MANAGING AMBIDEXTERITY ON FIRM-LEVEL: A META-ANALYSIS

(1)

MANAGING AMBIDEXTERITY ON FIRM-LEVEL:

A META-ANALYSIS

January 2014

Author A.W. de Boer Damsterdiep 269-55 9713 EE Groningen The Netherlands a.w.de.boer.1@student.rug.nl Student number: 1386654 Word count: 9088 University of Groningen

Faculty of Economics and Business MSc Strategic Innovation Management Master Thesis

(2)

1

Abstract

Innovation is pivotal to a firm’s survival. Nowadays firms have to be able to explore for new knowledge to create new innovations while simultaneously exploit its current knowledge to build on existing products. Ambidexterity can help a firm cope with challenges that come with managing innovation as it makes it possible for firms to simultaneously explore and exploit. In the academic literature, several studies have focused on organizational factors that can help a firm to perform exploratory and exploitative activities. However, many of the empirical results are fragmented and sometimes contradictory. This study sets out to create a more holistic picture of what organizational factors are beneficial to managing exploration and exploitation. In other words this study looks at possible firm antecedents that can benefit the ability to perform exploratory and/or exploitative activities. A meta-analysis was conducted in order to achieve this. Data was collected from 31 empirical studies and 7 factors were identified that possibly influenced the ability to explore and exploit. The meta-analysis showed that 3 factors are homogeneous positive significant for managing exploration: 1) managerial capabilities, 2) autonomy, and 3) strategic behavior. There are also 3 factors homogeneous positive significant for managing exploitation: 1) learning, 2) managerial capabilities, and 3) strategic behavior. The remaining factors are heterogeneous and for these factors a moderator analysis was conducted. For managing exploration 1 subgroup is homogeneous and positive significant: the combination of

decentralization and connectedness. For managing exploitation 3 subgroups are homogeneous and significant: 1) formalization, 2) structural differentiation, 3) the combination of connectedness and organizational capital.

Keywords: Innovation, exploration, exploitation, ambidexterity, meta-analysis, organizational

(3)

2

Introduction

In todays globalized economy it is pivotal for firms to be innovative in order to stay relevant and not become obsolete (Cardinal, 2001) as product-life cycles become shorter. To survive in dynamic and changing environments firms have to be able to simultaneously explore for new knowledge and products while also utilizing existing knowledge and build on current products (Chandrasekaran, Linderman, & Schroeder, 2012; De Clercq, Thongpapanl, & Dimov, 2013; Levinthal, March, & Wiley, 1993; March, 1991). It is of vital importance for firms to pursue exploration and exploitation at the same time because focusing on a single exploration or a single exploitation strategy will result in severe problems: “firms that neglect exploration and focus on the exploitation of established, safe technologies may lack the capabilities to adapt to an evolving environment. Firms that neglect exploitation and focus on the exploration of new, uncertain possibilities may become too slow at developing and refining existing competencies needed to compete in the current market” (Bierly III, Damanpour, & Santoro, 2009). However, exploration and exploitation are not concepts that are easily managed at the same because even though “exploration and exploitation are needed for the long-term survival of a firm, there is often a tension between the two due to constraints on resources and the firm’s strategic orientation” (Bierly III et al., 2009). In addition, firms experience problems with finding the right balance between exploration and exploitation or firms become path-dependent and fail to adapt to changing environments (Levinthal et al., 1993; Walrave, van Oorschot, & Romme, 2011). To remedy this problem firms need to find a balance between exploration and exploitation and one approach to achieve this is by creating ‘organizational ambidexterity.’ This concept proposes “firms to consider dual structures; different structures to initiate versus execute innovation. In this view, ambidexterity occurs sequentially as organizations switch structures as innovations evolve” (O’Reilly & Tushman, 2008: 193). Or in other words ambidexterity is the “synchronous pursuit of both exploration and exploitation via loosely coupled and differentiated subunits or individuals, each of which specializes in either exploration or exploitation” (Gupta, Smith, & Shalley, 2006).

The problem is that much of the research regarding the management of ambidexterity is fragmented as many different topics on firm-level are addressed, including structure (Cardinal, 2001;

(5)

4 Commandeur, 2004)and myriad others. Also previous research has been done in different industries and used different research methods. In addition the results of these studies are occasionally contradictory. For example Jansen et al. (2006) found a negative relationship between exploration and centralization while Cardinal (2001) results show a positive relationship between exploration and centralization.

Another issue is that the term ambidexterity is defined and measured differently by researchers, “the most notable differences in the conceptualizations of OA concern whether it refers to achieving an optimal balance between exploration and exploitation or whether it involves a combination of high levels of both exploration and exploitation” (Junni et al., 2013).

Therefore it is important to create a holistic view of the research on managing ambidexterity on a firm-level. This paper tries to resolve this problem by using a meta-analysis in order to bring together these different studies and thereby finding out the antecedents for successfully managing an

ambidextrous organization. Due to the different definitions and measurements of ambidexterity this research will focus on the two dimensions of ambidexterity; exploration and exploitation, as there is a consensus among researchers regarding the definition of these terms. The results of the meta-analysis can make several contributions to the ambidexterity research:

1. Combine quantitative research as to develop a holistic view on ambidexterity management on firm-level.

2. Identify organizational factors that positively or negatively impact the management of exploration, exploitation, and ambidexterity.

In addition, the results can help managers better understand the characteristics of an organization that inhibit or promote the ability of a company to simultaneously pursue exploration and exploitation.

(6)

5

Literature Review

Exploration

Innovation can be distinguished in two separate parts, exploration and exploitation. Exploration is concerned with finding new knowledge, products, markets, technologies etc. (Miller & Friesen, 1983). Therefore exploration is often characterized by presenting radical innovations or products/services that are novel to the industry. By exploring firms move away from their current knowledge and exhibit “behaviors characterized by search, discovery, experimentation, risk taking and innovation“ (He & Wong, 2004). The importance in exploration lies in the fact that it can help firms to stay relevant and ahead of the competition. By acquiring new knowledge firms can anticipate on the future needs of customers and even create new markets (Jansen et al., 2006). Exploration is a necessary but risky endeavor in order to obtain new knowledge, products, technologies, and resources (Ireland & Webb, 2009). The exploration process involves “efficiently managing a breadth of resources while managing the uncertainty that surrounds the potential effectiveness of the resources” (Ireland & Webb, 2009).

Exploitation

Exploitation is characterized by incremental innovations and builds on “existing knowledge and skills, improve established designs, expand existing products and services, and increase the efficiency of existing distribution channels” (Jansen et al., 2006: 1662). Exploitation is often a more comfortable strategy for firms as it deals with the known and makes it attractive due the fact that less risk is

involved. However, when a firm sticks to expanding on its current knowledge and capabilities it can lead to organizational myopia. When this happens and circumstances in the environment change the core capabilities of a firm can turn to core rigidities (He & Wong, 2004). In order to exploit successfully firms have to “be able to efficiently manage a relatively narrow set of resources to facilitate speed and

accuracy and the process is characterized by fewer and less influential sources of uncertainty” (Ireland & Webb, 2009).

Ambidexterity

(7)

6 concepts of alignment and adaptability. The former deals with a firm’s efficiency and builds on the firm’s existing capabilities and knowledge while the latter focusses on experimentation with new knowledge. In other words alignment aims to “refine its current competencies to improve operational efficiency” and adaptability aims to “reconfigure the firm’s current activity set, with an emphasis on

experimentation with radically new innovations” (De Clercq et al., 2013). Ambidexterity can help to manage exploratory and exploitative activities and thus result in more innovation and better

performance (O’Reilly & Tushman, 2004). However, this study looks not at the outcome, for example performance, but at the factors that can benefit the ability to perform exploratory or exploitative activities.

Previous research has attempted to identify organizational antecedents that can promote or inhibit exploration and exploitation including the following: structure, learning, autonomy, partner

characteristics, technology and strategic behavior. The following section will discuss these organizational antecedents.

Structure

The structure of an organization can help managers to guide the activities towards exploration or exploitation. Structure is comprised of several concepts, including centralization, formalization, structural differentiation, decentralization, organizational capital, organizational complexity, and connectedness (Cardinal, 2001; Chandrasekaran et al., 2012; de Visser et al., 2010; Dewar & Dutton, 1986; Ettlie, Bridges, & Keefe, 1984; Foss, Lyngsie, & Zahra, 2013; Jansen et al., 2006; Menguc & Auh, 2010; Subramaniam & Youndt, 2005). Centralization can be described as the way a firm organizes its decision-making authority and this has an impact on the innovation process because “the concentration of decision-making authority prevents innovative solutions, while the dispersion of power is necessary for innovation” (Damanpour, 1991). With high centralization communication flows of knowledge and ideas in a firm will be reduced and therefore hurt explorative activities, while for exploitation high centralization will ensure efficiency and the speeding up of processes (Damanpour, 1991; Jansen et al., 2006).

(8)

7 regulations (Cardinal, 2001; J. Jansen et al., 2006). Formalization can help reduce variation and thereby increasing the efficiency of processes and thus increasing the ability to exploit (Jansen et al., 2006).

Structural differentiation can help to avoid exploration and exploitation conflicting with each other within a firm as “differentiation occurs through physical separation in the organization”

(Chandrasekaran et al., 2012). Exploration and exploitation projects are separated so that the focus can stay on both and not on one strategy and firms can do this by functional separation – departments working separately – or by cross-functional structures, i.e. departments working together and sharing information (de Visser et al., 2010). Exploratory activities can benefit from cross-functional structures as it promotes knowledge sharing and cooperation between departments which consequently can lead to reduction in risk and uncertainty (de Visser et al., 2010). For exploitation and incremental innovations there is less interdependence between departments and therefore a functional structure will suffice.

Decentralization deals with the distribution of decision making authority in a firm and a high level of decentralization leads to a high degree of delegation which in turn is beneficial for exploration because employees need to have the empowerment to act swiftly on new opportunities (Foss et al., 2013).

The concept of organizational capital is defined as “the institutionalized knowledge and codified experience residing within and utilized through databases, patents, manuals, structures, systems, and processes” (Subramaniam & Youndt, 2005). This knowledge is intrinsic to the firm and will be preserved even as employees come and go. In addition organizational capital “is the set of intangibles of explicit as implicit nature that structure and develop the organizational activity of the firm and is part of the firm’s structure” (López et al., 2006). Organizational capital is utilized through manuals, distinct guidelines and routines and will therefore increase the use of existing knowledge and thus promote exploitation (Subramaniam & Youndt, 2005). In contrast, organizational capital will not encourage the search for new knowledge and therefore hinder exploratory activities.

Another part of structure is organizational complexity which is the “the number of distinct

occupational specialties in the organization” (Dewar & Dutton, 1986). Organizational complexity should according to Dewar and Dutton (1986) be more important for managing exploration than for managing exploitation because complexity deals with specialists in the firm that can help make sense of new knowledge acquired by the firm from the external environment.

(9)

8 density or connectedness” (Jansen et al., 2006). According to Jansen et al. (2006) and Subramaniam & Youndt (2005) connectedness can stimulate exploration and exploitation as it encourages individuals in the firm to interact with each other which can help with the transfer of new and existing knowledge. However, too much interaction between familiar individuals can limit exposure to new knowledge and therefore hamper exploration.

Taking all this information regarding structure into account the following hypothesis can be formulated:

Hypothesis 1: The structure a firm applies will significantly impact a firm’s ability to manage

exploration and exploitation.

Learning

To be able to innovate a firm’s ability to acquire and integrate knowledge is a critical competency (van Wijk et al., 2012). Knowledge moves around in a firm between different units and each of these units may use the knowledge in distinct ways leading to diverse outcomes. How a firm deals with its knowledge acquisition and diffusion can impact the firm’s ability to explore and exploit as knowledge needs to be present in the knowledge stocks before innovation is possible (Dewar & Dutton, 1986; van Wijk et al., 2012).

These knowledge stocks of different business units vary in depth and breadth (Argyres, 1996; George, Kotha, & Zheng, 2008; van Wijk et al., 2012). Depth of knowledge can be considered as specialist

knowledge regarding a certain subject and it involves knowledge that is “extensive and sophisticated but

narrowly focused on a single or limited set of activities and fields” (van Wijk et al., 2012: 931). To have a deep understanding of a certain subject or domain can have three advantages according to George et al. (2008: 1455):

 Firms will be able to understand linkages and are therefore able to select the appropriate technology components to yield innovation. This will result in an increased reliability of the output.

 Depth of knowledge can increase the absorptive capacity of a firm regarding knowledge gathered from within or outside the firm.

(10)

9 Breadth of knowledge is the opposite as it concerns knowledge that is sophisticated knowledge that can be used for a variety of activities but it lacks sophistication to be specifically useful within one certain activity (van Wijk et al., 2012). By expanding the breadth of knowledge a firm can acquire new knowledge and possibly venture into “new areas of competence” (George et al., 2008). In addition, the expansion of the breadth of knowledge can help solve problems by increasing the variety of knowledge available and thus enlarging the possibility of finding the right solution.

Once knowledge is obtained it is transferred within the firm. These knowledge flows occur horizontally and vertically, where the former is concerned with the knowledge transfer from organizational unit to peer units and the latter deals with knowledge flows from the corporate headquarter to an organizational unit (van Wijk et al., 2012). Horizontal knowledge flows can help to increase unit’s breadth of knowledge because different units focus on different issues and have distinct knowledge regarding certain subjects relevant to their situation. When these units interact with each other they can transfer knowledge and units may obtain new knowledge that can help them battle unresolved problems or help to become more innovative (George et al., 2008; van Wijk et al., 2012). On the other hand, vertical knowledge flows can expand a unit’s depth of knowledge because the corporate headquarter has a large knowledge stock to choose from. Units can be reluctant when it comes to knowledge obtained from the corporate headquarter as the unit has specific knowledge of its local market and the unit will therefore only use knowledge that is relevant to the unit’s existing knowledge and thus expanding its depth of knowledge (van Wijk et al., 2012).

In addition, firms can also learn from exposure to the external environment which can happen “through membership in trade or professional associations or contact with manufacturer's

representatives” (Dewar & Dutton, 1986: 1424). Exposure to external information and knowledge can keep a firm up to date regarding advancements in its current products and processes and therefore increase its knowledge that can be used in innovative activities (Dewar & Dutton, 1986; George et al., 2008; van Wijk et al., 2012). Also, exposing units to external information and knowledge keeps units posted regarding “new technologies and serves as a link to the scientific network-enhancing innovation capabilities” (Cardinal, 2001:23).

(11)

10

Hypothesis 2: Learning is positively related to a firm’s ability to manage both exploration and

exploitation.

Partners & Network Characteristics

Research has shown that firms collaborating with other firms, institutions, or individuals can help to manage exploration and exploitation (Gilsing et al., 2008; Nielsen & Gudergan, 2012). There are several characteristics of the partner that impact the ability to manage exploration and exploitation;

technological distance, network density, partner experience, competence similarity, partner trust, customer-, supplier-, competitor-, university-, and independent expert collaboration (Gilsing et al., 2008; Nielsen & Gudergan, 2012; Schamberger, Cleven, & Brettel, 2013).

The technological distance between collaborating firms deals with the differences and or similarities in the firms’ technological profiles. Gilsing et al. (2008) argue that a larger technological distance between firms can have a positive effect on exploration because the firms have the opportunity to create new products or processes by combining new information and complementary resources. However, the technological distance between firms should not be too fast because there needs to be sufficient mutual understanding present (Gilsing et al., 2008).

When a firm has trouble with absorbing the novel knowledge acquired by a partner in a network it can be helpful to have another firm nearby in the network to give meaning to this novel knowledge. A dense network means that the firm’s partners are linked and they can complement each other and increase the absorptive capacity of the firm, which in turn can benefit exploratory activities (Gilsing et al., 2008). A side note is that too much network density can eventually lead to all partners having the same knowledge and therefore novel information will decrease.

The experience a firm has with a partnering firm can also have an impact on the ability to manage exploration and exploitation. It can be positive for exploitation because by repeated interaction it can help to increase the knowledge depth of the current technologies and knowledge. This repeated interaction can be detrimental for exploration because there will less novel knowledge coming into the firm and therefore partnering up with new firms is the better choice to increase the exposure to new knowledge (Nielsen & Gudergan, 2012).

(12)

11 influence on managing exploration and exploitation as more knowledge and resources are available for these activities.

The amount of trust placed in a partner influences a partnership and thus the ability to manage exploration and exploitation. When there is no trust present cooperation will decrease and knowledge flows decline, while the opposite is true when a great amount of trust between partners is present (Nielsen & Gudergan, 2012).

The type of partner a firm cooperates with can impact exploration and exploitation activities as well. Collaborating with customers can help exploitation by improving the current products, but customer collaboration will have a negative impact on exploration (Schamberger et al., 2013). This is the same for supplier collaboration as suppliers can help exploitation by identifying problems in current products (Schamberger et al., 2013). When it comes to competitor collaboration it can benefit exploration because by combining resources and investments it can reduce the chance of betting on the wrong technologies (Schamberger et al., 2013). Working together with universities can help a firm to acquire knowledge regarding new technologies as universities often work at the frontier regarding science and technology. A firm having access to this kind of knowledge can decrease the firm’s R&D expenditures while improving exploratory activities. This same sort of knowledge can be acquired by collaborating with independent experts (start-ups, engineering and design consultancies) who can provide new expert knowledge and capabilities to a firm and therefore positively influence exploration.

Thus, this study claims the following:

Hypothesis 3a: Partner characteristics (including technological distance, network density,

competence similarity, partner trust, and collaboration with

competitors/universities/independent experts) are positively related to a firm’s ability to manage exploration.

Hypothesis 3b: Partner characteristics (including partner experience, competence similarity,

(13)

12

Strategic Behavior

The strategy a firm enacts can impact the ability to manage exploration and exploitation. According to Lin, Lin & Chu (2008) the type of market orientation of a firm – taking into account the needs and preferences of customers –influences a firm’s exploration and exploitation activities. They mention two different strategic approaches: a proactive market orientation and a responsive market orientation. The former strategy enables exploratory activities as it “accentuates discovering and satisfying the latent and/or emerging needs of customers through undertaking market experiments” (Li, Lin, & Chu, 2008). The latter strategy responds on signals from the market regarding current products and is concerned with the expressed needs of customers (Li et al., 2008) and therefore benefits exploitation activities.

Another strategic orientation that can benefit exploration is a firm taking on the role of a prospector. Meaning that these type of firms engage in “elaborate and aggressive information-acquisition and they have been noted for their drive to search for and experiment with new opportunities” (Sidhu et al., 2004), which are activities beneficial to exploratory innovation. For exploitation a reactive strategic orientation is better for the same reasons as the responsive market orientation.

This leads to the following hypotheses:

Hypothesis 4a: A proactive strategic behavior is positively related to a firm’s ability to manage

exploration.

Hypothesis 4b: A responsive strategic behavior is positively related to a firm’s ability to manage

exploitation.

Autonomy

Previous research has argued that a firm can try to manage exploration activities through the use of autonomy and thus providing employees the freedom to pursue certain goals. A firm can do this by using certain concepts, including: organization mission, goal specificity, goal autonomy, and supervision autonomy (Cardinal, 2001; McGrath, 2001; Sidhu et al., 2004). The first, organization mission, is the “organization-wide shared agreement on the vision, business domain and competencies of an organization” (Sidhu et al., 2004). It can positively influence exploration by fostering collective knowledge seeking activities and a shared vision can help the firm to partake in acquiring external information and knowledge (Slater & Narver, 1995).

(14)

13 Goal autonomy can be described as to what extent managers define the goals, objectives and

outcome criteria for exploratory activities (McGrath, 2001). When exploring for new knowledge the firm operates in unfamiliar areas and it is thus difficult to define goals, objectives and outcomes. Therefore high goal autonomy will benefit exploratory activities.

Supervision autonomy is concerned with how management uses oversight through the observation of exploratory activities (McGrath, 2001). When the firm allows for greater autonomy during exploratory undertakings it will speed up decision making and the ability to improvise, which are necessary for successful new knowledge gathering activities (Cardinal, 2001; McGrath, 2001).

This information suggests the following hypothesis:

Hypothesis 5: Autonomy is positively related to a firm’s ability to manage exploration. Technology

Ahuja and Lampert (2001) argue that firms can manage and foster their exploratory activities by experimenting with novel, emerging and pioneering technologies. This behavior can prevent firms from falling into “the familiarity trap - favoring the familiar; the maturity trap - favoring the mature; and the propinquity trap - favoring search for solutions near to existing solutions” (Ahuja & Lampert, 2001).

Experimenting with novel technologies means that firms explore technologies that are unfamiliar to the firm and thus acquiring new knowledge and/or capabilities. Trying out emerging technologies – recently developed technologies – can help shift the firm away from mature technologies and into exploring possible opportunities for these emerging technologies (Ahuja & Lampert, 2001). Exploration can also be stimulated by the use of pioneering technologies, which are completely new technologies and the firm has to find problems this technology can be the solution for (Ahuja & Lampert, 2001).

In addition, Quintana-Garcia and Benavides-Velasco (2008) complement the argument made by Ahuja and Lampert by stating that the technological diversification of a firm is important to promote exploration. Technological diversification is defined as “the diversity in the knowledge system and principles underlying the nature of products and their methods of production” (Quintana-García & Benavides-Velasco, 2008). Being active in a wide variety of technologies can help solve unrelated problems and promote exploratory innovations. Thus:

(15)

14

Managerial Capabilities

The managers of a firm can use their capabilities to influence the behavior of their employees and thereby impacting exploration and exploitation (Chandrasekaran et al., 2012; Jansen, Vera, & Crossan, 2009). One way to do this is by transformational and transactional leadership behavior of strategic leaders – the people who bear the overall responsibility for an organization (Jansen et al., 2009). Transformational leadership involves the “intellectual stimulation, individualized consideration,

idealized influence, and inspirational motivation” (Jansen et al., 2009). Intellectual stimulation stands for the ability of leaders to have their employees be innovative and searching for new knowledge;

individualized consideration focuses on leaders being aware of the employees’ needs for achievement and growth; idealized influence refers to the extent employees respect and trust the leader; and inspirational motivation deals with the ability of the leader to act according to the vision he/she articulates and thereby motivates employees (Jansen et al., 2009). This type of leadership is suggested to be positive for exploratory activities as transformational leadership encourages risk taking, to challenge assumptions, and being innovative.

Transactional leadership is concerned with leaders making clear to employees what they have to do in order to be rewarded, and leaders observing the employees’ performance and taking corrective actions when necessary (Jansen et al., 2009). Transformational leadership can help exploitation as this type of leadership promotes efficiency of current activities and routines.

The strategic leaders also need to be able to make decisions regarding pursuing exploration or exploitation activities. Managers with the right “decision risk capability – the ability to constantly evaluate risks when making strategic choices between exploration and exploitation – can help organizations resolve the conflicting tensions when deciding between exploration and exploitation opportunities” (Chandrasekaran et al., 2012).

This leads to the following hypotheses:

Hypothesis 7a: The managerial capabilities, including decision risk capability and

transformational leadership, are positively related to a firm’s ability to manage exploration.

Hypothesis 7a: The managerial capabilities, including decision risk capability and transactional

(16)

15

Conceptual Model

The conceptual model in figure 1 provides a visual representation of the above mentioned theoretical constructs.

This model shows that structure and technology are related to managing exploration and that learning, partner characteristics, autonomy, proactive strategic behavior, and managerial capabilities are

positively related to managing exploration according to the literature review. For exploitation it shows that structure is related to managing exploitation and that learning, partner characteristics, responsive strategic behavior, and managerial capabilities are positively related to managing exploitation according to the literature review. The scope of the study is the influence of the organizational factors on

(17)

16

Data Collection and Methodology

To test the impact of different organizational antecedents on exploration and exploitation a meta-analysis was conducted. A meta-meta-analysis is a statistical research integration method (Song et al., 2008). A meta-analysis uses quantitative data from previous empirical studies and “applies statistical

procedures that are specifically designed to integrate the results of a set of primary empirical studies” (Song et al., 2008). This makes it possible to integrate the results of different studies and examine “relationships not investigated in the original primary studies, to correct for study artifacts, to test the effects of various moderators, and to uncover subtle trends that may be obscured in other approaches to summarizing research findings” (Junni et al., 2013). This research used a type of meta-analysis that studies the metafactors related to a particular construct, namely exploration and exploitation. Thus this study tries to find common metafactors that influence the ability to simultaneously pursue exploration and exploitation activities.

Studies selection process

The first step was to find literature that addressed organizational antecedents influencing either exploration, exploitation or both by using search-engines on the Internet, including EBSCO-host, Business Source Premier, and Google Scholar. Keywords that were used: innovation, exploration,

exploitation, ambidexterity, structure, and organizational antecedents. There was no difference made

between the journals the articles were published in as to not betray the spirit of a meta-analysis (Song et al., 2008). The objective was to collect as many studies as possible and later on the studies would be reviewed regarding the quality.

After the studies had been collected from EBSCO-host, Business Source Premier, and Google Scholar the references in the studies were checked in order to obtain more relevant studies for the meta-analysis. There were 115 studies that met the search criteria but these studies still had to be controlled for (1) using the correct level of analysis (firm-level), (2) using correct measurements of exploration and exploitation, and (3) presenting a correlation matrix with at least one antecedent of

(18)

17

Meta-analysis process

This study used the same protocol for the meta-analysis as the study done by Song et al. (2008) who based it on Hunter and Schmidt’s (1990) protocol. The Pearson correlations between a metafactor and the dependent variable were used in order to be able to make comparisons across research studies (Song et al., 2008). According to Hunter and Schmidt (1990) using the Pearson correlation is superior to using the regression coefficient as the Pearson correlation is not influenced by other variables in the model.

The next step in the meta-analysis according to Hunter and Schmidt (1990) it to correct metafactors for sample size differences and measurement errors. To correct for sampling error, the sample

correlation was weighted by sample size. The formula for the weighted average of correlations corrected for sample size is

where Ni is the sample size of the primary study i (Song et al., 2008).

The second step is to correct for measurement errors for which Cronbach’s alphas were used. The correlation coefficient was divided by the product of the square root of the reliability of the metafactor and the square root of the reliability of exploration/exploitation. The formula for real population correlation is

where is the compound reliability correction factor; is the average of the square roots of reliabilities of independent variables composing a given metafactor; and is the average of the square roots of reliabilities of dependent variables composing a given metafactor (Song et al., 2008).

(19)

18 according to Hunter and Schmidt (1990) this can be tested by using a variance based test. “The total variance in the correlation coefficient has three sources: variance due to artifacts (measurement errors), variance due to sampling error, and real variance due to heterogeneity of the metafactor. The

metafactor is assumed to be homogeneous, if the real variance is no more than 30 percent of the total variance” (Song et al., 2008). See Appendix 1 for the used formulas.

After it was determined whether metafactors were homogeneous the next step was to calculate if the metafactors were significant. This was done by checking whether the whole confidence interval (based on the real standard deviation) was above zero (Song et al., 2008). If this was the case, the metafactor would be considered significant.

(20)

19

Analysis and Results

After the literature review there were 13 metafactors for managing exploration and 10 metafactors for managing exploitation but due to metafactors having only one available study for analysis there were several discarded (see Appendix 2). This resulted in 7 metafactors for managing exploration and 5 metafactors for managing exploitation that could be used in the meta-analysis. Table 1 lists the definitions of the 7 distinct metafactors.

Table 1. Definitions of the 7 Meta-factors

Meta-factors Definitions Selected references

Structure

The relatively enduring allocation of work roles and administrative mechanisms that creates a pattern of interrelated work activities

Foss, Lyngsie & Zahra (2013)

Learning

The accumulation and transfer of knowledge from within and outside

the firm Van Wijk et al. (2012)

Autonomy

Freedom allowed to employees to

pursue certain goals McGrath (2001)

Technology

The knowledge system and principles underlying the nature of products and their methods of production

Quintana-Garcia & Benavides-Velasco (2008)

Partner Characteristics

The relational and structural characteristics of the firm’s

partners Nielsen & Gudergan (2012)

Strategic Behavior

The way a firm sets out to reach its

goals Li, Lin & Chu (2008)

Managerial Capabilities

The ability of managers to use leadership and decision making to impact the behavior of employees

(21)

20 Table 2 shows the results of the meta-analysis regarding metafactors impacting the ability to manage exploration and exploitation. The tables show total N, the aggregate sample size; and K, the number of correlations that build a given metafactor; and p, an estimate of the real population correlation. The spread of the real correlation variance is 95 percent confidence interval (Song et al., 2008). The real variance ([real variance/total variance]*100%) has to be below 30 percent in order for the metafactor to be homogeneous. ”In that case, the real variance is less than 30 percent of the total variance of

correlations from the primary studies and the remaining variance is likely due to other unknown and uncorrected artifacts, and therefore it can be neglected” (Song et al., 2008).The metafactors are divided between the concepts of exploration and exploitation.

Table 2. Results of the Meta-analysis

Metafactor Total N K p 95% Confidence Interval Real Variance (%) Moderator Exploration Learning 533 6 0.20 (0.04,0.36) 70 Yes Managerial Capabilities 157 3 0.43* 0 Autonomy 281 4 0.18* 0

Partner Characteristics 712 8 -0.02 (-0.13,0.09) 58 Yes

Strategic Behavior 312 2 0.17* 0 Structure 2430 13 0.32 (0.07,0.57) 93 Yes Technology 416 4 0.34 (0.32,0.35) 77 Yes Exploitation Learning 533 6 0.20* 0 Managerial Capabilities 157 3 0.53* 0

Partner Characteristics 1750 9 0.17 (0.00,0.32) 82 Yes

Strategic Behavior 352 2 0.37* 0

Structure 1496 13 0.14 (-0.04,0.33) 80 Yes

(22)

21 The results in Table 2 show that 6 metafactors are homogenous and all of those are significant. The following are suggested to be homogenous and positive significant metafactors correlated to managing exploration:

(1) Managerial Capabilities (p = 0.43, p<0.05) (2) Autonomy (p = 0.18, p<0.05)

(3) Strategic behavior (p=0.17, p<0.05)

Thus hypotheses 4, 5 and 7 are supported for managing exploration.

The following are suggested to be homogenous and positive significant metafactors correlated to managing exploitation:

(1) Learning (p = 0.20, p<0.05)

(2) Managerial Capabilities (p = 0.53, p<0.05) (3) Strategic behavior (p=0.37, p<0.05)

Thus hypotheses 2, 4 and 7 are supported for managing exploitation. Moderators

Table 2 shows that 6 of the 12 metafactors were heterogeneous, meaning that the relationship of the factors with managing exploration and exploitation is influenced by other variables. Moderators are variables that might explain the mixed results of the selected studies. Of these 6 heterogeneous metafactors 3 metafactors were able to be divided into subgroups in order to perform a moderator analysis, 2 metafactors for managing exploration and 1 metafactors for managing exploitation. For the other 3 metafactors there was not sufficient information present in the research to make subgroups as for certain metafactors only one study showed the results, for example, of different industries, countries or firm size while the other studies did not.

Table 3 shows the results from the moderator analysis. It shows total N, the aggregate sample size; K, the number of correlations that build a given metafactor; p, an estimate of the real population

(23)

22 The results in Table 3 show that of the 6 heterogeneous metafactors, 3 metafactors could be divided into subgroups while still having more than two studies to extract the necessary information from; for exploration: learning and structure; for exploitation: structure.

The metafactor learning (for exploration) was able to be divided in one subgroup: exposure to external knowledge. The metafactor structure, for both exploration and exploitation, could be divided into 4 subgroups: centralization, formalization, structural differentiation and mixed (decentralization and connectedness for exploration; connectedness and organizational capital for exploitation). The concepts of decentralization and connectedness for exploration; and connectedness and organizational capital for exploitation were grouped together by the author as the literature suggests possible

interconnectedness (Foss et al., 2013; Subramaniam & Youndt, 2005). Table 3. Results Moderator analysis

Metafactor Moderator p Total

N K 95% Confidence Interval Real Variance Exploration Learning Exposure to external knowledge 0.16 97 2 (0.02,0.30) 48 Mixed 4 Structure Centralization -0.17 437 4 (-0.27,-0.08) 51 Formalization -0.03 397 3 (-0.13,0.07) 58 Structural differentiation 0.16 684 3 (-0.02,0.35) 89 Autonomy & social

density: Decentralization and Connectedness 0.22* 376 2 0 Exploitation Structure Centralization -0.03 437 4 (-0.14,0.07) 54 Formalization 0.23* 397 3 (0.21,0.25) 6 Structural differentiation -0.26* 189 3 0

Social density & routines:

Connectedness and Organizational capital

0.26* 376 2 0

(24)

23 The results suggest that after the moderator-analysis there still remain heterogeneous subgroups and that 4 of the 9 subgroups showed to be homogeneous. Table 3 indicates that the combination of

decentralization and connectedness (autonomy and social density) are significantly and positively related

to managing exploration. Formalization is significantly and positively related to managing exploitation. Structural differentiation is significantly and negatively related to managing exploitation. The

combination of connectedness and organizational capital is significantly and positively related to managing exploitation.

Conclusion and Implications

Major Research Results

The results of the meta-analysis show that 6 of the 12 metafactors are homogeneous and significant. This means that autonomy, managerial capabilities, and strategic behavior help to manage exploration and that learning, managerial capabilities, and strategic behavior help facilitate exploitation. Of the 12 metafactors, 6 were heterogeneous and 3 of those 6 metafactors could be divided into subgroups for a moderator analysis. The results of the moderator analysis show that the 3 metafactors, 2 for

exploration and 1 for exploitation, could be split up into 9 subgroups. Regarding managing exploration there were 5 subgroups; 1 subgroup for learning, and 4 for structure. Concerning managing exploitation there were 4 subgroups, all of them for structure. Table 3 shows that 1 subgroup was homogenous and significant regarding managing exploration and 3 subgroups for exploitation. Of the 9 subgroups 5 subgroups remained heterogeneous after the moderator analysis. It is therefore necessary to conduct research on the heterogeneous, moderated factors listed in Table 2.

Theoretical and Managerial Implications

The meta-analysis showed that there are several heterogeneous factors and this may mean that the correlations of the studies used are contradictory. It is therefore important for a future study to analyze the correlation differences between studies regarding the factor and the dependent variable and taking into account various subgroups.

(25)

24 The managerial capabilities can foster exploration as well according to the results meaning that managers with the right abilities, including transformational leadership which can help to motivate employees to pursue exploratory activities. In addition, the results show that managers have to able to know when it is the right time or moment to go for exploratory opportunities (Chandrasekaran et al., 2012; Jansen et al., 2009).

When it comes to managing exploitation the results suggest that learning can have a positive effect. It is therefore important for firms to make sure the knowledge flows within the firm function properly and that employees are open and have the proper access, both formal and informal, to information and knowledge (van Wijk et al., 2012).

Just like with managing exploration, managerial capabilities can have a positive impact on managing exploitation according to the results. The only difference is that for exploitation transactional leadership is more appropriate to yield positive results as this type of leadership is more focused on achieving routines and efficiency and this promotes exploitation (Ireland & Webb, 2009; Jansen et al., 2009).

The meta- analysis insinuates that a proactive strategic orientation/behavior can benefit exploration. Using a strategy that encourages experimentation and new knowledge seeking can help a firm with the development of new innovations that cater to emerging needs of customers (Li et al., 2008; Sidhu et al., 2004). For managing exploitation a responsive strategic orientation/behavior is best fitting as it

facilitates the ability to use incremental innovation in order to respond to the current needs of customers.

(26)

25 manage exploitation. Thus connectedness can help transfer the institutionalized and codified knowledge in the firm and thereby increase the ability to exploit (Foss et al., 2013; Subramaniam & Youndt, 2005).

The above mentioned factors differ in managerial controllability. For example, the autonomy units experience can be controlled by managers as they can be set and changed by managers when necessary. A factor that is possibly less controllable is the social interaction between employees as it is hard to resolve relationship conflict which may occur between employees (Clercq, Thongpapanl, & Dimov, 2008).

Limitations & Future Research Directions

This meta-analysis has several limitations including the results showing that there are several

heterogeneous metafactors that remained heterogeneous even after the moderator analysis. This might mean that for certain variables there is still no consensus among the chosen studies in this research about the relationship of these variables with the dependent variables. It would therefore be beneficial to study these variables in order to find out if there are possible moderators or measurement

differences (type of industry, firm size etc.) at play that influence the different outcomes between studies. For example, looking at the individual outcomes of the studies used regarding the effect of centralization on exploration it shows that of the 4 studies 2 studies suggested a positive relationship; 1 study showed a non-significant relationship; and 1 study demonstrated a negative relationship. This example shows that there is more research needed regarding the effect of centralization on exploration and this might well be the case for several of the other heterogeneous factors.

Another limitation is the fact that this study only focuses on the organizational factors of a firm and does not take into account the industry factors (competiveness, industry life cycle stage, dynamism, etc.) that can influence exploration and exploitation. It would therefore be beneficial to conduct a meta-analysis regarding the influence of industry factors on exploration and exploitation.

(27)

26

References

Ahuja, G., & Lampert, M. C. (2001). Entrepreneurship in the large corporation: a longitudinal study of how established firms create breakthrough inventions. Strategic Management Journal, 22(6-7), 521–543. doi:10.1002/smj.176

Aragón-Correa, J. A., García-Morales, V. J., & Cordón-Pozo, E. (2007). Leadership and organizational learning’s role on innovation and performance: Lessons from Spain. Industrial Marketing Management, 36(3), 349–359. doi:10.1016/j.indmarman.2005.09.006

Argyres, N. (1996). Technological Diversification and Divisionalization. Strategic Management Journal, 17(5), 395– 410.

Bierly III, P. E., Damanpour, F., & Santoro, M. D. (2009). The Application of External Knowledge : Organizational Conditions for Exploration and Exploitation. Journal of Management Studies, 46:3(May), 481–509. Cardinal, L. B. (2001). Technological Innovation Industry : The Use of in the Pharmaceutical in Organizational

Control and Development Managing. Organization Science, 12(1), 19–36.

Chandrasekaran, A., Linderman, K., & Schroeder, R. (2012). Antecedents to ambidexterity competency in high technology organizations. Journal of Operations Management, 30(1-2), 134–151.

doi:10.1016/j.jom.2011.10.002

Chang, Y.-Y., & Hughes, M. (2012). Drivers of innovation ambidexterity in small- to medium-sized firms. European

Management Journal, 30(1), 1–17. doi:10.1016/j.emj.2011.08.003

Clercq, D., Thongpapanl, N., & Dimov, D. (2008). When good conflict gets better and bad conflict becomes worse: the role of social capital in the conflict–innovation relationship. Journal of the Academy of Marketing Science,

37(3), 283–297. doi:10.1007/s11747-008-0122-0

Damanpour, F. (1991). Organizational Innovation : A Meta-Analysis of Effects of Determinants and Moderators.

Academy of Management Journal, 34(3), 555–590.

De Clercq, D., Thongpapanl, N. (Tek), & Dimov, D. (2013). Shedding new light on the relationship between contextual ambidexterity and firm performance: An investigation of internal contingencies. Technovation,

33(4-5), 119–132. doi:10.1016/j.technovation.2012.12.002

De Visser, M., de Weerd-Nederhof, P., Faems, D., Song, M., van Looy, B., & Visscher, K. (2010). Structural ambidexterity in NPD processes: A firm-level assessment of the impact of differentiated structures on innovation performance. Technovation, 30(5-6), 291–299. doi:10.1016/j.technovation.2009.09.008 Dewar, R. D., & Dutton, J. E. (1986). The Adoption of Radical and Incremental Innovations: An Empirical Analysis.

Management Science, 32(11), 1422–1433. doi:10.1287/mnsc.32.11.1422

Ettlie, J. E., Bridges, W. P., & Keefe, R. D. O. (1984). Organization Strategy and Structural Differences for Radical versus Incremental Innovation. Management Science, 30(6), 682–695.

Foss, N., Lyngsie, J., & Zahra, S. (2013). The Role of External Knowlegde Sources and Organizational Design in the Process of Opportunity Exploitation. Strategic Management Journal, 1471(April), 1453–1471.

(28)

27 George, G., Kotha, R., & Zheng, Y. (2008). Entry into Insular Domains : A Longitudinal Study of Knowledge

Structuration and Innovation in Biotechnology Firms, (December).

Gilsing, V., Nooteboom, B., Vanhaverbeke, W., Duysters, G., & van den Oord, A. (2008). Network embeddedness and the exploration of novel technologies: Technological distance, betweenness centrality and density.

Research Policy, 37(10), 1717–1731. doi:10.1016/j.respol.2008.08.010

Gumusluoglu, L., & Ilsev, A. (2009). Transformational leadership, creativity, and organizational innovation. Journal

of Business Research, 62(4), 461–473. doi:10.1016/j.jbusres.2007.07.032

Gupta, a. K., Smith, K. G., & Shalley, C. E. (2006). the Interplay Between Exploration and Exploitation. Academy of

Management Journal, 49(4), 693–706. doi:10.5465/AMJ.2006.22083026

He, Z.-L., & Wong, P.-K. (2004). Exploration vs. Exploitation: An Empirical Test of the Ambidexterity Hypothesis.

Organization Science, 15(4), 481–494. doi:10.1287/orsc.1040.0078

Ireland, R. D., & Webb, J. W. (2009). Crossing the great divide of strategic entrepreneurship: Transitioning between exploration and exploitation. Business Horizons, 52(5), 469–479. doi:10.1016/j.bushor.2009.05.002

Jansen, J., Van Den Bosch, F. a. J., & Volberda, H. W. (2006). Exploratory Innovation, Exploitative Innovation, and Performance: Effects of Organizational Antecedents and Environmental Moderators. Management Science,

52(11), 1661–1674. doi:10.1287/mnsc.1060.0576

Jansen, J., Vera, D., & Crossan, M. (2009). Strategic leadership for exploration and exploitation: The moderating role of environmental dynamism. The Leadership Quarterly, 20(1), 5–18. doi:10.1016/j.leaqua.2008.11.008 Jung, D. (Don), Wu, A., & Chow, C. W. (2008). Towards understanding the direct and indirect effects of CEOs’

transformational leadership on firm innovation. The Leadership Quarterly, 19(5), 582–594. doi:10.1016/j.leaqua.2008.07.007

Junni, P., Sarala, R. M., Vas, T., & Tarba, S. (2013). Organizational Ambidexterity and Performance: A Meta-Analysis,

27(4), 299–312.

Levinthal, D. A., March, J. G., & Wiley, J. (1993). The Myopia of Learning. Strategic Management Journal, 14, 95– 112.

Li, C.-R., Lin, C.-J., & Chu, C.-P. (2008). The nature of market orientation and the ambidexterity of innovations.

Management Decision, 46(7), 1002–1026. doi:10.1108/00251740810890186

López, J. E. N., Salazar, E. A., Castro, G. M. de, & Sáe, P. L. (2006). Organizational Capital as Competitive Advantage of the Firm. Journal of Intellectual Capital, 7(3), 324 – 337.

March, J. G. (1991). Exploration and Exploitation in Organizational Learning. Organization Science, 2(1), 71–87. doi:10.1287/orsc.2.1.71

McGrath, R. G. (2001). Exploratory Learning, Innovative Capacity, and Managerial Oversight. Academy of

(29)

28 Menguc, B., & Auh, S. (2010). Development and return on execution of product innovation capabilities: The role of

organizational structure. Industrial Marketing Management, 39(5), 820–831. doi:10.1016/j.indmarman.2009.08.004

Miller, D., & Friesen, P. H. (1983). Strategy-Making and Environment: The Third Link. Strategic Management

Journal, 4(3), 221–235.

Nielsen, B. B., & Gudergan, S. (2012). Exploration and exploitation fit and performance in international strategic alliances. International Business Review, 21(4), 558–574. doi:10.1016/j.ibusrev.2011.07.001

O’Reilly, C. A., & Tushman, M. L. (2004). The Ambidextrous Organization. Harvard Business Review, 81(4), 74–82. O’Reilly, C. a., & Tushman, M. L. (2008). Ambidexterity as a dynamic capability: Resolving the innovator’s dilemma.

Research in Organizational Behavior, 28, 185–206. doi:10.1016/j.riob.2008.06.002

Quintana-García, C., & Benavides-Velasco, C. a. (2008). Innovative competence, exploration and exploitation: The influence of technological diversification. Research Policy, 37(3), 492–507. doi:10.1016/j.respol.2007.12.002 Schamberger, D. K., Cleven, N. J., & Brettel, M. (2013). Performance Effects of Exploratory and Exploitative

Innovation Strategies and the Moderating Role of External Innovation Partners. Industry & Innovation, 20(4), 336–356. doi:10.1080/13662716.2013.805928

Sidhu, J. S., Volberda, H. W., & Commandeur, H. R. (2004). Exploring Exploration Orientation and its Determinants : Some Empirical Evidence *. Journal of Management Studies, 41:6(September).

Slater, S. F., & Narver, J. C. (1995). Market Orientation and the Learning Organization. Journal of Marketing,

59(July), 63–74.

Song, M., Podoynitsyna, K., Van Der Bij, H., & Halman, J. I. M. (2008). Success Factors in New Ventures: A Meta-analysis. Journal of Product Innovation Management, 25(1), 7–27. Retrieved from

http://doi.wiley.com/10.1111/j.1540-5885.2007.00280.x

Subramaniam, M., & Youndt, M. a. (2005). the Influence of Intellectual Capital on the Types of Innovative Capabilities. Academy of Management Journal, 48(3), 450–463. doi:10.5465/AMJ.2005.17407911 Van Wijk, R., Jansen, J. J. P., Van Den Bosch, F. a. J., & Volberda, H. W. (2012). How firms shape knowledge to

explore and exploit: a study of knowledge flows, knowledge stocks and innovative performance across units.

Technology Analysis & Strategic Management, 24(9), 929–950. doi:10.1080/09537325.2012.718666

Walrave, B., van Oorschot, K. E., & Romme, a. G. L. (2011). Getting Trapped in the Suppression of Exploration: A Simulation Model. Journal of Management Studies, 48(8), 1727–1751.

(30)

29

Appendix 1

(31)

30

Appendix 2

List of discarded metafactors due to insufficient information to perform a meta-analysis

Flexibility Human Resources Input-control Partners Size Social Network

Appendix 3

Publication sources of the articles included in the meta-analysis

Publication Source Number of Studies in Analysis

Academy of Management Journal 4

Druid Society 1

European Management Journal 1

Industrial Marketing Management 2

Industry & Innovation 1

International Business Review 1

International Journal of Innovation Management 1

Journal of Business Research 1

Journal of Business-to-Business Marketing 1

Journal of Management Studies 1

Journal of Operations Management 1

Management Decision 1

Management Science 3

Organization Science 2

Research Policy 2

Strategic Management Journal 3

Technology Analysis & Strategic Management 1

Technovation 2

The Leadership Quarterly 2

MANAGING AMBIDEXTERITY ON FIRM-LEVEL: A META-ANALYSIS