Individual ambidexterity and team climate : the moderating role of supportive leadership

(1)

“Individual ambidexterity and team

climate: the moderating role of supportive

leadership”

Master thesis: MSc in Business Administration - Strategy track

Faculty of Economics and Business Student name: Havi Kurda

Student number: 10769927 Supervisor: B. Lima

Academic year: 2017 – 2018 Date: 22ndJune 2018

(2)

Statement of Originality

This document is written by Student Havi Kurda 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 that 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.

Amsterdam, 22nd June 2018 Havi Kurda

(3)

Abstract

The phenomenon of organizational ambidexterity has become very popular in the recent years. However, there are not many studies on the emergence of ambidexterity at lower hierarchical levels. This is strange, since the capabilities to exploit and explore are capabilities that in fact lie in individuals themselves. We theorize that team climate factors (i.e. climate for trust, climate for excellence and climate for affiliation), may positively contribute to individuals’ ability to deal with the conflicting demands associated with ambidexterity. In addition, we propose that in such teams, team leaders’ supportive attitudes may play an important role in further encouraging and facilitating ambidextrous behavior, and that the extent to which individuals are able to achieve ambidexterity in such teams is contingent upon team leaders’ supportive behavior. Using data collect by a survey, our study reveals that particularly team climate for excellence matters for the emergence of individual ambidexterity. It also reveals that with regard to the climate for trust, it is better for team leaders to not show supportive behavior as this has a negative effect on individual ambidexterity.

(5)

1. Introduction

In today’s dynamic business environment, no business is longer ensured of its dominant and favourable position in the market. Managers of large established companies continuously face the challenge of how to respond effectively to the requirements of flexibility on the one hand, and efficiency on the other. Sustainable success therefore lies in the capability to exploit current competencies and explore new possibilities – a capability referred to as organizational ambidexterity (Raisch, Birkinshaw, Probst & Tushman, 2009).

In the past two decades, prior research has mainly focused on reaching ambidexterity at the organizational or unit level. Even though throughout the years more research has been done on determining important antecedents to ambidexterity, studies about ambidexterity at the individual level are still scarce. Although organizational ambidexterity has been linked to success, the capabilities to exploit and explore are in fact capabilities that lie in the individuals themselves (Keller & Weibler, 2014). Since most research has taken a macro perspective in studying ambidexterity, investigating the micro level foundations of ambidexterity can assist in understanding the phenomenon of organizational ambidexterity even further (Raisch et al., 2009). Therefore, studying ambidexterity at the individual level can be particularly helpful for companies that are facing the opposing demands of exploration and exploitation in their environments.

Another important argument to investigate the individual level of ambidexterity, is that when only the organizational level of analysis is taken into account, individual homogeneity is assumed, while individuals’ capabilities and perceptions differ from person to person (Keller & Weibler, 2014). The notion of heterogeneity across individuals and as a consequence heterogeneity in individual ambidexterity, is important to understand, since employees can cumulatively influence organizations’ ambidexterity and as a result, firm performance as well (Raisch et al., 2009; Laureiro-Martínez, Brusoni & Zollo, 2010). Studying individual

(6)

ambidexterity thus is highly relevant, as ambidextrous individuals play a key role in achieving organizational success and consequently the long-term survival of firms. Consistent with existing literature, we conceptualize individual ambidexterity as the individual’s ability to pursue both exploration and exploitation and flexibly shift between them (Mom, Van den Bosch & Volberda, 2009; Good & Michel, 2013).

Since the majority of previous research has focused solely on structural antecedents and macro levels of ambidexterity, important determinants for ambidexterity at the individual level and knowledge about its interactions with other levels of analysis is a relatively underdeveloped research area (Laureiro-Martínez, Brusoni & Zollo, 2010). As a result, there is a lack of insight in what antecedents may cause individual ambidexterity (Good & Michel, 2013; Andriopoulos & Lewis, 2009; Laureiro-Martínez, Brusoni & Zollo, 2010; Turner, Swart & Maylor, 2013). Previous research may have overlooked antecedents that influence ambidextrous behavior of individuals. In Birkinshaw and Gibson’s (2004) influential article about contextual ambidexterity, it is suggested that the demand for ambidexterity can be resolved by creating a supportive context that helps individuals to make better choices between alignment- and adaptability-oriented activities (Birkinshaw & Gibson, 2004). Despite their call for more research into contextual factors and individual ambidexterity, there is still an important factor which has – to the best of our knowledge - not been researched yet in combination with individual ambidexterity. This overlooked contextual issue within individual ambidexterity literature, is the role that the direct working environment may have on individuals, such as individuals’ perceived functioning of their teams.

A study has already demonstrated that socio-psychological factors such as team cohesion and team efficacy help team members to deal with the paradoxical challenge of combining explorative and exploitative learning efforts (Jansen, Kostopoulos, Mihalache & Papalexandris, 2016). Jansen et al. (2016) argue that particularly these two team-level factors

(7)

positively influence team-level ambidexterity. Other studies in the socio-psychology research field seem to be consistent with this view. It is proven that individuals’ social environment can influence individuals’ actions and as a consequence, their behaviors as well (Salancik & Pfeffer, 1978). Furthermore, team members’ shared perceptions about cognitive, emotional and affective states (Marks, Mathieu & Zaccaro, 2001; Mathieu, Maynard, Rapp & Gilson, 2008) can help in explaining their ability to deal with the conflicting demands associated with ambidexterity. Smith and Tushman (2005), for example, demonstrate that teams with shared mental models and collective cognitive processes are better able to understand complex goals and conflicts between contrasting learning demands, such as exploration and exploitation. Thus, it appears that team-level factors and individuals’ perceived functioning of their teams can directly influence individual behavior.

A concept that captures many team-level antecedents to individual ambidexterity, is team climate. As the terms organizational climate, team climate and group climate are often used interchangeably, no consensus has been reached on the definition of team climate yet. However, the majority of the studies tend to define team climate as: ‘the norms, atmosphere, practices, interpersonal relationships, enacted rituals and ways of working developed by a team’ (Anderson & West, 1994; Edmondson, Kramer & Cook, 2004; Anvuur & Kumaraswamy, 2007). We argue that among many antecedents to individual ambidexterity, especially team climate is the most important, as teams are part of individuals’ direct working environments. Furthermore, team climates characterized by trust, affiliation and teams promoting high excellence, may stimulate and assist individuals in achieving better outcomes (Pirola-Merlo et al., 2002; Bock, Zmud, Kim & Lee, 2005). Incorporating team climate factors thus could help understand the phenomena behind individuals’ ambidextrous behavior. When it comes to teams and ambidexterity literature, prior research has only focused on the role of top-management teams and managers’ ambidexterity (Lubatkin, Simsek, Ling & Veiga, 2006; Jansen, George,

(8)

Van den Bosch & Volberda, 2008; Carmeli & Halevi, 2009). So far, there are no studies that have combined team climate factors and regular employee ambidexterity.

To address this gap in the literature, this study seeks to advance the understanding of the somehow unaddressed, yet important micro level of individual ambidexterity by testing the relationships between individual ambidexterity and team climate. This study therefore seeks to answer the following research question:

RQ1: “What is the relationship between team climate and individual ambidexterity?”

In addition to the examined main relationship between time climate and individual ambidexterity, it is important to consider the attitude of the team leader, as he or she might account for differences in work behavior as well (Jansen et al., 2016). Many studies have already pointed out the importance of leadership and its link with organizational success in general (Lowe, Kroeck, & Sivasubramaniam, 1996; Lovelace, Shapiro & Weingart, 2001; Jung & Sosik, 2002). With regard to ambidexterity, there are studies that have examined different leadership characteristics. Nemanich and Vera (2009), for example, found proof for their hypothesis that transformational leadership positively contributes to team ambidexterity. On the other hand, a study by Jansen et al. (2016), demonstrates that supportive leadership moderates the relationship between team level factors and team level ambidexterity. Yet another leadership style, which is referred to as ambidextrous leadership, has been found to have positive effects on individual level ambidexterity (Rosing, Frese & Bausch, 2011l; Kauppila & Tempelaar, 2016). These studies have in common that they focus on characteristics of leadership in general. The complex processes that are behind the combination of leadership and ambidexterity, however, are reflected in the inconsistent results presented above.

(9)

Although prior research has investigated several leadership characteristics alongside ambidexterity in general, we argue that specifically supportive leadership should be studied alongside team climate and individual ambidexterity, as it covers “the encouragement of individual initiatives, clarification of individual responsibilities, provision of clear and complete performance evaluation feedback, a strong task orientation, emphasis on quality group relationships and trust in organizational members” (Van de Ven & Chu, 1989, p. 340). As we study team level antecedents to individual ambidexterity, we believe that supportive attitudes of team leaders are most relevant to incorporate. As team leaders may further encourage, facilitate and stimulate team members to express more critical or diverse opinions, as well as encourage teams to recover bad situations, it might be interesting to observe whether individuals in team climates with more supportive leaders are able to cope better with ambidexterity demands and show a different, possibly higher level of individual ambidexterity, compared to teams with lower levels of supportive leadership. Therefore, this study tests the moderating effect of supportive leadership on the focal relationship between team climate and individual ambidexterity. This study also answers a second research question, namely:

RQ2: “What is the moderating effect of supportive leadership on the relationship between team climate and individual ambidexterity?”

By studying the relationships between team climate, individual ambidexterity and supportive leadership, this study makes several contributions. First, this study extends the understanding of micro level ambidexterity by elaborating on how and under what conditions individuals will be able to achieve ambidexterity. Studying ambidexterity at lower hierarchical levels will also contribute to a better general understanding of ambidexterity at the organizational level. Second, this study extends the literature on individual ambidexterity by its

(10)

multi level approach. It does so by adding team climate as a relevant contextual factor that might influence individuals’ ambidexterity level. When including team climate principles, also better insights can be provided into team level antecedents to individual ambidexterity, and the antecedents that drive variation in individuals’ ambidexterity. This study thus combines different levels of analysis and a multi-level approach to ambidexterity research.

Furthermore, we move beyond direct effects of team level antecedents to individual ambidexterity, by examining the role of supportive leadership. Incorporating leadership style as a potential moderator gives better insights into how team climate, individual ambidexterity and supportive leadership behaviors are related. Lastly, the empirically tested results of this study can help organizations to recognize and to create supportive team climates in which ambidexterity is stimulated. The insights related to leadership and individual ambidexterity can be particularly helpful for organizations to recognize supportive leaders and ambidextrous employees, as this will maybe help them in achieving firm performance.

This research is structured as follows: first, it provides a literature review that elaborates on the concepts of ambidexterity, the influence of team climate as well as the moderating effect of leadership. Consequently, the hypotheses and theoretical framework are discussed. Second, the method section provides insights into the process of data collection and methods used to conduct this study. Then, we discuss the analyzed data in the results section, followed by the discussion and conclusion, in which limitations and future research directions are also presented.

(11)

2. Literature review

2.1 Exploration and exploitation

When researching organizational learning, the concepts of ambidexterity, exploration and exploitation often go hand in hand. After the introduction of March’ famous framework on exploration and exploitation, many scholars have dived into further researching organizational phenomena related to exploration and exploitation (March, 1991). Over the years, this framework has been extended and applied to many fields of research, focusing on diverse contexts and multiple levels of analysis (Lavie, Stettner & Tushman, 2010). The following section therefore describes and defines the the concepts of exploration and exploitation.

When it comes to exploration and exploitation, organizations tend to focus on either one of those dimensions and its associated activities, which in turn can lead to a blinded focus and organizational myopia (Levinthal & March, 1993). Focusing on only one of those can lead to missed opportunities as firms may not be fully utilizing new and existing knowledge. However, engaging in both exploration and exploitation can also lead to many tensions, requiring firms to make trade-offs between those activities (Raisch et al., 2009; He & Wong, 2004; March, 1991).

A common mistake would be to call the combination of explorative activities and exploitative activities ambidexterity, as these concepts are mixed many times. It is therefore important to distinguish the concept of ambidexterity from the combination of both exploration and exploitation for two reasons. First, by only having exploration and exploitation related activities in a company, it cannot be specified what the relationship between the two is and more importantly, not be understood to what extent each of those lead to success. The other reason is that it is the balance of exploration and exploitation that most likely leads to firm performance (March, 1991). Therefore, ambidexterity should be seen as the appropriate balance between exploration and exploitation, rather than just the presence of those two alongside each

(12)

other. This view is consistent with March’ (1991) view in his original work about ambidexterity not being orthogonal, but rather two ends of a continuum (Gupta, Smith & Shalley, 2006; Lavie & Rosenkopf, 2006; Birkinshaw & Gibson, 2004; Mom, Van den Bosch & Volberda, 2007). Nevertheless, both concepts are crucial parts of ambidexterity research and therefore should be analysed more thoroughly.

Exploration and exploitation have in common that both are related to learning activities (Gupta, Smith & Shalley, 2006). Exploration tends to primarily focus on gaining new knowledge by focusing on experimentation and searching for new opportunities, whereas exploitation can be seen as activities that use already existing knowledge, such as implementation and efficiency related activities (Mom, Van den Bosch & Volberda, 2009). The combination of those two activies also share commonalities with other logics, such as the the demand for flexibility versus efficiency (Adler et al., 1999) as well as discontinuous versus incremental innovations (Benner & Tushman, 2003; Smith & Tushman, 2005; Simsek et al., 2009). Although both are related to learning, the learning processes can be different in the sense that different organizational structures, competencies and mental models might be needed, while the resources needed for both activities are scarce (Jansen et al., 2008; Kang & Snell, 2009; March, 1991). As a consequence of the competition for scarce resources, firms risk that focusing too much on one dimension can drive the other dimension out.

Activities associated with exploration often involve uncertainty and risk. This is in contrast with exploitation, which builds on existing knowledge and focuses more on increasing productivity, efficiency, and enhancing current capabilities – thereby reducing risk (Benner & Tushman, 2003). In line with the definitions of March (1991), we define exploration as activities that include ‘things captured by terms such as search, variation, risk taking, experimentation, play, flexibility, discovery, innovation’ and exploitation as activities that include ‘such things as refinement, choice, production, efficiency, selection, implementation,

(13)

execution’ (March 1991, p. 71). Thus, exploration focuses on expanding knowledge and competencies and exploitation focuses on activities that deepen existing competencies.

2.2 Organizational ambidexterity

In order to understand the ambidexterity phenomenon further, we start with the most prominent works in the literature on ambidexterity. Since the first use of the term ambidexterity by Duncan (1976), many scholars have elaborated further on the subject. According to Duncan, organizations should focus both on alignment and adaptation in order to deal with conflicting demands (Raisch & Birkinshaw, 2008). This can be done by establishing certain structures in business units or within business units (Duncan, 1976 as cited in Raisch & Birkinshaw, 2008). This type of ambidexterity is referred to as structural ambidexterity by Birkinshaw and Gibson (2004), who take a slightly different approach by explaining ambidexterity with regard to the context (Birkinshaw & Gibson, 2004). Their view is also referred to as the contextual approach to ambidexterity.

According to He and Wong (2004), firms are seen as ambidextrous whenever a firm has high scores for both explorative and exploitative innovation. They propose to look at the difference between the scores, and argue that firms should only be considered ambidextrous when both dimensions are equally emphasized (He & Wong, 2004). Another view that builds on Duncan’s (1976) work, is Raisch and Birkinshaw’s view who describe ambidextrous firms as ‘aligned and efficient in their management of today’s business demands while simultaneously adaptive to changes in the environment’ (Raisch & Birkinshaw, 2008, p. 375). Yet another perspective on ambidexterity is the one of the balance and combination dimension. Firms should not only aim for a combination of explorative and exploitative activites, but also try to reach a balance between the two dimensions (Cao, Gedajlociv & Zhang, 2009). As can

(14)

be seen, the definitions, explanations and approaches for ambidexterity can differ, depending on which view is chosen.

This study builds on the ambidexterity concept as defined by O’reilly and Tushman (2008), in which they describe ambidexterty as not being solely about engagement in exploration and exploitation, but also about the ability to shift between these different modes. Firms should not only aim for more efficiency and innovation, but rather focus on developing the competencies deemed necessary to compete in new markets in order to ensure firm survival (O’Reilly & Tushman, 2008). This definition of ambidexterity at the organizational level provides a base for a better general understanding of ambidexterity at the individual level, which we discuss in the next section.

2.3 Ambidexterity at the individual level

As can be seen, the majority of prior research on ambidexterity has primarily focused on reaching ambidexterity at the organizational or business unit level. Whereas the previous section provided insights into the general ambidexterity phenomenon through an organization’s point of view, this section seeks to clarify how balancing and shifting between the different modes of exploration and exploitation leads to ambidexterity at the individual level.

Our understanding of ambidexterity has only recently been expanded by adding multiple levels of analysis, such as ambidexterity at the team level, but also at the level of individuals (Birkinshaw & Gibson, 2004; Nemanich & Vera, 2009; Good & Michel, 2013). Studies on individual ambidexterity are present within the ambidexterity literature, but have not been studied extensively. While organizational ambidexterity can be easier connected to organizational outputs, and thus firm performance, individual ambidexterity has a less clear pathway towards firms’ profitability. However, insights into factors that drive ambidextrous behaviors also lead to a better understanding of ambidexterity at the organizational level, which

(15)

eventually leads to firm performance (Tushman & O’Reilly, 1996; Auh & Menguc, 2005; Lubatkin et al., 2006). Therefore, organizational success also lies in individuals’ ability to combine and balance explorative and exploitative efforts. This makes more research into this area of high importance.

Consistent with prior research, this study conceptualizes individual ambidexterity as the ability of individuals to engage in both explorative and exploitative activities in a balanced manner (Bledow et al., 2009; Kauppila & Tempelaar, 2016) and the ability to shift between them (Mom, Van den Bosch & Volberda, 2009; Good & Michel, 2013). This also includes individuals’ ability to combine and deal with opposing task performances, as well as the learning efforts that are required for ambidextrous behavior. Individuals thus are required to continuously cross-fertilize knowledge between explorative and exploitative demands (Taylor & Helfat, 2009). Since the ambidexterity construct consists of two dimensions, a factor that leads to an increase in one of its dimensions should also lead to an increase in individual ambidexterity. This implies that exploration and exploitation are mutually enabling (Kauppila & Tempelaar, 2016). Previous research has pointed out mechanisms that help individuals to deal with these opposing demands. The majority of available research can be divided into two substreams; one that draws upon psychological and behavioral theory, and one that builds on organizational context to explain individual ambidexterity.

The degree to which individuals are able to reach higher levels of ambidexterity is for a large part influenced by individual characteristics (Raisch et al., 2009). Kao and Chen (2016), for example, have found relationships between individuals’ intrinsic motivation, personality traits and individual ambidexterity. They argue that intrinsic motivation is an important driver of individual ambidexterity and that individuals’ emotional intelligence plays an important role in explaining this relationship. This is confirmed by several other studies, which show that a high degree of individual’s emotional intelligence accounts for differences in task adaptive

(16)

performance and also individual ambidexterity (Rosing, Frese & Bausch, 2011; Good & Michel, 2013). As ambidexterity requires individuals to deal with contradictory demands, scholars have also studied the influence of cognitive ability and cognitive flexibility with regard to this matter. This is what Smith and Tushman (2005) call “paradoxical thinking”. Other characteristics, such as divergent thinking, self-efficacy and the different identities that individuals hold are likely to influence ambidextrous behavior as well (Good & Michel, 2013; Keller & Weibler, 2015; Kauppila & Tempelaar, 2016; Tempelaar & Rosenkranz, 2017). These views are in line with general findings about individual behavior being heavily influenced by personality traits. One explanation for the variance in individuals’ ability to behave ambidextrous, thus can be traced back to individual characteristics.

Although individual characteristics and personality play a role in individual ambidexterity, another stream of research concentrates on the organizational context and shows that the context can induce individuals to engage in both exploration and exploitation. Birkinshaw and Gibson (2004) were the first ones to introduce the concept of contextual ambidexterity, which they define as ‘the capacity to simultaneously achieve alignment and adaptability at a business-unit level’ (p. 209). In their approach they argue that contextual factors such as stretch, discipline, support and trust can facilitate ambidexterity (Birkinshaw & Gibson, 2004). In such supportive contexts, individuals are encouraged to make their own decisions between alignment- and adaptability-oriented activities. In this case, the concepts of alignment and adaptability closely resemble exploitation and exploration. Building further on the organizational context, prior research emphasized contextual antecedents to individual ambidexterity, such as team behavioral integration (Lubatkin et al., 2006), leadership style (Nemanich & Vera, 2009; Kauppila & Tempelaar, 2016), shared team vision and team reward structures (Jansen et al., 2008). These studies have in common that a focus on social means could assist individuals in handling the paradoxical demands associated with exploration and

(17)

exploitation.

To conclude, there are many factors that likely influence individual ambidexterity. As can be seen, the organizational context provides a fruitful area for setting and explaining conditions that may lead to ambidextrous behavior. We argue that especially contextual factors should be investigated alongside individual ambidexterity, as there is such a variety and diversity within organizational contexts, which may not have been explored (fully) yet. In accordance with the contextual approach to ambidexterity, we suggest that the team level antecedents and leadership style are particularly important to investigate in combination with individual ambidexterity. The following parts will therefore be dedicated to elaborate further on these subjects.

2.4 Team climate

Although research has been done on organizational context in combination with ambidexterity, team level antecedents to individual ambidexterity seems to be a poorly developed research field. As mentioned previously, several team level antecedents to ambidexterity have been investigated, such as behavioral integration and teams’ shared vision (Lubatkin et al., 2006; Jansen et al., 2008). However, a majority of these studies have only focused on managers and top-management teams (Carmeli & Halevi, 2009; Mihalache, Jansen, Van den Bosch & Volberda, 2014). There is fewer research available on antecedents to ambidexterity in combination with non-managerial teams.

Although empirical research on individual ambidexterity in combination with team level antecedents is very limited, one study by Jansen et al. (2016) has already demonstrated the importance of team level antecedents, by arguing that team cohesion and team efficacy can positively contribute to a team’s ambidexterity. This conclusion is in line with socio-psychological research, in which often the immediate social environment is proven to influence how individuals formulate perceptions, attitudes and behaviors (Salancik & Pfeffer, 1978).

(18)

Anderson and West (1996) also suggested that teams’ characteristic and social processes are important determinants of the quality of innovations and performance. Prior research shows that especially the social context is a fundamental driver of individuals’ behavior (Salancik & Pfeffer, 1978; Robinson & O’Leary-Kelly, 1998). Individuals usually tend to analyze their work environments and adjust their actions based on these environments (Schneider, 1975). Furthermore, Salancik and Pfeffer (1978) argue that the effect of the social environment on individuals’ behavior primarily depends on the degree to which individuals share common beliefs with their social environment (Salancik & Pfeffer, 1978). Therefore, it could be that organizational teams could influence the degree of individual ambidexterity.

Based on these arguments, we suggest that there is room for broadening our knowledge on the role of organizational teams in providing an enabling context for ambidexterity. A way to do this is by integrating team climate principles into ambidexterity research. The question yet to be answered is, what is team climate? Although there is evidence that organizational climate influences individuals’ actions and behavior, scholars have not yet reached consensus about an overall construct of organizational climate. Previous research shows that the organizational climate concept is often composed of multiple factors. As previously demonstrated, one way to describe team climate is as follows: ‘the norms, atmosphere, practices, interpersonal relationships, enacted rituals and ways of working developed by a team’ (Anderson & West, 1994; Edmondson, Kramer & Cook, 2004; Anvuur & Kumaraswamy, 2007).

Firms can stimulate their employees by formulating a certain team climate that encourages them to communicate freely, share ideas and to explore non-routine alternatives (Edmondson, 1999; Jaw & Liu, 2003). Other studies confirm this by arguing that team climate shapes individuals’ behavior by common practices, shared beliefs and value systems (Janz, Colquett & Noe, 1997; Sveiby & Simons, 2002; Chen & Huang; 2007; Xue, Bradley & Liang,

(19)

2011). Furthermore, team climate can also play a large role in influencing existing routines, individual outcomes (Kuenzi & Schminke, 2009) and also individual performance (McKay, Avery, & Morris, 2008). We argue that the team climate for trust, excellence and affiliation are most important to study alongside individual ambidexterity, because it is specifically these team climate factors that encourage discussion, knowledge sharing and information exchange (Bock et al., 2005; Jansen et al., 2006; Nemanich & Vera, 2009). Bock et al. (2005), for example, showed that individuals who perceived a climate to be trusting, innovative and trustworthy, more likely adhered to the subjective norm of sharing knowledge. A supportive team context that fosters communication and collaboration, thus may enable individuals to better keep up with explorative and exploitative activities (Jansen et al., 2006). Furthermore, these team climate factors could enhance the feeling of psychological safety and as a consequence, support ambidexterity by positively influencing explorative and exploitative processes (Nemanich & Vera, 2009). We argue that such team climates may help individuals to deal with difficult task demands.

Anderson and West (1998) have attempted to develop a validated measure for team climate earlier. In their study they propose that team climate consists of vision, participative safety, support for innovation and climate for excellence. Here, vision refers to ‘a valued outcome, which represents a higher-order goal and motivating force at work (Anderson & West, 1998, p. 310). Participative safety refers to feelings of safety, lack of fear and affection (Pirola-Merlo et al., 2002). Support for innovation measures the degree to which new innovative ideas are encouraged and climate for excellence captures a group’s shared norms concerning excellence (Anderson & West, 1998). In contrast, Bock et al. (2005) created a climate construct that only consists of affiliation, fairness and innovation. Our argument for choosing the team climate for trust, excellence and affiliation, is that these factors might influence both the exploitation and exploration dimension, and thus are more relevant to study in combination

(20)

with individual ambidexterity than the other climate factors. There are for example climate factors like ‘support for innovation’ and ‘innovativeness’, which we expect to enable explorative behavior more than exploitative behavior. Since we are looking for indicators that can capture both dimensions of the ambidexterity measure and the ones that could help individuals with their ability to shift between different modes, we choose to focus on the team climate for trust, excellence and affiliation specifically. We thus adapt separate validated measures of organizational climate studies to fit the ambidexterity context. In the next chapter we seek to elaborate on these dimensions in combination with individual ambidexterity, as well as the expected relationships and hypotheses.

2.5 Supportive leadership

Organizational research has pointed out the important role of leadership behavior in team and organizational outcomes (Bass, 1990; Lowe, Kroeck, & Sivasubramaniam, 1996). Leaders can influence employees’ growth and achievement by allowing them to share knowledge and experiences among each other (Oldham & Cummings, 1996). Leaders’ behaviors can influence the atmosphere within a unit, but also within teams. At the group or team level of analysis, team leaders play a significant role in the way team processes and outcomes develop (Zaccaro, Rittman & Marks, 2001; Gil, Rico, Alcover & Barrasa, 2005). Numerous studies have reported that team leaders can significantly contribute to individual, group and organizational performance (Lowe, Kroeck, & Sivasubramaniam, 1996; Lovelace, Shapiro & Weingart, 2001; Jung & Sosik, 2002). Team leaders can directly influence the way members formulate perceptions, as well as set examples for how to behave (Edmondson, Kramer & Cook, 2004).

Because of this influence of leadership on organizational outcomes and team functioning, we argue that leadership is an important factor to study alongside team climate and ambidexterity. As pointed out in the introduction already, research on leadership and

(21)

ambidexterity is available, but limited. Nemanich and Vera (2009) argue that transformational leadership, which is a leadership style that consists (1) charisma, (2) inspirational motivation, (3) intellectual stimulation and (4) individualized consideration, positively influences ambidexterity at the team level (Bass, 1990; Nemanich & Vera, 2009). In contrast, a research by Jansen et al. (2008) on senior team attributes and organizational ambidexterity, showed that transformational leadership moderates this relationship (Jansen et al., 2008). Another study investigated the role of paradoxical leadership in combination with individual ambidexterity. It was concluded that paradoxical leaders – leaders who combine strong managerial support with high performance expectations – directly influence individuals’ ambidextrous behavior (Kauppila & Tempelaar, 2016). Yet another study by Jansen et al. (2016) found support for the hypothesis that supportive leadership moderates the relationship between team cohesion, team efficacy and team level ambidexterity, instead of a having a direct effect on ambidexterity.

Although these studies differ in the levels of analysis and leadership styles, they share the notion that leaders play an important role in enabling a supportive context for ambidexterity. The lack of consensus on the role of leadership, however, makes empirical research on this topic of high importance. More specifically, the absence of empirical studies on the role of leadership alongside team level antecedents to individual ambidexterity, indicates that further research is needed.

As the leadership literature distinguishes many types of leadership styles, we choose to focus on the role of supportive leadership because this one is most likely to occur at the team level. Available literature shows that there are many types and conceptualizations of leadership styles. Examples are transformational, transactional and paradoxical leadership styles (Hartog, Muijen & Koopman, 1997; Rosing, Frese & Bausch, 2011; Kauppila & Tempelaar, 2016). We argue that it is important to incorporate a supportive leadership style specifically, as we focus on the team leaders instead and not the managerial levels. As team leaders are more engaged

(22)

with regular employees, their supportive behaviours might assist individals in teams to further develop themselves with regard to learning activities. Supportive leadership thus “covers the encouragement of individual initiatives, clarification of individual responsibilities, provision of clear and complete performance evaluation feedback, a strong task orientation, emphasis on quality group relationships and trust in organizational members” (Van de Ven & Chu, 1989, p. 340). In the next chapter the relationships and supporting arguments with regard to team climate factors, individual ambidexterity and supportive leadership are discussed more thoroughly.

(23)

3. Theoretical framework & Hypotheses

In the previous chapter it became clear that team climate could possibly influence individuals’ ambidextrous behavior. We argue that specifically the climate for trust, climate for excellence and climate for affiliation are relevant climate indicators to study in combination with individual ambidexterity. Furthermore, we propose that supportive leadership may moderate the relationship between team climate and individual ambidexterity. This chapter is dedicated to the hypotheses and supporting arguments.

3.1 Team climate factors and individual ambidexterity

Climate for trust. The climate for trust refers to organizational members’ perception

of how trusting the climate is. Based on Costa and Anderson’s measure of trust between team members, we look at trust as the individual’s perceived trustworthiness of the other team members, which is referred to as ‘the extent to which individuals expect others to be and to behave according to their claims’ (Costa & Anderson, 2011, p. 125).

Team climates that are perceived as more trustworthy, will more likely increase the extent of knowledge sharing among team members. Exchanging information and experiences regarding work activities helps individuals to reflect upon each other’s work (Mohammed & Dumville, 2001). The support and encouragement that results from trusting climates may benefit individuals in coping with explorative and exploitative demands, as members of such climates are more commited to difficult task goals (Klein & Mulvey, 1995). This view is consistent with Birkinshaw and Gibson’s article (2004), in which they argue that individuals in a supportive organization context characterized by trust, will engage in both exploration-oriented and exploitation-exploration-oriented actions. In a trusting climate, individuals’ behavior will enable more cooperation, learning and monitoring between team members (Ghoshal & Barlett, 2004; Costa & Anderson, 2011). Thus, teams that have a trusting climate could possibly have

(24)

higher levels of individual ambidexterity. Conversely, teams characterized by low climates for trust will likely have a negative impact on individuals’ ability to become more ambidextrous. According to Edmondson, Bohmer and Pisano (2001), the less trusting the climate, the more it will inhibit learning. An organizational climate that is characterized by distrust and that is perceived as unfair, will not stimulate team members to share and exchange knowledge (Kim & Mauborgne, 1998). This is possibly a lost opportunity to gain valuable knowledge that facilitates ambidextrous behavior. Therefore, we expect:

Hypothesis 1a: Team climate for trust is positively associated with individual ambidexterity.

Climate for excellence. Another team climate factor that we consider in this study is

the climate for excellence. Climate for excellence is a team-level concept, which is defined as:

“The shared group norms about excellence of quality of task performance and each team member’s commitment to high quality standards, critical appraisals, monitoring, and clear performance criteria within the team” (Eisenbeiss, Knippenberg & Boerner, 2008, p. 1348).

Under high climates for excellence, team members will be more likely to build on each other’s ideas and to monitor each other, since they are highly concerned about the excellence of task performance (Eisenbeiss, Knippenberg & Boerner, 2008). This can be explained by their shared norms regarding the quality of these ideas. In high climates for excellence, individuals will express the clear criteria that they have in mind in order to achieve excellence and will consequently share their novel and useful ideas with other team members (Anderson & West, 1998). The general concern about excellence of task performance in turn may also result in

(25)

more feedback from the team members on proposed ideas and suggestion by others. This will eventually result in careful selection of ideas and lead to ideas that are most promising and have highest quality (Anderson & West, 1998).

Whenever there is the demand for individual ambidexterity, we expect that the shared norms about task excellence and other norms associated with a high climate for excellence, will stimulate individuals more to provide feedback and share ideas about the opposing learning demands associated with ambidexterity. Thus, under a high climate for excellence, it is more likely that individuals engage in explorative and exploitative activities in a balanced manner.

Hypothesis 1b: Team climate for excellence is positively associated with individual ambidexterity.

Climate for affiliation. Finally, the last team climate factor that we consider is the

affiliation climate. This climate factor measures the extent to which team members cooperate and feel connected on a social level. Bock et al. (2005) define it as: “the perception of a sense of togetherness among an organization's members, reflects the caring and pro-social behavior critical to inducing an organization's members to help one another.” (Bock et al., 2005, p. 94). It is important for team members to perceive a sense of togetherness as this will likely result in more motivation to exchange complex information (Wong, 2004). Team members that experience a sense of togetherness are more likely to negotiate, adjust, and to integrate conflicting agendas and learning attitudes (Jansen et al. 2016). Due to this team environment, it is more likely that individuals learn from the other team members. A general sense of awareness about their coworkers’ different approaches to ambidexterity can be helpful in enabling them to achieve ambidexterity by themselves. Furthermore, a climate characterized by pro-social norms may also provide norms for constructive feedback in explorative and

(26)

exploitative learning (Wong, 2004). We expect that a team with a high affiliation climate engenders its team members to actively engage, cooperate and reflect on proposed problems associated with explorative and exploitative activities. Therefore, we hypothesize:

Hypothesis 1c: Team climate for affiliation is positively associated with individual ambidexterity.

3.2 The role of supportive leadership

Supportive team leaders may further influence the relationship between individual ambidexterity and team climate by participating and facilitating teams that are characterized by climates for trust, excellence and affiliation to resolve conflicts and contradictory demands that could arise from such climates (Jansen, George, Van den Bosch & Volberda, 2008). By encouraging individual initiatives, clarifying responsibilities, emphasizing relationships and encouraging feedback exchange, supportive leaders can direct team members to resolve conflicts and help them overcoming the paradoxical demands associated with ambidexterity. Supportive leaders may also induce team members to freely discuss and evaluate each other’s ideas. By stimulating collaborative behavior and responding to team members’ possibly conflicting agendas, team leaders can increase the impact of positive team climates on individual ambidexterity.

In a climate for trust, highly supportive leaders could possibly stimulate individuals more to critically reflect on one another’s work by creating a comforting culture (Anderson & West, 1998). This may in turn make individuals more conscious about their ambidextrous behaviors. Therefore, we propose that supportive leadership behavior could modify the relationship between the climate for trust and individual ambidexterity. Whenever there is the presence of few trust among team members, team leaders could set the salient example for how to behave by demonstrating trust in single team members and emphasizing the group

(27)

relationships (Van de Ven & Chu, 1989; Carmeli, Gelbard & Gefen, 2010). This exemplary role, in turn, could reinstate perceptions of trust among team members. Supportive leadership thus could possibly amplify the positive relationship between the climate for trust and individual ambidexterity. As the effect of a high climate for trust on individual ambidexterity can be amplified under situations in which team leaders actively support the team to further share their ideas, we expect:

Hypothesis 2a: Supportive leadership enhances the positive effect of team climate for trust on individual ambidexterity.

In high climates for excellence, the pressure to come up with innovative and high quality solutions might be high. Although this could benefit individual ambidexterity on one hand, the high norms of excellence may also result in not sharing ideas, because of hesitation and insecurity. Supportive leaders can foster open communication, show caring behaviors (Jansen et al., 2016) and support team members that might want to share task-related problems, but are hesitant to do so. Furthermore, supportive leaders could encourage team members to further express diverging viewpoints without feelings of threat or challenge (Lovelace, Shapiro & Weingart, 2001). Without the presence of such effective leaders, discussing task-related problems regarding contradictory tasks such as exploration and exploitation, might result faster in extensive and dysfunctional conflict (Lovelace, Shapiro & Weingart, 2001). Therefore, we hypothesize:

Hypothesis 2b: Supportive leadership enhances the positive effect of team climate for excellence on individual ambidexterity.

(28)

Leadership literature provides extensive studies on the role of leadership and how team members experience senses of belonging and togetherness. Specifically, leaders that emphasize harmonious relationships, may further support teams with a high climate for affiliation to express their thoughts, contribute, cooperate and share their knowledge even further (Jansen et al., 2016). The facilitative aspect of supportive leadership in team communication and collaboration can in turn influence how individuals deal with ambidexterity. Furthermore, we argue that supportive leadership helps individuals within such teams to take specific actions and that it enhances their ability to share and combine explorative and exploitative activities (Jansen et al., 2016). By encouraging team members to discuss and, as a consequence, re-evaluate ideas, supportive leaders can eliminate possible threats of groupthink. They can do so by stimulating team members to reflect on each other’s opposing views, and as a result, reach consensus regarding the conflicting goals that ambidexterity entails (Barrick, Bradley, Kristof-Brown & Colbert, 2007; Stewart, Courtright & Barrick, 2012). Thus, it is expected that supportive leadership amplifies the positive relationship between team climate for affiliation and individual ambidexterity. Taken together, we propose:

Hypothesis 2c: Supportive leadership enhances the positive effect of team climate for affiliation on individual ambidexterity.

(29)

3.3 Conceptual model

The hypotheses and proposed relationships between the variables lead to the following conceptual model:

(30)

4. Research design & Methods

The following paragraphs present the main information about the research approach of this study. It does so by providing insights into the data collection method, as well as the sample of this study. Lastly, it elaborates on the measurements of the independent variables, dependent variables, moderator and control variables.

4.1 Data collection and Sample

Since this research seeks to examine possible relationships between team level antecedents and individual ambidexterity, one of the criteria was to find a large group of respondents that work in groups or teams. Another factor that we considered to be important, was that there should be at least one team leader in the picture, in order to assess the individuals’ perception about their leaders’ supportive attitude. Finally, the last criterion related to the task characteristics of the employees, since it was of high importance that their tasks included both exploration and exploitation related activities. Therefore, companies were needed that did not solely focus on either exploration or exploitation - such as efficiency focused companies like factories - but instead, companies that allowed their employees to engage in both activities.

An industry that fits the criteria set above, is the financial services industry. This industry can be characterized as knowledge intensive, since technologies, competition and customer demands are continuously changing (Flier et al., 2001). On the other hand, the companies in this industry also need to focus on the efficiency and optimization of their processes, resulting in a focus on exploitation. Therefore, we consider the financial services industry as one that meets the criteria for this research.

Respondents were carefully selected based on convenience sampling. This resulted in a relatively heterogeneous sample of respondents, representing employees of several companies

(31)

in the financial services industry. The respondents were questioned by means of an online survey, designed and distributed via Qualtrics. The survey can be found in Appendix D. The time frame for the data collection period was two weeks. In total, 197 respondents clicked on the link of the survey, of which 132 respondents actually started and dropped out at some point. In the end, we obtained fully completed surveys from 109 respondents. The final sample included respondents from multiple types of companies, categorized into: banking (53.2%), consulting (20.2%), insurance (5.5%) and others (21.1%). The respondents were mostly employed at large-sized enterprises (73.4%), medium-sized enterprises (9.2%) and small-sized enterprises (17.4%).

As this study used individuals’ responses to measure team-level variables, as well as leadership levels, a bias that needed to be checked for is the common method bias. We performed Harman’s single factor test to check whether this bias influenced our relationships. Table III in Appendix A presents the findings. These indicated that the first factor did account for 45% of the variance, which indicates a good result as it is lower than the cut off point of 50% (Podsakoff et al. 2003).

4.2 Measurement and Validation of Constructs

This section reflects upon the the validation and measurements of the main variables: individual ambidexterity, team climate, supportive leadership and the control variables. In order to ensure well-tested and validated scales, we adopted existing scales from previous research, which we verified through reliability and exploratory factor analysis. The items and scales can be found in Appendix C.

(32)

4.2.1 Dependent variable: Individual Ambidexterity

In accordance with previous research (Birkinshaw & Gibson, 2004; He & Wong, 2004), we use two steps to develop a measure for individual ambidexterity. First, individuals provide information regarding their exploration activities. We adopted a five-item scale for individuals’ exploration activities developed by Mom et al. (2007, 2009). This scale captures the extent to which individuals depart from existing knowledge and engage in explorative activities. The reliability for this scale based on the Cronbach’s alpha is α = 0.82. Sample items are ‘Searching for new possibilities with respect to products/ services, processes or markets’ and ‘Focusing on strong renewal of products/services or processes.’.

A second six-item scale measures the exploitation dimension of individual ambidexterity and captures the extent to which individuals build on available knowledge. This scale was also adopted from Mom et al.’s (2007, 2009) study and has a Cronbach’s alpha of α = 0.76. Sample items are ‘Activities of which a lot of experience has been accumulated by yourself’ and ‘Activities of which it is clear to you how to conduct them’. Both scales were measured by a seven-point Likert scale ranging from (1 = to a very small extent to 7 = to a very large extent).

In order to assess the construct validity of individual ambidexterity and its exploration and exploitation dimension, we performed an exploratory factor analysis. Initially, due to one ambiguous item of the Exploration dimension, the results did not replicate the intended two-factor structure. However, after removal of this item, both the reliability and two-factor analysis yielded better results, as all items loaded cleanly on their intended factors. All factor loadings were above .30 with cross loadings below .27. Also, the analysis showed that only after the removal of this item the factors had eigenvalues greater than one. Therefore, we deemed the removal of this item necessary in order to proceed with further analysis. The results are presented in Table IV in Appendix A.

(33)

Lastly, following previous literature on individual ambidexterity, we multiplied the scores of the exploration and exploitation dimension. Previous scholars have constructed ambidexterity in multiple ways, namely subtracting the two dimensions (He & Wong, 2004), multiplying the dimensions (Gibson & Birkinshaw, 2004; Mom et al., 2007) and summation (Lubatkin et al., 2006). Since this empirical study looks into individual ambidexterity specifically, we decided to multiply the exploration and exploitation dimension, as this approach is in accordance with most available studies on individual ambidexterity and follows the same pathway as Mom et al. (2009), who originally developed the scales.

4.2.2 Independent variables: Team Climate

In this study we focus on the climate for trust, climate for excellence and climate for affiliation. These team level factors are adopted from prior research and have been used widely in the literature.

Climate for trust

To measure the climate for trust in organizational teams, this study adopted the six-item scale of Costa and Anderson (2011), who based their items on Cummings and Bromiley’s (1996) seven-item scale to measure team members’ perceptions of other members’ trustworthiness. An example of an item states ‘In this team people can rely on each other’ and ‘In this team people will keep their word’. All items are measured using a seven-point Likert scale (1 = strongly disagree to 7 = strongly agree). The Cronbach’s alpha for this scale is α = 0.85.

Climate for excellence

To measure the climate for excellence, we adopted a seven-item scale from Anderson and West’s (1998) Team Climate Inventory. The scale refers to the the extent to which team

(34)

members interacted in order to promote excellence in the team's work (Burningham and West, 1995). An example of an item of this scale is: ‘Do your team colleagues provide useful ideas and practical help to enable you to do the job to the best of your ability?’. The seven-point Likert scale ranged from l = to a very little extent to 7= to a very great extent. The Cronbach’s alpha for this scale is α = 0.91.

Climate for affiliation

The affiliation climate can be measured using four items adopted from Bock et al. (2005). This measurement measures the perception of togetherness among team members. A climate that is characterized by pro-social norms, can be seen as affiliation climate (Bock et al., 2005). A sample item is: ‘Members in my department keep close ties with each other’. Respondents were instructed to rate this statement using a five-point Likert scale, ranging from 1 = strongly disagree to 5 = strongly agree. The Cronbach’s alpha for this scale is α = 0.81.

In order to assess the goodness of the scales of the team climate dimensions, we performed an exploratory factor analysis. Initially, due to two ambiguous items and high cross loadings of one item of the trust dimension and one of the affiliation dimension, the results did not replicate the intended three-factor structure at first. However, after removal of these two items, both the reliability and factor analysis yielded better results, as all items loaded clearly on their intended factors. All factor loadings were above .50 with cross loadings below .24. The eigenvalues of these factors were greater than one. The results are presented in Table V in Appendix A.

4.2.3 Moderating variable: Supportive Leadership

(35)

involvement within teams, but other researchers also used this measure for rating the degree of team leader support (Wickramasinghe & Widyaratne, 2012). An example of an item is: ‘The leader of my work team encourages a climate of trust and co-operation among the

members of the team’. Respondents were requested to base their answers on a seven-point Likert scale ranging from 1 = strongly disagree to 7 = strongly agree. The Cronbach’s alpha for this scale is α = 0.83.

4.2.4 Control variables

In this study we include relevant control variables to diminish the possibility of confounding effects. The control variables that were added in this research are: (1) gender, (2) age, (3) education, (4) firm size, (5) team size, (6) tenure in current function, (7) tenure in organization and (8) supervisory role. The following paragraph provides arguments for these choices.

The first variable that needs to be controlled for is gender. Prior research shows that men and women approach activities differently. For example, men are often associated with risk-taking, whilst women tend to be more precise and focused (Byrnes, Miller & Schafer, 1999). In a sense, risk-taking can be interpreted as explorative behavior, whereas precision more resembles exploitation. As these differences may account for different levels of ambidexterity between men and women, we include gender as a control variable.

Second, prior research demonstrated that older people tend to focus more on efficiency and less on exploring new areas (Nystrom, Ramamurthy & Wilson, 2002). Therefore, we include employee age to diminish a possible confounding effect.

Third, we include education as a control, since it might influence individuals’ cognitive ability to switch between exploration and exploitation. Furthermore, since most of the employees work in knowledge intensive industries, we expect that these employees should

(36)

continuously be able to switch between the opposing demands of exploration and exploitation. Their educational level might influence this abilty.

Fourth, we control for firm size. Larger organizations may have more available resources, which could enable individuals to engage in ambidexterity. It is likely that larger organizations are more dynamic, and therefore demand more ambidextrous behavior from employees.

Fifth, as this empirical study incorporates the team context, we control for team size. Team members in larger teams may have more opportunities to shirk, or to stay focused on either exploration or exploitation related activites, without their coworkers noticing. Their behavior may be less observed in comparison to team members in smaller teams.

Furthermore, we include the tenure in organization and the tenure inside employee function as a control. As employees experience in the organization or function increases, this leads to more specialization, which can cause a higher focus on one of the two dimensions of ambidexterity. Mom et al., for example demonstrated that tenure is associated with specialization (Mom et al., 2007). More specialization can lead to a higher focus on exploration or exploitation, depending on the profession.

Lastly, we asked the respondents whether or not they had a supervisory role in their team. Hierarchical level is proven to have a positive relationship with employee ambidexterity (Floyd & Lane, 2000). Another reason to include this control, is because team leaders might perceive their climate differently than the regular team members. Due to these factors the results for these individuals might differ from the others.

(37)

5. Results

In this section the results of this study are presented. First, the descriptive statistics and correlations of the variables used in this study are presented and will be analyzed by Table I. Second, Table II presents the results of the hierarchical regression analysis for individual ambidexterity. After reporting on the effects, we illustrate the steps that were taken for the post-hoc analysis.

5.1 Correlation analysis

The correlation analysis shows no extremely high values, thus implying less concern for multicollinearity issues. There is a significant positive effect between the climate for excellence and individual ambidexterity, r = .20, p < .05. Also, the team climate variables are significantly and positively correlated to each other. Furthermore, it is noteworthy that the more supportive the leader, the higher the team climate for trust, r =.29, p < .01, excellence, r =.50, p < .01 and affiliation, r =.50, p < .01. Also, this sample shows that gender has a negative and significant relationship with indivudual ambidexterity, r = -.20, p < 0.05. This means that the male participants in this study reported higher levels of individual ambidexterity compared to females. We also observe that dducation level has a positive and significant correlation with all the three team climate variables. Furthermore, the respondents with a supervisory role seem to have higher levels of individual ambidexterity, r = .25, p < 0.01. The analyses also show that these leaders are often older males who have been working in their organizations and current functions for a longer period of time.

(38)

Table I - Means, standard deviations, and correlations

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

1. Individual Ambidexterity 22.72 7.52 -

2. Climate for Trust 5.16 1.11 .11 (.85)

3. Climate for Excellence 4.85 1.12 .20* .65** (.91)

4. Climate for Affiliation 3.71 .79 .16 .54** .59** (.82)

5. Supportive Leadership 5.27 1.25 -.03 .29** .50** .50** (.83)

6. Gender (1= female; 0= male) .36 .48 -.20* .03 .10 -.00 .02 -

7. Age 2.43 1.21 .14 .08 -.18 -.13 -.03 -.20* -

8. Educational level 2.38 .64 -.07 .24* .20* .21* .13 .04 -.14 -

9. Tenure in organization 10.38 10.79 .09 .16 -.05 -.033 -.092 -.11 .68** -.28** -

10. Tenure in function 5.05 6.69 .17 .10 -.03 .01 .10 -.15 .63** -.17 .52** -

11. Supervisory Role (1 = yes; 0 = no) .31 .47 .25** .05 .02 .11 .10 -.30** .32** -.03 .25** .34** -

12. Team size 22.48 48.83 .16 .08 .002 .06 -.08 -.15 .15 .13 .21* .10 .19 -

13. Firm size 2.56 .78 .02 .00 .054 -.06 -.14 -.09 .02 .19* .13 -.29** -.18 .13 -

Notes: Exploration α=0.82, Exploitation α=0.76. Individual Ambidexterity = Exploration * Exploitation. * Correlation is significant at the 0.05 level (2-tailed).

(39)

5.2 Hierarchical regression results

Table II contains the results of the hierarchical regression analysis. The first model includes the effects of the control variables. The second model presents the main effects of the independent variables, which are the climate for trust, climate for excellence and climate for affiliation. Model 3 adds the interaction effects of the moderator supportive leadership on individual ambidexterity.

To examine multicollinearity, we calculated variance inflation factors (VIF) for each of the regression equations. All of the VIF scores were well below the rule-of-thumb cutoff of 10 (Field, 2009), with the highest score being 2.49. Furthermore, the tolerance levels were all higher than 0.10, indicating few concern for multicollinearity.

As presented in model 2, the climate for trust does not have a significant relationship with individual ambidexterity (β = -.14, p > .05), thus, Hypothesis 1a is not supported. Hypothesis 1b on the other hand, which suggested that the climate for excellence is positively associated with individual ambidexterity, is supported (β = .36, p < .05). Accordingly, Hypothesis 1c, which posited that climate for affiliation would contribute to individual ambidexterity, was not supported (β = .13, p > .05).

In addition to the direct effects of team climate factors on individual ambidexterity, we argued that the impact of this relationship will be enhanced when team leaders demonstrate supportive leadership. The proposed interaction effect between team climate for trust and supportive leadership is significantly related to individual ambidexterity (β = -.17, p < .05). However, only in the opposite direction, which contradicts our expectation. The second interaction effect between team climate for excellence and supportive leadership is positive, but not significant (β = .26, p > .05), thus Hypothesis 2b is not supported. Consistent with

(40)

Hypothesis 2b, the interaction effect between climate for affiliation and supportive leadership (Hypothesis 2c) is slightly negative, but also non-significant, hence not supported.

Contrary to our expectations, Hypothesis 2a, which suggests a positive moderating effect of supportive leadership on the relationship between climate for trust and individual ambidexterity, is negative. As the interaction proved significant, we conducted a simple slope analysis by using the process macro for SPSS and by following the procedures described and developed by Hayes (2013). Before creating the plots, the variables were mean-centered. Figure 2 shows the results for low, medium and high values of supportive leadership. The values are based on the mean and 1 SD above and below the mean (Aiken & West, 1991). As can be seen in Figure 2, the slopes for low, medium and high levels of supportive leadership are negative. In team with high levels of supportive leadership, the team members are less likely to achieve individual ambidexterity. Conversely, teams with lower levels of supportive leadership are more conducive to achieve individual ambidexterity. This indicates that the effect of team climate for trust on individual ambidexterity is more negative when team leaders demonstrate more supportive attitudes.

Individual ambidexterity and team climate : the moderating role of supportive leadership