External network ties and individual ambidexterity : the moderating effects of collectivistic cultures

(1)

External Network Ties and Individual Ambidexterity: The

Moderating Effects of Collectivistic Cultures

Master Thesis

Student:

Danielle Garcia de Albuquerque Moreira

Student № 11373997

MSc. Business Administration (MSc BA), Strategy track

University of Amsterdam, Faculty of Economics and Business

Supervisor: MSc. B. Silveira Barbosa Correia Lima

University of Amsterdam, Amsterdam Business School

(2)

Statement of originality

This document is written by Student Danielle Garcia de Albuquerque Moreira

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.

(3)

Abstract

Further research in the individual level of ambidexterity has been required. Managing exploration and exploitation is not an easy task, therefore this study aims to answer what is the effect of employees’ interaction with external network ties on their ambidextrous behaviour and what is the moderator effect of culture in this relationship. This research has been conducted in different countries, mainly in Brazil and the Netherlands, to analyse whether the individualism or collectivism of the country would moderate the effect of informality and heterogeneity of the external network in individuals’ ambidexterity. The findings suggest that the effects of informality of external contacts is not significant for employees’ ambidexterity. However, the effect of heterogeneity of external ties in individuals’ ambidexterity was shown as significant, as well as the moderator effect of culture collectivism in the relationship between heterogeneity of external ties and individual ambidexterity.

Key words: Ambidexterity, Network, External Network Ties, Informality, Heterogeneity,

Employees, Exploration, Exploitation, Knowledge, Innovation, Culture, Collectivism, Individualism

(5)

Introduction

Ambidexterity has been a trendy topic that has sparkled more research in the last years. It can be evidenced at organizational level, business unit/team level, down to the individual employees’ level (Raisch et al., 2009; Turner et al., 2013). Some researchers (e.g. Mom et al., 2009, O’Reilly and Tushman 2004), however, did research individual level, although mostly focused on managers. Ambidexterity in the individual level is considered as the extent to which individual combine the exploration of new competence areas with the exploitation of existing competences in their work role (Bledow et al., 2009; Mom et al., 2009).

Ambidexterity is important since one should engage in both, exploration and exploitation. Due to the fact that exploring to the exclusion of exploitation will probably lead to suffer the costs of experimentation without gaining many of the benefits. While engage in exploitation to the exclusion of exploration, likely will lead to be trapped in suboptimal stable equilibria (March, 1991). In addition, ambidextrous firms need ambidextrous managers and teams (O’Reilly & Tushman 2004). However, although individuals are key for contextual ambidexterity, they have been neglected in ambidexterity researches, that were mostly in organizational level and when they were in individual level focused on managers only.

Previous researches have analysed what can influence ambidexterity. They were mostly focused on individual characteristics (Raisch et al., 2009), but there are other aspects that are also antecedents of ambidexterity. Such as refining and renewing knowledge, skills and expertise (Floyd & Lane, 2000; Sheremata, 2000; Hansen et al., 2001). Similarly, exploration is related to sourcing of new knowledge and exploitation is related to sourcing of existing knowledge (Rothaermel & Alexandre, 2009).

Other aspects that were previously studied was organizational level ambidexterity. Most of the literature on the topic focus on the organizational level, where ambidexterity is referred as a firm’s or business unit’s ability to combine exploration and exploitation related activities (e.g.

(6)

Gibson & Birkinshaw 2004, O’Reilly & Tushman 2004). Although, O’Reilly and Tushman (2004, p. 81) conclude that “one of the most important lessons is that ambidextrous organizations need ambidextrous senior teams and managers”.

The individual level ambidexterity was researched mostly among managers. Mom et al. (2009) added that ambidexterity is conceptualised a multidimensional construct that captures the extent to which employees pursue both explorative and exploitative activities in their work tasks (Mom et al., 2009). Individual’s networks play a crucial role in shaping individual-level ambidexterity, but it has been under-researched (Rogan & Mors, 2014). Most of the researches considered network structure, instead of its qualitative characteristics. In the culture aspect, Yang et al. (2015) researched whether collectivism of an organization would affect its ambidexterity regarding innovations. But no research was found regarding the effects that collectivism would have in the relationship between external network of employees and individual ambidexterity.

Therefore, this study aims to find what is the effect of employees’ interaction with external network ties on their ambidextrous behaviour. Besides that, by analysing the collectivism of countries, this research will provide the moderating effect of culture in the relationship between external network ties and individual ambidexterity. The data for this study was gathered to test its arguments via survey with 122 employees in Europe and America in different hierarchical levels. Consequently, the date was analysed through a multiple regression.

Gibson and Birkinshaw (2004) argue that ambidexterity manifests itself in the specific actions of individuals throughout the organization and in their daily work, individuals often face choices of how they should spend their time, therefore it is important to have further research on the individual level in general, instead of exclusively on managers. Additionally, it is still missing in the literature what is the influence that the interactions with external actors,

(7)

outside firm’s boundaries, such as suppliers and clients for instance, can increase the level of ambidexterity in employees in general. Furthermore, the moderating effect that culture can have in this relationship has not been addressed considering external network and individual ambidextrous behaviour.

With the findings of this research we will know the effects that interactions in the external network have in the employees’ ambidexterity, as well as how the country they work can affect this relationship. The findings of this study are relevant for most of the direct stakeholders of companies, such as shareholders, managers, employees and clients. Due to its importance, individual-level ambidexterity is necessary for combining and gaining synergies between explorative and exploitative activities at higher organizational levels as well (Adler et al., 1999; Miron-Spektor et al., 2011). For managers, this study will be relevant because ambidextrous behaviour is associated with higher performance in dynamically changing tasks (Good and Michel, 2013), thus if they manage employees or has dynamic tasks it will be important to know what in the external network can improve this behaviour. This study also benefits employees, as well as the fact that ambidexterity is an important way to enrich the work for employees in unexciting jobs because the combination of explorative and exploitative tasks makes the work more meaningful (Parker, 2014).

This study will also benefit theorists, since the ones in the design and motivation field have noted that contemporary work roles require individuals to attend to increasingly diverse tasks (Griffin et al., 2007) and to take personal initiative in developing and pursuing their work goals (Frese and Fay, 2001). These researchers have asked for job roles that involves ambidextrous behaviours to make jobs more interesting and motivating for employees (Adler et al., 1999).

This paper will first provide a review of the literature, what has been found and discussed so far regarding individual level ambidexterity and the influence of network ties as

(8)

well as the moderating effect of culture. In addition, the research question and conceptual model of this study will be provided. Secondly, the methodology of this research will be explained, followed by its findings, results, discussion and conclusion. Consequently, the limitations and suggestions for future research will be presented.

Literature Review

The following section presents and discusses the most relevant insights from existing literature on ambidexterity and external network ties, as well as present the research questions of this research. First, concepts of ambidexterity, external network ties and collectivist and individualist cultures will be introduced. Furthermore, findings of previous studies in which these three concepts are linked will be presented. Then, I address the research gap and research question. Finally, I present the benefits and importance of further research in this topic.

Individual ambidexterity

Ambidexterity is a trendy topic that has been mostly researched at the firm level, where ambidexterity is referred as a firm’s or business unit’s ability to combine exploration and exploitation related activities (e.g. Gibson & Birkinshaw, 2004; O’Reilly & Tushman, 2004). However, O’Reilly and Tushman (2004, p. 81) conclude that “one of the most important lessons is that ambidextrous organizations need ambidextrous senior teams and managers”. Hence, on individual level, ambidexterity is conceptualised a multidimensional construct that captures the extent to which employees pursue both explorative and exploitative activities in their work tasks (Mom et al., 2009). Employees need to behave towards both exploration and exploitation in order to be ambidextrous. Therefore, the concept of exploration and exploration will be demonstrated in this study.

Previous studies provided some characteristics of ambidexterity that inspire further research. Specifically, two aspects will be addressed in this study.

(9)

Firstly, ambidextrous individuals are multitaskers (Mom et al., 2009; Birkinshaw & Gibson, 2004, Floyd & Lane, 2000). Regarding this point, March (1991) states that one should engage in both, exploration and exploitation. This is due to the fact that, exploring to the exclusion of exploitation will probably lead to suffer the costs of experimentation without gaining many of the benefits, while engage in exploitation to the exclusion of exploration, likely will lead to be trapped in suboptimal stable equilibria. Returns from exploration are less certain and more remote in time than the returns from exploitation, consequently individuals tend to prioritize exploitation over exploration. Therefore, employees should engage in both exploration and exploitation activities within a certain period of time (Floyd & Lane, 2000) to be ambidextrous. Even though the trade-off between exploration and exploitation involves conflicts between short-run and long-run concerns (March, 1991). Lavie, Stettner, and Tushman (2010) adds that “Ambidexterity resolves the tension between exploration and exploitation by suggesting these activities are maintained simultaneously” (p. 129). The time restrain is perceived differently in other studies. Thus, defining ambidexterity as individual behavioural orientation toward combining exploration and exploitation related activities within a certain period of time (cf. Gibson & Birkinshaw, 2004, p. 210; O’Reilly & Tushman, 2004, p. 81; Tushman & O’Reilly, 1996, p.24). This study will follow this concept of individual ambidexterity regarding multitask situations. In addition, other authors argue that individual ambidexterity is defined as the extent to which individuals combine exploration of new competences with the exploitation of existing competences in their work role (Bledow, Frese, Anderson, Erez & Farr, 2009; Kauppila & Tempelaar, 2016).

The second aspect considered in this study is that ambidexterity is knowledge related. Since both exploration and exploitation are learning-related activities (Gupta et al., 2006). According to Rothaermel and Alexandre (2009), exploration is related to sourcing of new knowledge or technology and exploitation is related to sourcing of existing knowledge or

(10)

technology. Consequently, ambidextrous individuals refine and renew their knowledge, skills and expertise (Floyd & Lane, 2000; Sheremata, 2000; Hansen et al., 2001). Miller et al. (2006) add arguments about the role of direct interpersonal learning. Where tacit knowledge cannot be transmitted through an organizational code, but only by individuals sharing their knowledge directly with others (Miller, Zhao & Calantone, 2006). Therefore, it is important to recognise the value of individual learning and how interpersonal interactions are critical for knowledge transfer. For instance, exploitation implies behaviours such as refinement of existing assets and knowledge (Benner & Tushman, 2003), while exploration draw in behaviours like gaining broader knowledge and advancing new opportunities (Jansen, George, Van den Bosch & Volberda, 2008). Any factor that increases exploration or exploitation, considering that this increase is not at the expense of decreasing the other, can enhance individual ambidexterity. Actually, factors that increase exploration without lowering an existing high level of exploitation are particularly important antecedents of ambidexterity (Kaupilla & Tempelaar, 2016). There are studies that explicitly embrace the idea that both exploration and exploitation are associated with learning and innovation, despite of different types (Baum et al., 2000; Benner & Tushman, 2002; He & Wong, 2004). Gupta et al. (2006) also affirms that both exploration and exploitation are learning-related activities, where explorative learning increases the breadth of knowledge, creating prospects for radical changes, and exploitative learning increases the depth of knowledge, which usually leads to incremental development and enhanced reliability (Benner & Tushman, 2003). Therefore, ambidextrous employees engage in both reliability-enhancing and variety-increasing learning activities (Holmqvist, 2004; McGrath, 2001).

Studies in the individual level of ambidexterity are still scarce (Mom et al., 2009). Most of the existing studies in the individual level have focused only on the managers. Contemplating that managers who allocate resources between exploration and exploitation are

(11)

also considered to be acting ambidextrously (O’Reilly & Tushman 2004, p. 81). Managers are likely to have more resources, to allocate ambidextrously, than lower levels employees. Gibson and Birkinshaw (2004) argue that although ambidexterity is a characteristic of a business unit, it manifests itself in the specific actions of individuals throughout the organization and in their daily work, individuals often face choices of how they should spend their time— for instance, should they continue to focus on an existing customer account to meet quota, or should they nurture a new customer who has a slightly different need? Thus, individuals in general, not only managers, are important drivers for ambidexterity, also due to that they contend with the dynamic context (Good & Michel, 2013).

Considering interorganizational learning and external industry forces such as uncertainty and turbulence, knowledge acquired with individuals’ external ties can lead them to be more exploitative (Beckman, Haunschild, & Phillips, 2004), explorative (Rowley, Behrens, & Krackhardt, 2000) or both (Koza & Lewin, 1998), thus being ambidextrous. Also, if they increase one not at the expense of decreasing the other, it can enhance their ambidexterity (Kaupilla et al., 2016). Therefore, interpersonal interactions that employees have in their external network may lead to effects in their ambidexterity.

Individual Ambidexterity and External Network Ties

A network is “any collection of actors that pursue repeated, enduring exchange relations with one another and, at the same time, lack a legitimate organizational authority to arbitrate and resolve disputes that may arise during the exchange” (Podolny & Page, 1998, p. 59). Social network analysis defines networks as sets of ties linking several individuals. Ties or contacts may be of different kinds, formal or informal, frequent or infrequent, affect-laden or purely utilitarian (Granovetter, 1973).

(12)

The use of network theory in the ambidexterity field is not new (e.g., Atuahene-Gima and Murray 2007, Lavie and Rosenkopf 2006, Lazer and Friedman 2007, Tiwana 2008), but applications in the individual level are very limited (Mom et al., 2009, Raisch & Birkinshaw 2008). Most of the current literature focuses on ambidexterity at the business unit and firm level of analysis, conceptual and empirically.

Networks provide more access to information and improve the quality, relevance, and timeliness of this information (Adler & Kwon, 2002). As well as they can provide access to knowledge and resources that are not readily available via market exchanges (Gulati et al., 2000). The challenge to reach individual ambidexterity then becomes a matter of integrating internal and external knowledge, which relies on external brokerage and internal absorptive capacity (Cao, Gedajlovic, & Zhang, 2009; Raisch et al., 2009). Aligned to this challenge, the main highlight of network theory is that the actions of individuals are shaped by the networks in which they are embedded (e.g., Burt, 1992; Granovetter, 1985; Brass et al., 2004). Thus, it shows us the importance of analysing how characteristics of individuals’ ties may affect their ambidextrous behaviour. In a study with the police, Fors and Hansén (2014) concluded that ambidexterity is foremost achieved externally.

External networks refer to common topics such as social interaction, relationships, connectedness, collaboration, collective action, trust, and cooperation (Provan, Fish & Sydow, 2007). External ties can be customers, suppliers, producers, services providers and competitors. In this research, the focus will be on external network ties that employees discuss important matters with. Podolny and Page (1998) identify that networks can facilitate learning via the transfer of knowledge from one firm to another, in which the network acts as a conduit for processing and moving knowledge. Firms have ties with other firms, which gives their employees exposure to various types of knowledge that are potentially valuable. As Powell states; "The most useful information is rarely that which flows down the formal chain of

(13)

command in an organization, or that which can be inferred from price signals. Rather, it is that which is obtained from someone you have dealt with in the past and found to be reliable" (1990, p. 304). Individuals engage in interactions and networking, which facilitates the flow of information. Normally, social ties in certain strategic locations can provide an individual with useful information about opportunities and choices otherwise not available (Lin, 2001). External contacts perform a very important role to obtain the assets and the identification of entrepreneurial opportunities, since economic actions are embedded within larger interorganizational networks (Burt, 1992; Granovetter, 1985). For instance, a buyer may benefit indirectly through the transfer of knowledge via interorganizational learning, not only directly from the innovation itself (Hamel, 1991; Child, 2001). If a buyer is selling an innovative product or service, you learn from the innovation itself, but even when the product or service is not innovative, you can still learn in the interaction with this actor that could affect your ambidexterity.

This knowledge can be received or gathered from external actors through different channels, such as interorganizational channels, conversations and social interactions with individuals with similar goals (Dutta & Crossan 2005, p. 437), also formal and informal meetings, mails, telephone calls and e-mails. Enkel et al. (2016) found empirical evidence that identifying external knowledge from various sources provides individuals access to diverse knowledge, enabling them to create new combinations of internal and external knowledge, which contributes to both exploratory and exploitative innovation. Whenever this contribution happens within a certain timeframe, this will enforce ambidextrous behaviour. However, their research was focused in only in employees from one company (Bosch), which can lead to bias, therefore it is necessary to confirm these findings with more empirical research in different industries and organizations in varying sizes. Due to the fact that these aspects can change the individuals’ behaviours towards ambidexterity.

(14)

Rogan and Mors (2014) analysed ties in the internal network (colleagues in the same company) and external networks (ties with work related people outside the boundaries of the firm) of senior managers in a management consulting firm. Their study revealed differences in the contact heterogeneity and informality of ties in the networks of senior managers who engaged in ambidexterity compared with managers that predominately explored or exploited. This study will go further and add other layers to these finding. The effect of culture in the relationship of the influence of external network ties in individual ambidexterity will be added. As well as, by controlling diverse layers in the hierarchy, it is possible to evaluate how employees in each level of the hierarchy behave towards ambidexterity. This study will enrich the literature in understanding the combination of exploration and exploitation at the individual level of analysis (Gibson & Birkinshaw, 2004; Jansen, George, Van den Bosch, & Volberda, 2008; Lubatkin et al., 2006; Smith & Tushman, 2005; Mom et al, 2007; 2009).

This study will support the literature by researching about the effects caused by interactions with the external network in individuals’ ambidexterity, going beyond current findings by considering this relationship in broader levels in the hierarchy (employees in different levels), instead of only managers (Mom et al., 2009). As well as the characteristics of the network, instead of its structure, as antecedent to individual ambidexterity. Additionally, how the culture of individuals’ countries impacts the influence of their external network ties in their ambidextrous behaviour. By analysing individuals from countries and investigating whether their culture influence their ambidexterity, will also add a broader perspective of influence on individuals’ ambidexterity.

Relevant aspects that could have influence in ambidextrous behaviour will be controlled in this study. The level of employees in the organization hierarchy can influence their external network (who they will interact with, as well as the types of interaction) and the effects that these individuals of their network will influence their knowledge to be ambidextrous. Kaupilla

(15)

and Tempelaar (2016) argue that their study was the first to investigate the antecedents of ambidexterity among non-managerial employees, thus being very relevant to contextual ambidexterity. Therefore, the results from this study intend to go beyond and find the influence of external networks ties in employees’ ambidexterity. The size of the company and functional area might impact as well considering the amount of interactions these employees will have in their network. The environment the company is insert in also can have influence since some companies are embedded in more dynamic environment requiring constantly innovations. Consequently, all these differences might vary the heterogeneity and informality of ties.

Therefore, this research attempts to enrich the organizational literature and support management with the effects that interactions with external actors can influence ambidextrous behaviour on individual level, developing the theory of organization learning. Researching the impact that collectivism has on the influence of external network in individuals’ ambidexterity. Thus, this study will improve our knowledge regarding individual ambidexterity, giving us the grasp to enhance it. Hence, the research questions this paper aims to answer is the following:

RQ 1: What is the effect of employees’ interaction with external network ties on their ambidextrous behaviour?

RQ 2: What is the moderating effect of culture in this relationship?

These findings will allow employees to understand the drivers to their ambidextrous behaviour from their network interactions. Additionally, managers can use this result to improve knowledge sharing within the company’s external network, for example inducing interactions with external actors that employees can have more access to tacit knowledge that can increase their levels of exploration and exploitation. Thus, companies will learn how employees’ interactions with external actors can affect effective learning, adaptation, and survival in the long term. Furthermore, network is a central area in the effort to achieve both

(16)

integration and separation between exploration and exploitation (Stadler et al., 2014). This implicates the importance of studying the field of networks in relation to ambidexterity.

Previous studies in the individual level of ambidexterity analysed exclusively managers (e.g., Mom et al., 2009; Rogan & Mors, 2014). Kaupilla and Tempelaar (2016) argue that they are the first ones to investigate the antecedents of ambidexterity among non-managerial employees, but their research is relating ambidexterity to learning orientation and paradoxal leadership. Therefore, this study aims to be one of the first ones to add to the current literature the knowledge about the effects of the external network in the individual ambidexterity among employees, not only managers, due to the fact that ambidexterity is demonstrated in the specific actions of individuals throughout the organization and in their daily work, individuals often face choices of how they should spend their time (Gibson and Birkinshaw, 2004).

As well as contribute to future research on contextual ambidexterity (Adler et al., 1999; Gibson and Birkinshaw, 2004), which suggests that each employee’s ambidextrous behaviour adds to overall organizational ambidexterity (Simsek, 2009). Consequently, based on the findings, managers can develop and implement a framework for employees to improve their interactions.

Finally, by learning how culture influence in ambidexterity, companies can decide in which country they should develop and execute certain activities that require ambidexterity. In addition, that employees’ external network would have more influence in sharing information relevant for ambidextrous behaviour.

Informality of ties

Informal ties the ones independent of the employees’ role, when they use mostly personal resources to maintain the relationship with this external contact. Rogan & Mors (2014) found a positive relationship between informal ties and ambidexterity (and exploration).

(17)

Furthermore, they found that if the managers in a firm maintain primarily formal ties (related with the employees’ formal role in the company), they are likely to tend more toward exploitation instead of ambidexterity or exploration (Rogan & Mors, 2014). This research examines whether the effect of informality is also positive in individuals in diverse levels in the hierarchy, functions and countries. Additionally, to whether the collective or individualistic culture of the employees’ country influences this relationship.

Informal ties were chosen to analyse this effect, due to the fact that they are the result of nonprescribed social interaction processes and is more critical than formal ties when managers need to go outside existing boundaries (within or outside the firm) to accomplish tasks (Ibarra, 1993; Tushman & Romanelli, 1983). Additionally, Keinbaum and Tushman (2007) demonstrate that individuals use their informal network, instead of formal, to initiate cross-line-of-business. Which relates to ambidexterity in the lenses that cross-fertilization of knowledge between new and existing business of the company selection is due to ambidextrous behaviour (O’Reilly & Tushman, 2004, Taylor & Helfat, 2009).

Kanter (1983, 1988) argued that although some of these resources may be attached to an individual's formal organizational position, most of them must be acquired through informal network connections. In addition, informal ties lead individuals to avoid bureaucracy that restrict access to resources (Lovas & Sorenson, 2008) that can be crucial for ambidextrous behaviour. Previous research suggests that the broader entrepreneurs’ informal networks, the more likely they are to gather relevant information (e.g., Johansson, 2000), informal ties also adds cross-fertilization of knowledge between new and existing (Kleinbaum and Tushman, 2007), which is a characteristic of ambidexterity.

Therefore, in this study we will analyse whether:

(18)

Heterogeneity of ties

Employees with heterogeneous partners have access to more complementary resources and knowledge how to use these resources in their operations (Simsek 2009). In contrast, employees that has a homogenous network has little opportunity to consider different perspectives because the contacts think and act similar (Simsek 2009). Therefore, in order to behave ambidextrously, it is important to deal with heterogeneous contacts, since ambidextrous individuals refine and renew their knowledge, skills and expertise (Floyd & Lane, 2000; Sheremata, 2000; Hansen et al., 2001). Previous studies overlooked the roles that individuals play in exploring, assimilating and exploiting external knowledge (ter Wal, Criscuolo & Salter, 2011). Therefore, it is relevant to explore this field as antecedent for ambidextrous behaviour. The qualitative nature of ties in the network, therefore the content in the network is known to affect behaviour (Rodan & Galunic, 2004). According to them, the heterogeneous knowledge shared in the network is important to access new information and thus access a vast range of knowledge experts and resources that helped individuals to implement innovations, aligned with exploration. Rogan and Mors (2014) argue that since cross-fertilization of knowledge and ideas is aligned to ambidexterity, heterogeneous contacts in the network, thus the shared knowledge within these contacts, might affect individuals’ tendency to be ambidextrous. Going beyond their research, this study will investigate this relationship in different levels of the hierarchy, functions and countries. Previously to the influence of culture in this relationship, it will be analysed whether:

Hypothesis 2 (H2): External ties heterogeneity positively affect individual ambidexterity.

The moderating effect of Collectivism

Informality and heterogeneity of ties may influence the individual ambidexterity as discussed in the previous sections. In this section, the influence of culture in this relationship will be discussed.

(19)

National cultures differ regarding uncertainty, risk taking, and individualism (Hofstede, 1984; Kogut & Singh, 1988), which affect entrepreneurship, innovation, and cooperation (Shane, Venkataraman, & MacMillan, 1995). This point is also made by Furman et al. (2002), who suggested that institutional differences lead to cross-country variations in innovation activity in general (Furman, Porter, & Stern, 2002; Shane, 1994) and specially in exploratory and exploitative innovation (Hall & Soskice, 2001). Schneider and De Meyer (1991) add that culture influences individual work behaviour and decisions. Which may influence employees on their decision to behave ambidextrous. Other studies also addressed the effects of culture in innovation (Pandey & Sharma, 2009; Roth, 1995; Shane, 1994), which has been addressed by Tushman and O’Reilly as a feature of ambidexterity when both incremental and discontinuous innovation is pursued ambidexterity is achieved (1996, p. 24).

There are five culture dimensions according to Hofstede (1980): individualism, power distance, masculinity, uncertainty avoidance, long term orientation, and indulgence. These dimensions vary from country to country. His study was further developed in the GLOBE study (House, Hanges, Javidan, Dorfman, & Gupta, 2004) that designed different constructs and scales. In this study, the classifications of countries’ collectivism of Hofstede (2010) will be used and his argumentation why the GLOBE constructs are not suitable is presented further.

Individualism refers to the "relationship between the individual and the collectivity which prevails in a given society" (Hofstede 1980, p. 213). Individualism on the one side versus its opposite, collectivism, as a societal, not an individual characteristic, is the degree to which people in a society are integrated into groups. Individualism correlated strongly with national wealth (Gross National Product per capita). In Hofstede et al. (2010) study, individualism index scores are listed for 76 countries; individualism tends to prevail in developed and Western countries, while collectivism prevails in less developed and Eastern countries (Hofstede, 2011). According to Hofstede et al. (2010), the countries that the participants of this study live have

(20)

the following levels of collectivism, in a scale from 0 to 100, Brazil (38), Netherlands (80), UK (89), Belgium (75), Germany (67), Denmark (74) and Mexico (30).

Individualism/collectivism was considered as a key dimension of national culture (Hofstede, 1984), “perhaps the most important dimension of cultural differences in social behavior” (Triandis, 1988, p. 60) and has been strongly grounded in cross-cultural research (Hui & Triandis, 1986; Shane, Venkataraman, & MacMillan, 1995). It has also been addressed by organizational researchers and broadly studied at the organizational level (Chen et al., 1998; Earley & Gibson, 1998; House et al., 2004; Robert & Wasti, 2002).

Individualism involves the idea that in society everyone is supposed to take care of yourself and immediate family, they have an “I” consciousness, the identity based on the individual, they have emotion independence from organizations, focus on individual initiative, achievement and autonomy. While collectivism draws in the concept that people are born in extended families or clans who protect them in exchange for loyalty, they have a “we” consciousness, the identity is based in the social system, they have emotional dependence and moral involvement with organizations, they believe in belonging to the company and in group decisions. Collectivism, in contrast to individualism, refers to the degree that individuals care about group goals and base their identities on group memberships (Hofstede, 1984; House et al., 2004; Wagner & Moch, 1986). Collectivism indicates an orientation of individuals prioritizing group welfare and emphasizing sharing, cooperation and group harmony (Chen et al., 1998; Triandis, 1995).

The cultural dimension of collectivism has been extensively researched in social sciences (Earley and Gibson, 1998) and it was suggested to have cultural factors incorporated into research on organizational issues within an international context. In this study, I will focus of the indirect effect of individualism vs. collectivism on the relationship between external ties

(21)

informality and individual ambidexterity, as well as between external ties heterogeneity and individual ambidexterity.

Although some researchers suggested that individualism and creativity are positively related (Goncalo & Staw, 2006; Rank, Pace, & Frese, 2004) because individualistic culture emphasizes independence, uniqueness, autonomy, and achievement (Triandis & Suh, 2002). Thus, would decrease conformity pressures and contribute to generate novel ideas (Goncalo & Staw, 2006). Also, that collectivism can suppress the variety of ideas and potential actions that is fundamental for innovation (Herbig & Dunphy, 1998) and that collectivism stresses usefulness and appropriateness, while individualism encourages idea novelty and originality (Erez & Nouri, 2010). Which could decrease the levels of individual ambidexterity.

On the other hand, it has been argued about the positive effect that collectivism can have on ambidexterity (Yang et al., 2015, Muller et al., 2013). In collectivist cultures, individuals heavily attain their identity from the social system they are embedded in, such as their team, organization, and nation. Their actions are aimed at collective goals and aligned with the norms of the social system (Triandis, 1995). Collectivist cultures boost teamwork and good relationships between co-workers as well as between members of the organization and stakeholders, such as customers (Muller et al. 2013). By emphasizing the importance of creativity and innovation for the social context wellbeing, motivation for innovation can increase, leading to ambidexterity.

At the individual or group level, the relationship between collectivism and creativity has been extensively discussed and researches imply that collectivism can also benefit creativity (Bechtoldt, De Dreu, Nijstad, & Choi, 2010; Erez & Nouri, 2010; Wang, Xue, & Su, 2010). Members under high pro-social motivation (rather than pro-self-motivation) will produce more ideas (Bechtoldt et al., 2010). Collectivism can also alleviate the tension in knowledge transfer process, by leading to more organizational citizenship behaviour and a

(22)

lower level of social loafing, employees will make more efforts to balance the discipline and passion drivers in what they are going to learn, and how to learn. They will not choose their learning target based solely on the similarity of experience or expertise, instead they will choose their learning target based on the criteria of whether the learning outcome is beneficial to the organization, achieving the balance between exploration and exploitation in the knowledge transfer process (Yang et al. 2015), generating ambidexterity. In addition, Chatman et al.’s (1998) findings suggest that people in collectivist cultures—within which this firm is located (Hofstede, 1997)—will benefit more from heterogeneity than individualist cultures.

Mueller et al. (2013) analysed three of these culture dimensions, institutional collectivism, power distance, and uncertainty avoidance, because these are the three dimensions that are most likely to influence resource allocations to innovation projects. They hypothesized a negative relationship between collectivism and explorative innovation, but found that companies in collectivist countries benefit more from exploratory innovation than firms in individualist countries. They concluded that collectivistic culture can reduce the resistance from employees and middle managers in high-risk explorative activities by promoting teamwork and good relationships between organizational members. They also stated the necessity to examine the relationship between collectivism and ambidexterity.

Yang et al. (2015) researched the collectivism of an organization affect its ambidexterity regarding innovations and if centralization moderates this relationship. They found that creating collectivistic culture in the organization can help decrease the tension between exploration and exploitation in organizational learning and facilitate ambidexterity within the firm, as well as that this effect is weakened in a centralized hierarchy system. They used the Globe (House et al., 2004) construct that they ask the respondent to answer how individualistic or collectivistic the society and the company are and should be in a seven-likert scale, but it can be problematic if they don’t have the knowledge to compare to other country.

(23)

Therefore, this study considers Hofstede (1980) scores of collectivistic and individualistic countries to do this analyse. Yang et al. (2015) examined only in-group collectivism instead of societal one, their analyse was in the organizational level, considering the scales of He and Wong (2004) to measure exploration and exploitation, which when done simultaneously lead to ambidexterity (March, 1991). They also suggest that by adding the culture concerns in ambidexterity, we can extent the contextual ambidexterity literature (Gibson & Birkinshaw, 2004), and show that collectivism can help overcome the tension in personal drivers and then achieve balance between exploration and exploitation in all aspects of organizational learning. Therefore, impacting employees’ ambidexterity.

Collectivism can alleviate the tensions between exploration and exploitation that are embedded in all the sub-processes of organizational learning, namely knowledge creation, retention and transfer (Argote & Miron-Spektor, 2011), through balancing the discipline and passion of organizational members’ personal drivers (Kane, Argote, & Levine, 2005; Levin & Cross, 2004; Wong & Tjosvold, 2006). Therefore, it is hypothesised:

Hypothesis 3 (H3) - Collectivism moderates the relationship between external ties informality and individual ambidexterity, so that this relationship is stronger in countries with higher level of collectivism.

Hypothesis 4 (H4) - Collectivism moderates the relationship between external ties heterogeneity and individual ambidexterity, so that this relationship is stronger in countries with higher level of collectivism.

This research enriches the literature by considering the societal level of collectivism of the country the employees work at, with pre-measured levels of societal collectivism to avoid respondents’ ambiguousness, also because with this sample would not be possible to determine precisely the country level of collectivism. Hofstede (2006) criticized the GLOBE (House et

(24)

al. 2004) approach by saying that ask ‘as is’ questions basically assumed that these people could compare their society with other societies. It takes international experience and an unusually open mind to produce a credible comparison between one’s own society and others. Additionaly, as well as the items are too abstract (p.886). This study also adds to the literature by considering a continuous scale to measure the ambidexterity (Mom et al., 2009) instead of two scales, one for exploration and one for exploitation, which gives a more concrete measurement of ambidexterity since it accesses it by computing the multiplicative interaction between employees’ exploration and exploitation activities.

The findings of this study will be relevant also for managers. Since the world is in increasing globalization, managers often have a choice regarding the location of their firms’ innovation activities. Hence, knowledge about the advantages and disadvantages of specific locations, due to their institutional environment, enables managers to make more informed decisions on locations.

Conceptual model

The topics previously discussed, lead to this conceptual model as presented in figure 1. In this model, I explore the effects of interactions in the external network of employees on their individual ambidexterity.

Figure 1 Conceptual Model

H2 H1 H4 H3 Individual Ambidexterity External Ties Informality

External Ties Heterogeineity

(25)

Methodology

In this section, the research approach and design of this study will be presented. Initially, the sample and data collection will be explained, covering the strategies to collect information and respondents. Secondly, the measures for each of the variables will be presented. Consequently, this section clarifies the scales used for each variable. Moreover, all the measures were based on existing literature. Additionally, the exploratory factor analysis (EFA) for the dependent variable will be examined in this section. Lastly, the final part of this section will discuss how the raw data was assembled and analysed.

Data collection and sample

The purpose of this research is to find empirical evidence about the effects of external ties informality and heterogeneity on individual ambidexterity. To test the hypotheses, it was used only one data collection technique to collect the quantitative data. This choice was made taking some aspects into account. For instance, the fact that quantitative research is more scientific, larger amount of data could be gathered and analysed, thus having a wider perspective about the country effect. Besides that, it leads to less bias than in qualitative, since it would depend on interviewer interpretation. The data was collected, efficiently and economically, through an online survey via Qualtrics.com. Survey was chosen as to collect the data, since it has been used by researches to study the antecedents of individual ambidexterity (e.g. Mom et al, 2009), thus reliable constructs were used to gather the data. The survey was tested to last roughly 11 minutes to complete. The questions of the survey where all based on existing measures in the literature. The introduction in the survey and the messages we send to ask for responses stated the anonymity and confidentiality. Incentives were not provided, but it was stressed that their participation would be especially important to complete this study, well as for valuable findings about factors that impact individuals’ ambidextrous behaviour. The questionnaire was

(26)

administered in English, thus people from different nationalities could be reached and do not face issues due to translation of the items. The research was conducted with people from different companies working in different hierarchical levels mostly in the Netherlands and Brazil, but some in other countries as well. The goal was to collect data with as much variance across different contexts, to present results that are generalizable for various firms. The employees were from 92 different firms, most of them have in a medium level authority (48%), roughly equally distributed over gender 57% male, their average age is 33 years and their names were not asked to ensure confidentiality. Using a non-probability convenience sampling approach, respondents were contact throughout the personal network of the researcher, also from the respondents who forwarded the survey to their own network. This way more respondents could be reached and characterized a Snowball and network samples, involving participants via “referrals made among people who share or know of others who possess some characteristics that are of research interest” (Biernacki & Waldorf, 1981, p. 141). The invitation to fill in the survey was sent via e-mail, LinkedIn and Facebook, including its link to Qualtrics website. The responses were collected from May 2017 until middle June 2017 and during this time reminders were sent. Because of this approach it is not possible to know the response rate. From a total of 201 respondents, 29 only clicked in the survey link but did not answer it. There were 32 individuals who randomly stopped answering the survey, therefore they were considered as item non-response. And 18 answered that they did not have access to external network. Thus, throughout a list-wise deletation, the final sample contained 122 respondents. List-wise deletation was selected since it is an usual method to clean missing data (Peugh & Enders, 2004). Therefore, most of the missing values were due to that the employee stopped answering the questionnaire for an unknown reason (32 individuals) or because they do not interact with external ties, 18 employees answered that they did not have interactions with

(27)

external network through their job. Nevertheless, this sample is still adequate, since Kass and Tinsley (1979) recommended having between 5 and 10 participants per variable.

Measures

The measures and scales are selected as result of the literature review. The questions were selected in the way that similar questions were avoided to prevent the questionnaire being too large. Only one kind of questionnaire was sent and used to assess the employees of different firms. All independent variables (heterogeneity and informality) and dependent variable (individual ambidexterity) were assessed trough assessment and reflect therefore a self-assessment of the employees. The moderator was accessed through a pre-determined score for each country of the respondents.

Dependent variable

Individual ambidexterity is defined as the dependent variable of this study, which means

the behavioural orientation toward combining exploration and exploitation related activities within a certain period of time (cf. Gibson & Birkinshaw, 2004, p. 210; O’Reilly & Tushman, 2004, p. 81; Tushman & O’Reilly, 1996, p. 24). Most of the constructs to measure ambidexterity are for organizational level and consider separated exploitation and exploration measures and combine them (Gibson & Birkinshaw, 2004; He & Wong, 2004). To measure individual ambidexterity, the construct Mom et al (2009) was used. This construct is a continuum measure that accesses ambidexterity by computing the multiplicative interaction between employees’ exploration and exploitation activities. It has 14 items with a seven-point Likert scale (1 = not at all; 7 = to a very great extend) in which employees were requested to indicate “To what extent did you, last year, engage in work related activities that can be characterized as follows”. Seven of these items measure exploration (α=.812) and the other seven measure exploitation (α=.734). The items of this construct can be found in the Table 1.

(28)

To check for convergent and discriminant validity, an exploratory factor analysis (EFA) was performed with orthogonal rotation (Varimax) with all 14 items. The Kaiser–Meyer–Olkin measure certified that the sampling is adequate for the analysis, KMO = .725. Bartlett’s test of sphericity χ² (91) = 620,786, p < .001, indicated that correlations between items were large enough for EFA. This analysis showed two factors retained after rotation, one exploration scale with the seven exploration items (Eigenvalue = 3.9) and one exploitation scale with the seven exploitation items (Eigenvalue = 2.5). And no item cross-loading was greater than 0,35. According to Stevens’ (2002) loadings greater than 0.4 represent substantive values and all of them were apart of the “Activities which serve existing (internal) customers with existing services/products” that scored 0,37 that rounded would be 0,4, therefore it can be considered substantive enough. Even though the almost cluster of this item might be due to its content and how it is perceived by the employees. The exclusion of the item was performed, but since relevant improvement was not observed, all of the items were kept.

Table 1 - Factor Analysis for Individual Ambidexterity (Mom et al., 2009)

Factorᵇ Itemsª

To what extent did you engage in work related activities that can be characterized as follows:

1 2

An employee's exploration activities (α=.81)

Searching for new possibilities with respect to products/services, processes, or markets 0,78 0,06

Evaluating diverse options with respect to products/services, processes, or markets 0,79 0,04 Focusing on strong renewal of products/services or processes 0,71 0,08 Activities of which the associated yields or costs are currently unclear 0,70 0,00

(29)

Activities requiring you to learn new skills or knowledge 0,45 0,29 Activities that are not (yet) clearly existing company policy 0,73 -0,17 An employee's exploration activities (α=.73)

Activities of which a lot of experience has been accumulated by yourself 0,34 0,52 Activities which you carry out as if it were routine 0,07 0,70

Activities which serve existing (internal) customers with existing services/products 0,29 0,37

Activities of which it is clear to you how to conduct them -0,03 0,73 Activities primarily focused on achieving short-term goals 0,27 0,64

Activities which you can properly conduct by using your present knowledge -0,03 0,75

Activities which clearly fit into existing company policy -0,21 0,59

ª Items are quoted from survey.

ᵇ Extraction Method: principal component analysis. Rotation Method: Varimax with Kaiser normalization.

Independent variables

External ties informality was measured with an one item scale by asking the employees

the resources used to build and maintain each of their relationships with their external contacts, with a five-point Likert scale, the combination of resources they used to build and maintain each relationship with 5 indicating “independent of formal” and 1 indicating “formal”, the higher the value more informal the ties are (Rogan & Mors, 2014). Before asking about the independent variables, I asked the respondents to name 5 of their contacts that they work with, who are not employed at their company, that they would discuss important matters with. It was informed to them that if they were not comfortable with using the real name, they could use initials or nicknames. The use of five alters has been prominent in a number of network papers including Aldrich, Rosen, and Woodward (1987), Podolny and Baron (1997), Nicolaou and

(30)

Birley, (2003), and Ostgaard and Birley (1994). From the generation of contacts, respondent describes the relationship with each cited contact (Fischer, 1982; Burt, 2002). This variable was calculated with the average of these up to five contacts informality level of each employee. Hence, greater values indicate more informality of ties in the employees’ network.

External ties heterogeneity is about how heterogeneous the knowledge flowing across

the ties is (Rodan & Galunic, 2004; Mors, 2010). For its measurement, it was asked “how similar is your contacts’ knowledge compared to your own knowledge?”. To be answered in a 5- Likert-scale from very similar to very different (Rodan and Galunic, 2004). An example to make the question clearer was also added. In which explained that a football player and a football coach have very similar knowledge, while an airline pilot and a computer scientist have very different knowledge.

Moderator

Collectivism/Individualism is one of the dimensions of culture presented by Hofstede (1980). In his study, he gave a score for 76 countries. The scale is from 0 to 100 where the closer to 0 more the country is collectivistic and the closer to 100 higher the level of individualism in that country (Hofstede, 2010). This study had participants in 10 different countries, which the majority was in Brazil (38,5%) and in the Netherlands (43,4%). According to Hofstede et al. (2010), the countries that the participants of this study live have the following levels of collectivism, in a scale from 0 to 100: Brazil (38), Netherlands (80), UK (89), Belgium (75), Germany (67), Denmark (74) and Mexico (30). Each of the employees received the Hofstede’s (2010) score for their countries.

Control variables

To control for inconsistent effects the variables, the following control variables were used in this study. Ambidexterity may be influenced by employees’ experience, that is one of

(31)

the preeminent drivers of behaviors (Sturman, 2003; Taras, Kirkman & Steel, 2010). The experience is measured as the number of months spent in their specific job (Quinones et al. 1995). Higher the level of experience in that organization, more knowledge about what it takes to manage conflicting tasks and whom to call on for support in tackling antagonistic demands (Reagans & McEvily, 2003). Therefore, besides age, it is important to control for tenure within

the organization since it can influence positively employees’ ambidextrous behaviour

(Tushman & O’Reilly 1996), on the other hand tenure in the current function is expected to affect it negatively, due to higher levels of specialization (Birkinshaw & Gibson 2004).

Hierarchical level in the organization can also influence employees’ ambidexterity, the higher

in the hierarch more ambidextrous he or she might be (Floyd and Lane, 2000). Dummy variables were created to control for low, medium and high hierarchical level, considering low as reference group. The functional area these employees work in is also controlled due to changes in their level of exploration and exploitation activities depending on the area (Duncan 1976). Therefore, to control for functional effects, three dummy variables were created, one for marketing and sales (Marketing & Sales), one for research and development (R&D) and operations is the reference group (Mom et al., 2009). Due to more access to resources, size of

the company might increase the positive effect of ambidexterity on firm performance (Yu &

Khessina, 2012). It was controlled by asking the number of employees working in their establishment (Audia & Greve, 2006) and dummy variables were created: small, medium, and large firms as reference group. The higher the degree of education, more cognitive ability to process information and learning (Papadakis et al., 1998), which according to Adler and colleagues (1999) might affect ambidexterity positively. To control for education, two dummies were created for employees with education lower than bachelor as control group, one indicating the ones that had bachelor or professional studies and other demonstrating the ones that studied master or PhD. Size of the external network, by asking “how many people that you

(32)

work with, who are not employed at your company would you discuss important matters with?”. This variable was used to control in case the size of their external network were responsible for changes in ambidexterity level. Tie strength that measured the closeness and interaction frequency (Levin & Cross, 2004) with the selected external ties in a three-items-scale (α =.69) in order to differentiate the effects of informality of ties and their strength.

Environmental dynamism, is also important to control since it may influence the degree of

employees’ engagement in exploration or exploitation activities or both (Jansen et al., 2006; Lewin et al., 1999). It was measured with a five-item (α =.67) in a 5-Likert-scale (1 = strongly disagree to 7 = strongly agree) by Dill (1958) to understand the changes in the market that they work in.

Controls were included based on two premises. Firstly, it was important to receive information about respondents to determine the heterogeneity of the sample. Second, some of these controls were used in the conceptual model based on relevant findings in existing literature.

Statistical model

Preparation of the data is necessary previous to its analysis and hypothesis test. Firstly, the data was cleaned, by running frequency tests for the previously mentioned variables for missing values. It is not possible to define the response rate due to the method the survey was shared, snowball and network sampling. From the employees that clicked in the survey, most of the lack of responses was partial non-response, which were people that opened the survey, but didn’t start answering. These ones were excluded and from the remaining answers, 32 were item non-response of respondents who missed some items randomly. These missing data was list wise deleted, as well as the ones that did not answer about the independent variables due to have no access to external network, leading to 122 respondents.

(33)

A normality text was performed to control the skewness and kurtosis of individual ambidexterity, external ties informalization and heterogeneity and collectivism. The test was performed showing normality of data, the only abnormality was a low kurtosis (-1.8) for collectivism, which is due to the fact that the majority of the respondents were in Brazil and Netherlands, having only few respondents in countries with scores in between these. Therefore, standardised variables were created to check for outliers, but none of them had Z>|3|. The only counter-indicative item, forth item of environmental dynamism, was recoded. To test if the set of items are closely related as a group, a reliability analysis was performed. It measured internal consistency for all the variables that have three or more items, indicating that they are reliable since they have Cronbach’s alpha roughly >.7. Variables with the means of the items of this study were created to test the hypotheses.

Finally, with the use of SPSS, a correlation matrix could be run, thus this study provides a table of correlation coefficients of all variables. It shows that informality (-.04, p < .05) and collectivism (.03, p < .05) have a small negative effect, while heterogeneity (.04, p < .05) has a small positive Pearson correlation to individual ambidexterity. Although, the correlation between the mentioned variables are relatively low (Cohen 1988, 1992), it still does not impact since they are not significant (p<.05). Regarding the control variables, some had relevant and significant correlation. Age was positively correlated to collectivism, which might be that older people from individualistic countries filled in the survey, since the higher in the scale more individualistic the country is. Having a master or higher degree had a positive correlation to informality of ties, therefore respondents with this degree of education had more informal contacts. High level employees had a positive correlation to collectivism also, probably for the same reason. Environmental dynamism is positively correlated to individual ambidexterity, probably due to the fact that as previously addressed dynamism induces individuals to behave more ambidextrously. This variable was also positively related to marketing employees, most likely due to the same

(34)

reason. Tie strength was positively correlated to high level and R&D employees, therefore these individuals probably had stronger relationships with their external ties.

Consequently, a hierarchical multiple regression analysis was performed to investigate the proportion of variance of the informality and heterogeneity of external ties, as well as the control variables, explained by the model. To analyse the moderating effect of the collectivism on the relationship between the level of informality and individual ambidexterity, as well as on the relationship between the level of informality and individual ambidexterity, the independent variables and moderator were mean-centered before analyzing the effect, thus avoiding any potential issue with multicollinearity (Lim et al., 2014). Model 1 of PROCESS on SPSS (Hayes, 2013) was used to test the hypotheses on the moderating effects.

(35)

Results

This section presents the results of this study. Firstly, the descriptive statistics for all the variables, used in this study, will be displayed in table 2. It contains also the correlation analysis of these variables, exposing whether the correlations are significant. Lastly, to test the hypotheses of this study, the hierarchal regression analysis is executed and shown in Table 3. Model 1 reports the control variables, model 2 incorporate the main effects, and model 3 adds the moderating effect.

Main effects

The main effects related to Hypothesis 1 and 2 are demonstrated in Model 2 of Table 3. External ties informality presented a negative coefficient, meaning that the higher the degree of formality, individual ambidexterity is positively affected. However, this effect is not significant (p>0.05) and in model 3 this effect is positive. Although in model 3 this relationship had the expected effect, it was not significant either (β = 0.43 p>0.05). Therefore, hypothesis 1 is not supported. On the other hand, external ties heterogeneity shows a positive coefficient, in model 3, which is aligned with the Hypothesis 2. Although, this relationship was not relevant in model 2, where the moderator effect was not present. It means that the more heterogeneous the external ties more ambidextrous the individual behaves. This relationship supports the H2 in model 3 (β = 0.63 p<0.10). Collectivism was not significant as main effect (β = -0.11 p>0.05) in none of the models, although only in model 2 it presented the expected effect.

Moderating effect

The moderating effect is shown in Model 3 demonstrating the effect of individualism in the relationship between external ties informality and individual ambidexterity, as well as between external ties heterogeneity and individual ambidexterity, respectively. The first interaction has a negative coefficient. Which means that the more collectivist the country is, the stronger will

(36)

be the effect of informality of their external ties on the employees’ level of ambidexterity. Although, Hypothesis 3 is not supported due to its insignificance (β = 0.20 p>0.05). The second interaction, on the other hand, is significant (β = -0.79 p>0.10), which means that the more collective the country is, the more positive will be the effect of heterogenous ties on individual ambidexterity. Thus, since this one is significant, it supports Hypothesis 4.

Control variables

All the three models show the control variables that were add in Model 1. Considering the values of Model 3, with the complete analysis, age would be negatively related to individual ambidexterity, but it is not significant (β = -0.09 p>0.05). Education, tenure in firm and in current function did not affect the ambidexterity (p>0.05). Employees who are low in the hierarchy and from big firms tended to be more ambidextrous, although it cannot be confirmed due to lack of significance (p>0.05). Nonetheless, people working in operations related jobs had more influence in their ambidexterity (p<0.10). Environmental Dynamism also is positively related to individual ambidexterity (β = 0.45 p<0.001).

(37)

Table 2 Descriptive Statistics

Mean SD Min. Max. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

1 Individual Ambidexterity 23.51 7.75 5.63 49.00 2 External Ties Informality 3.21 1.13 1.00 5.00 -0.04 3 External Ties Heterogeneity 28.02 1.02 1.00 5.00 0.04 0.06 4 Country Collectivism 62.48 21.45 27.00 91.00 -0.06 -0.14 0.03 5 Age 33.85 10.44 22.00 72.00 0.09 -0.19 -0.01 0.29** 6 Education - Master/ PhD 0.46 0.50 0.00 1.00 -0.11 0.31** -0.06 -0.15 -0.19* 7 Education - Bachelor/ Professional 0.11 0.31 0.00 1.00 0.01 -0.13 0.05 0.07 -0.03 -0.31** 8 Tenure in firm 77.40 72.30 1.00 405 0.08 -0.05 0.04 0.19* 0.58** -0.08 0.14 9 Tenure in current position 47.19 55.07 1.00 353 0.09 -0.04 -0.05 0.10 0.54** 0.04 0.07 0.75** 10 Firm Size - Small 0.44 0.50 0.00 1.00 -0.01 0.09 -0.05 0.00 0.00 -0.02 0.23* -0.16 -0.07 11

Firm Size - Medium

0.32 0.47 0.00 1.00 -0.04 0.02 -0.03 -0.02 -0.12 -0.06 -0.18* 0.04 -0.06 -0.60**

12 Hierarch - High 0.30 0.46 0.00 1.00 0.03 -0.02 -0.09 0.23** 0.43** -0.10 0.12 0.28* 0.22* 0.13 -0.26**

13 Hierarch - Medium 0.47 0.50 0.00 1.00 0.10 0.03 0.10 -0.02 -0.15 -0.01 -0.06 0.01 -0.03 -0.02 0.20* -0.62** 14

Function - Marketing &

Sales 0.35 0.48 0.00 1.00 0.09 -0.02 0.04 0.01 0.21* -0.16 0.14 -0.01 0.01 0.14 -0.02 0.11 0.02 15 Function - R&D 0.07 0.26 0.00 1.00 -0.18 -0.01 -0.06 0.00 -0.10 -0.01 0.00 -0.16 -0.11 0.00 0.01 -0.05 -0.02 -0.21* 16 Environmental Dynamism 5.13 1.06 1.25 7.00 0.39** 0.03 -0.03 -0.05 0.23* -0.04 0.01 0.10 0.12 -0.14 0.00 0.03 0.02 0.24** 0.02 (.67) 17 Tie Strength 2.91 0.57 1.20 4.22 0.00 -0.05 0.08 0.19 0.28* -0.24* -0.17 0.06 -0.05 -0.02 0.02 0.21* -0.11 -0.11 0.21* 0.01 (.69)

Notes: Exploration α=.81, Exploitation α=.73. Individual Ambidexterity = Exploration * Exploitation

External network ties and individual ambidexterity : the moderating effects of collectivistic cultures