain no notable outliers. Regarding the normality, visual inspection of the frequency histogram shows that the dependent variable and the moderator are both approximately normally distrusted (see appendix 2 and 3 for the histograms). In addition, the mean of ambidexterity is 27.24 and the median is 27.31. The mean of ST shared leadership is 5.52 and the median is 5.80, so we can conclude that the normality assumption is satisfied and we can conduct further analysis.

(36)

Table 1. Means, Standard deviations, Correlations and Reliabilities

Notes. Numbers in parentheses on the diagonal are Cronbach’s alphas of the composite scales. All correlations above |0.15| are significant at p < 0.05. n = 186. Table 1 depicts the descriptive statistics and correlations of the variables of this research and the control variables. The variance inflation factors (VIFs) for each of the regression equations were calculated in order to examine multicollinearity. The maximum VIF within the model was 1.19, which is well below the rule-of-thumb cut-off of 10 (Neter et al., 1990). The frequency tables of the control variables senior team size, firm size and firm age indicate skewness and accordingly they are transformed with natural log. See appendix 3 for the frequency tables of senior team size, firm age and firm size.

Table 2 presents the results of the regression analyses for organizational

ambidexterity. The baseline model (Model 1) contains the control variables (Firm size, Firm age, Senior team size, and Industry dummies). Following Baron and Kenny (1986), we estimated three regression equations in order to test for mediation. First, regressing the mediator on the independent variable (e.g. ST shared leadership on transformational leadership); second, regressing the dependent on the independent

M SD 1 2 3 4 5 6 7 8 9 10 11 12 13

1. Senior Team Size .49 .50 -

2. Firm Size .98 .43 .61 - 3. Firm Age 1.17 .35 .31 .50 - 4. Primary Industry 0.06 .25 .01 .04 .14 - 5. Manufact, Industry .16 .37 .02 .09 .13 -.12 - 6. Retail Industry .14 .35 -.13 -.03 .04 -.11 -.18 - 7. Financial Industry .11 .31 .08 .06 .06 -.09 -.15 -.14 - 8. Service Industry .43 .50 -.03 -.16 -.29 -.23 -.38 -.35 -.30 - 9. Public Industry .09 .29 .10 .12 .13 -.08 -.14 -.13 -.11 -.28 - 10. Ambidexterity 27.2 9.33 .15 .11 -.16 -.05 .01 -.05 -.04 .23 -.24 - 11. ST Shared Leadership 5.52 .92 -.14 -.18 -.13 -.04 -.05 -.06 -.06 .13 .04 .26 (.84) 12. Transfor. Leadership 5.28 1.13 .04 -.06 -.11 -.07 -.17 -.08 .04 .27 -.12 .39 .46 (.90) 13. Task Interdependence 4.14 0.68 -.07 -.07 .06 .08 -.10 -.13 -.02 .12 .06 .02 .28 .12 (.90)

(37)

variable (e.g. ambidexterity on transformational leadership); and third, regressing the dependent variable on both the independent and on the mediator variable (e.g. ambidexterity on both transformational leadership and ST shared leadership). The results of model 6 satisfy the first condition of Baron and Kenny’s (1986) approach for mediation, as the effect of transformational leadership on ST shared leadership is positive and statistically significant (β = .478, p < .001). Model 2 represents the second step of the mediation analysis, as it shows the effect of transformational leadership on ambidexterity. This positive and statistically significant influence of transformational leadership on ambidexterity (β = .35, p < .001) satisfies the second condition for mediation. Thereafter, we consider the influence of the proposed mediator on organizational ambidexterity. In model 4 we observe that ST shared leadership is positively associated with organizational ambidexterity (β = .31, p < .001). These findings represent confirmation of the third conditions required for mediation. Therefore, the empirical analysis suggests that transformational leadership influences organizational ambidexterity and that ST shared leadership mediates this relationship. Thus, hypotheses 1, 2 and 3 are supported.

(38)

Table 2. Results of hierarchical regression analyses: Transformational leadership, ST Shared Leadership, Task Interdependence, and Ambidexterity

Note. Standard regression coefficients are reported. * p < 0.05, ** p < 0.01, *** p < 0.001

Subsequently, the mediation effect is analysed with process model 4 (Hayes, 2013). The model explains a significant proportion of the variance in organizational ambidexterity, showing R2 of .288, F (6.39), p < .0001, which means that 28.8% of the variance in organizational ambidexterity is explained by this model. The effect of ST shared leadership on organizational ambidexterity (b = 2.91, SE = 0.76, t (186) = 2.19, p < .05) and the effect of transformational leadership on ST shared leadership (b = .387, SE = .06, t (186) = 7.00, p < .0001) is significant. Thus, both the hierarchical regressions and the process analysis show that the total mediation effect is significant.

Ambidexterity Shared Leadership

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Control variables Firm Size .18 .21 .23* .23* .21* -.11 -.09 Firm Age -.20 -.22 -.22* -.22* -.21* -.01 -.05

Senior Team Size .11 .10 .12 .15 .15 -.10 -.08

Primary .15 -.09 .07 .09 .11 .13 .04 Manufacturing .22 .19 .15 .17 .19 .23 .14 Retail .22 .14 .12 .12 .17 .15 .08 Financial .17 .06 .04 .10 .13 .10 .01 Service .44 .25 .21 .31 .34 .24 .09 Public -.03 -.08 -.12 -.17 -.10 .26 .16 Independent variable Transformational leadership .35*** .27*** .48*** .46*** Mediator variable Shared leadership .16* .29*** .29*** Moderator variable Interaction .08 -.05 Task Interdependence -.02 .23*** Adjusted R2 _.10 _.27 _.29 _.24 _.19 _.23 _.32 Δ Adjusted R2 .14 .11 .02 .09 .08 .21 .26

(39)

Therefore, it can be inferred that ST shared leadership mediates the relationship between transformational leadership and organizational ambidexterity and accordingly, hypotheses 1, 2 and 3 are supported.

Figure 2. The outcomes of the process analysis visualized in the conceptual model

Note. The bootstrapped model shows regression coefficients (b-values) and confidence intervals based on 1,000 samples.

p < .05*, p < .001**, p < .0001***

Next, we test the moderation hypotheses, starting with the moderating role of task interdependence on the relationship between the mediator and the dependent variable (respectively ST shared leadership and organizational ambidexterity). A hierarchical regression shows (model 5) that the moderating role of task interdependence on the relationship between ST shared leadership and organizational ambidexterity is statistically insignificant (β = .08, p > .05). Second, the moderating role of task interdependence on the relationship between transformational leadership and ST shared leadership is tested (model 7). The results of the regression indicate that the proposed moderating role of task interdependence on the relationship between the independent and the mediator is statistically insignificant (β = -.05, p > .05).

Hypotheses 4 and 5 are also analysed with process, based on model 58 (Hayes, 2013). The model explains a significant proportion of the variance in organizational ambidexterity, showing R2 of .277, F (5.06), p < .0001, which means that 27.7% of

(40)

the variance in organizational ambidexterity is explained by this model. However, the interaction effect of transformational leadership and task interdependence on ST shared leadership is statistically not significant (b = -.04, SE = .08, t (186) = -.71, p > .05). The interaction effect of ST shared leadership and task interdependence on organizational ambidexterity is also statistically insignificant (b = .70, SE = .96, t (186) = 1.50, p > .05). In sum, both the hierarchical regression analysis and the analysis with process indicate that the moderating role of task interdependence on the relationship between the dependent and moderating variable, and on the relationship between the independent and the moderating variable, is statiscally not significant. Therefore, hypotheses 4 and 5 are not supported.

Post hoc analysis

To further verify our findings, we conducted several additional analyses. First, we used process model 1 (Hayes, 2013) to assess whether there is a conditional effect of ST shared leadership on organizational ambidexterity at different values of the moderator task interdependence. The Johnson-Neyman technique is used to measure the moderator value that defines the significance region. The results show that the relationship between ST shared leadership and organizational ambidexterity is not significant for low values of the moderator task interdependence. This means the relationship between ST shared leadership and organizational ambidexterity is slightly conditional.

Second, we used process model 1 (Hayes, 2013) once more to assess whether our data shows a conditional effect for the relation between transformational leadership and ST shared leadership. Again, the Johnson-Neyman technique is used to measure the moderator value that defines the significance region. However, the results indicate that there is no statistical significance transition point within the observed

(41)

range of the moderator. This means that the relationship between transformational leadership and ST shared leadership is significant at all observed values of the moderator. So we can conclude that task interdependence does not have an influence on the positive relationship between transformational leadership and ST shared leadership. In addition, given the results of the second Johnson-Neyman test, we conducted an additional regression analysis to examine the direct effect of task interdependence on ST shared leadership. The results indicate that the main effect of task interdependence on ST shared leadership is positive and statistically significant (β = .278, p > .001). This pattern of findings leads us to suggest that task interdependence is an antecedent of ST shared leadership, rather than that task interdependence moderates the relation between transformational leadership and ST shared leadership.

Since the results revealed no statistically significant effects of task interdependence on the relationships between transformational leadership and ST shared leadership, and between ST shared leadership and organizational ambidexterity, an additional curvilinear regression was performed in order to test for a quadratic effect on the previous asserted relations. The curvilinear effect of task interdependence on the relation between ST shared leadership and organizational ambidexterity was tested with process model 1 (Hayes, 2013). The model explains a significant proportion of the variance in organizational ambidexterity, showing an R2

of .238, F (4.51), p < .0001, which means that 23.8% of the variance in organizational ambidexterity is explained by this model. However, the results pointed out that the quadratic effect of task interdependence is statistically insignificant (b = .03, SE = .49, t (186) = .05, p > .05). Another curvilinear regression in process (model 1) examined the quadratic effect of task interdependence on the relationship between

Senior team shared leadership and organizational ambidexterity : the influence of vertical transformational leadership and task interdependence