Environmental management in the Dutch food service industry: the role of organizational knowledge and collaborative learning
by Ellen Christine Oudejans 10561196
Reducing The Foodprint
Version: final Date: June 22, 2018
Supervisor: dhr. B. Szatmari Second reader: dhr. dr. G.T. Vinig
MSc BA: Entrepreneurship & Innovation
1
“The biggest threat to this planet is the belief that someone else will save it”
2
3 TABLE OF CONTENTS STATEMENT OF ORIGINALITY 2 Abstract 5 Introduction 6 Literature Review 9 2.1 Resource-Based View 9 2.1.1 Organizational Knowledge 10 2.2 Dynamic Capabilities 12 2.2.1 Collaborative Learning 13
2.3 Environmental Management Practices 14
Theoretical Framework 17 3.1 Hypothesis One 17 3.2 Hypothesis Two 19 3.2 Hypothesis Three 22 3.3 Hypothesis Four 25 3.4 Conceptual Model 27 Research Context 28 Research Design 30 5.1 Method 30
5.2 Data Collection and Sample 31
5.3 Independent Variables 31 5.4 Dependent Variable 32 5.5 Mediator Variable 33 5.6 Moderator Variables 34 5.7 Control Variables 35 5.8 Method of Analysis 36
Data Analysis and Results 37
6.1 Factor Analysis 37
6.2 Reliability Analysis 38
6.3 Check For Normality 39
6.4 Normality of Residuals 39 6.5 Homoscedasticity 39 6.6 Descriptive Statistics 40 6.7 Correlations 40 6.8 Regression 41 6.9 Mediation 42
4 6.10 Moderation 44 6.11 Post-Hoc Analyses 46 Discussion 48 7.1 Theoretical Implications 48 7.2 Practical Implications 52
7.3 Limitations and Future Research 53
Conclusion 54
References 55
5
Abstract
After the Brundtland report of 1987 on sustainable development, the consequences for organizations of being green in terms of financial and competitive performance gained increased attention. Both from CEOs as well as researchers. But the foundation for - rather than the antecedent of - adopting environmental management practices remains to be
determined. The current study thus looks into the underlying question of how to establish an environmental management system in the first place, by considering organizational
knowledge, collaborative learning, managerial systems and values + norms. The study domain involves the Dutch food service industry (n = 118), as research indicates that this industry severely contributes to the environmental burden and sustainable development could thus make a significant impact. It was found that tacit knowledge only positively affects the adoption of environmental management practices in combination with collaborative learning. No empirical evidence was found for a direct effect. On the other hand, explicit knowledge both directly and indirectly (through collaborative learning) contributes to the adoption of environmental management practices. Managerial systems (the organizational structure and technological infrastructure) negatively moderate the relationship between tacit knowledge and collaborative learning, indicating that a lower degree of decentralization and self-organization fosters the exploitation and exploration of tacit knowledge. For explicit knowledge, this effect has not been found. Lastly, values + norms do not moderate the relation between either tacit or explicit knowledge and collaborative learning. This indicates that the organizational culture of a food service establishment does not affect tacit knowledge and the mobilization of which.
Keywords: organizational knowledge, tacit, explicit, collaborative learning, sharing, environmental management practices, EMS, food service industry, Dutch, The Netherlands
6
chapter one Introduction
Despite the fact that the current president of the United States of America appears to disagree on the matter, several scholars have expressed the severity of global warming as an
undeniable phenomenon that is causing the temperature of the Earth to rise at an unusual pace (Root, Price, Hall, Schneider, Rosenzweig & Pounds, 2003; Stern, 2008). As a consequence, global warming is becoming an increasingly hot topic all around the world. Especially since the phenomenon, in combination with an ever-accelerating depletion of natural resources and beyond-steadily growing population, poses a significant risk on and considerable threat to the natural environment (Vörösmarty, Green, Salisbury & Lammers, 2000; McMichael,
Woodruff & Hales, 2006). The implications of these environmental changes are numerous: extreme weather conditions, dwindling biodiversity, political conflicts, deforestation,
pressure on food production, a rise of sea levels etc. (Stern, 2008). It becomes therefore rather evident that in order to reduce our environmental footprint, measures have to be taken.
“A historic moment for humanity” - the words used to conclude the Earth Summit of 1992; the first global conference sought to help governments and business rethink their practices and processes regarding the natural environment (UN, 1992). Although an ongoing debate exists about the right response to the various issues, the key denominator in these efforts is the concept of sustainable development. According to the Brundtland report, this refers to “meeting the needs of the present without compromising the ability of future generations to meet their own needs” (UN, 1987, p. 1). For governments, these sustainable development demands revolve around the issuance of different laws, standards and
regulations that aim to ease the environmental burden while stimulating economic growth at the same time. Simply complying with these decisions used to be sustainability policy of
7
many companies, but over recent years CEOs increasingly start to recognize the opportunities that respecting the environment brings (Tyteca, 1996; Hart, 1997).
A recent study by Porter and Kramer (2006) on the interdependency between business and society highlights the role of the Resource-Based View (RBV) in order to determine what it means to be green. The RBV states that the resources and capabilities a firm possesses determine its competitive advantage and overall performance (Barney, 1991). The natural RBV goes one step beyond this line of thinking, by highlighting the competitive advantages that derive from pollution prevention and resource reduction in order to reduce the footprint (Hart, 1995; Porter and Van der Linde, 1995; Christmann, 2000; Miles & Covin, 2000; King & Lennox, 2001). However, these studies are mainly focused on the capabilities and
complementary assets that are needed to create competitive benefits with green management as antecedent. This implies that environmental management has up until todays always been the independent variable. The underlying question of how to establish an environmental management system in the first place remains yet to be answered. The aim of this paper is therefore to determine the internal foundation that is needed for the adoption of
environmental management practices. In short: what does it take for a company to go green? The focus is hereby on the Dutch food service industry, as research indicates that this
industry severely contributes to the environmental burden and sustainable development could thus make a significant impact.
When it comes to the internal organizational foundation, numerous examples and definitions of resources and capabilities exist. However, it is widely recognized by scholars from the field of strategic management that knowledge is the common denominator of these resources and capabilities (Leonard-Barton, 1992; Grant, 1996; Gold, Malhotra & Segars, 2001). It follows that knowledge acts as facilitator as well as enabler, thereby affecting and shaping organizational actions. Hence the amount of knowledge residing within organizations
8
has a significant effect on the firm’s ability to ensure change and eventually to innovate (Grant, 1996; Teece, Pisano & Shuen, 1997; Subramaniam & Youndt, 2005). In a similar vein, it has been found that knowledge is the main determinant in the process of creating and leveraging organizational transformation, as knowledge equals the most strategically relevant resource of the firm (Nonaka,1994; Grant, 1999). It is therefore that moving towards
sustainable development - or more specifically, the adoption of an EMS - is first and foremost a knowledge-based process (Darnall & Edwards, 2006; Melville, 2010). However, given the high degree of velocity in today’s marketplace, knowledge on its own often appears to be insufficient in order to guarantee the effectiveness of environmental management. It has increasingly been found that the key determinant in the process of adopting sustainability practices revolves around the way knowledge is exchanged, assimilated and leveraged in order to ensure its success (Teece, 1986; Cash, Adger, Berkes, Lebel, Olsson, Pritchard & Young, 2003; Reed, 2008). Drawing upon this, the present study takes collaborative learning - a continuous attempt to learn together through the mobilization of knowledge - into account. This leads to the following research question: What is the role of organizational knowledge
and collaborative learning on the adoption of environmental management practices in the Dutch food service industry?
The remainder of this proposal is structured as follows: chapter two will give an overview of the current literature. Chapter three will elaborate on the proposed relationships between the variables and ends with a visual representation of those, by means of a
conceptual model. In chapter four the research context will be analyzed, whereas chapter five involves the research design. Furthermore, in chapter six the results of the current study can be found. In chapter seven and eight these results are respectively discussed and concluded.
9
chapter two Literature Review
This chapter presents the current research on the topic. This includes the Resource-Based View and Organizational Knowledge, Dynamic Capabilities and Collaborative Learning, and Environmental Management Practices.
2.1 Resource-Based View
Pasta sheets. Tomatoes. Garlic. Cheese. Basil. Some other optional vegetables. These are the ingredients (resources) of individual X required for tonight’s dinner. In order to turn these into an actual lasagne that either will not get burned in the oven or collapse during serving, he or she needs a certain amount of cooking expertise (capabilities).
Capabilities are the structuring and coordinating of organizational resources (Zander & Kogut, 1995). Or as Day (1994, p. 38) puts it: “assets are the resource endowments the business has accumulated (...); and capabilities are the glue that brings these assets together and enables them to be deployed advantageously.” Broadly speaking, these resources and capabilities can be looked at from one of two organizational perspectives. According to Porter (1991) and the Industrial Organization perspective, a company’s strategy and its competitive positioning are determined by external forces. On the other hand, the RBV looks at strengths and weaknesses rather than opportunities and threats. It thereby removes the assumption that firms are equal, as all have heterogenous resources that are distributed asymmetrically and thus pursue different strategies (Aral & Weill, 2007). Following from this, it is being argued that internal resources and capabilities determine organizational performance and competitive advantage (Wernerfelt, 1984; Barney, 1991). The current study takes an inside-out perspective, by considering organizational knowledge as the firm’s most strategic asset.
10
2.1.1 Organizational Knowledge
Knowledge, often wrongly used interchangeably with the concept of information, plays an increasingly crucial role as organizational resource nowadays due to its the widely recognized importance. Whereas information equals a certain flow of messages, knowledge is the
combination of these flows in accordance with the mindset of the beholder (Nonaka, 1994). This highlights the aspect of human action that is associated with knowledge, as knowledge is both created through human action and simultaneously affects human action. As such, due to its dynamic character, knowledge can be considered both an outcome (framework of ‘the truth’) as well as a process (absorbing new information) and is highly characterized by judgement and significance (Tskoukas & Vladimirou, 2001).
Nonaka (1994) states that these judgements are made at the level of the beholder, as it is the individual capability to draw distinctions. When the individual beholder is situated in a concrete context with additional members, the generation and creation of collective flows and eventually collective knowledge is taking place. An example of which is organizational knowledge, which is composed of the sum of knowledge generated by all employees within a certain context. Hereby the context refers to the organization (Spender, 1996; Tskoukas & Vladimirou, 2001). This implies that organizational knowledge is embedded between employees, rather than within, and thus embodies the collective brain and memory of the organization (Lam, 2000).
Over recent years organizational knowledge has gained increased attention due to its potential to create sustainable competitive advantage over time, due to its intangible nature. Hence it cannot be depleted, unlike physical assets (Collins, 1994; Grant, 1996; Teece, Pisano & Shuen, 1997; Alavi & Leidner, 2001; Subramaniam & Youndt, 2005). This has been found by Smith (2001), too. In her research, she states that 99 percent of the
11
same vein, Leonard-Barton (1992) argues that organizational capabilities are increasingly enabled through leveraging a set of knowledge as strategic resource. He defines four dimensions of this set, which are elaborated on below.
The first element is Tacit Knowledge. This revolves around people’s experiences and insights, implying the “know-how” element of information. As such, tacit knowledge is local, intuitive, unarticulated, contextual and manifested implicitly (Lam, 2000). The second
dimension is Explicit Knowledge. This consists of “know-what” or “know-about” information and is either codified or quantified. Rather than tacit knowledge, explicit knowledge is formal, systematic, technical and requires a level of academic understanding (Smith, 2001). Hence the main difference between tacit knowledge and explicit knowledge lies in the (ease of) transferability and the mechanisms of transferring. Tacit knowledge is exposed by means of active communication, whereas explicit knowledge is revealed through application (Grant, 1996; Lam, 2000; Smith, 2001). Important to mention is that a balance of both is needed in order to organize tasks (explicit knowledge) and foster creative
transformation (tacit knowledge) (Spender, 1996).
The third dimension is Managerial Systems. These are the management processes and activities, defining the way of creating and controlling of knowledge. According to several scholars, the process of managing knowledge predominantly aims for one of three things: raising awareness and transparency around knowledge, encouraging the sharing of knowledge or building an infrastructure for knowledge. All three processes should, at a minimum, consist of creating - storing - transferring - applying knowledge. When done in an effective manner, it enables a firm to forcefully compete while exploiting their existing knowledge and exploring the creation of new knowledge (Nonaka, 1994; Davenport & Prusak, 1998; Alavi & Leidner, 2001). Generally speaking, managerial systems are divided into technological infrastructure and organizational structure. Technical systems determine
12
how knowledge travels throughout an organization and how one can access it. Organizational structure determines the interactions among employees, which are affected by the level of organizational hierarchy. This implies that managerial systems either lean towards a higher degree of centralization and formalization or a higher degree of decentralization and self-organization, although exceptions and variations exist (Gold et al., 2001).
The last element is Values + Norms, which are generally considered to be the building blocks of the organizational culture. According to Schein (1984, p. 3), organizational culture refers to “the pattern of basic assumptions that a given group has invented (...) and that have worked well enough to be considered valid, and, therefore, to be taught to new members as the correct way to perceive, think, and feel in relation to those problems”. In other words, values are embodied in norms which in turn shape practical activities (Ipe, 2003). This implies that values + norms assign importance to the content and structure of knowledge, the means of collecting knowledge and the means of controlling knowledge in accordance with what the company deems to be relevant, convenient and of significance (Leonard-Barton, 1992; Gold et al., 2001).
This division in organizational knowledge is in line with Nonaka’s (2000, p. 15) perception of knowledge, arguing that it is “a multifaceted concept with multilayered meanings”. Important to mention is that the four dimensions are interrelated, meaning that managerial systems or values + norms can affect tacit knowledge as well as explicit
knowledge both positively and negatively. These interlinkages will be discussed in chapter three, as they are part of the hypotheses.
2.2 Dynamic Capabilities
With a change in geographical events such as globalization and the rapid rise of technology, the RBV is starting to lose its ability to adequately explain a firm’s performance.
13
Organizations nowadays often require more than resources that are simply valuable, rare, inimitable and non-substitutable. They need that type of capabilities, too (Eisenhardt & Martin, 2000; Teece, 2007). To describe these capabilities that are difficult to replicate, Teece et al. (1997, p. 501) introduced the notion of dynamic capabilities. This refers to “the firm's ability to integrate, build, and reconfigure internal and external competencies to address rapidly changing environments”. Exploited in its most effective way, dynamic capabilities have the potential to keep the firm’s prior accumulated resources and capabilities
continuously relevant and in doing so, positively regulate the implementation and adoption of new or improved organizational practices. All in order to meet the current and future needs of the market, while in pursuit of evolutionary fitness (Teece, 2007).
2.2.1 Collaborative Learning
Building upon knowledge being the main facilitator and enabler of organizational capabilities and the importance of dynamic capabilities in today’s ever-accelerating markets, the current study introduces collaborative learning as the capability that keeps existing knowledge relevant and engages employees to generate multifaceted knowledge together. Collaborative learning is also referred to as “a situation in which particular forms of interaction among people are expected to occur, which would trigger learning mechanisms, but there is no guarantee that the expected interactions will actually occur” (Dillenbourg, 2007, p. 5). It is accordingly a voluntary case of collectively exploiting (sharing) and exploring
(leveraging/learning) information, by mobilizing knowledge in complex settings. This
double-looped process of information exchange was initially developed as a teaching method with the idea of engaging students more deeply by grouping them together for academic purposes (Bruffee, 1984; March, 1991; Argyris, 1995). Nowadays it is more generally used in the sense of continuously “defining, analyzing, negotiating, organizing, implementing and
14
reflecting on activities” together (Keen, Brown & Dyball, 2005, p. 105). This can be applied to academic institutes, but can also be considered in an organizational context, as bringing together people with different backgrounds will lead to increased efficient decision making (Van den Bossche, Gijselaers, Segers & Kirschner, 2006).
Closely linked to the concept of collaborative learning is knowledge sharing. Knowledge sharing is commonly defined as the interorganizational process of exchanging information, data, experiences and/or insights among employees (Tsai, 2002; Lin, 2007). However, the main difference between knowledge sharing and collaborative learning deals with the fact that the latter is a continuous and collective process of transferring, assimilating and leveraging knowledge and thus has a more active, hybrid and bi-directional nature (Lee, 2001). Furthermore, Yin (2007, p. 83) argues that “knowledge sharing enables managers to keep the individual learning flowing throughout the company and to integrate it for practical applications”. Knowledge sharing can thus be seen as a (crucial) first step, after which the collaborative learning by leveraging organizational knowledge takes place (Cash et al, 2003). Learning hereby refers to the ability to encode “inferences from history into routines that guide behavior” (Levitt & March, 1988, p. 391). Hence the current research considers collaborative learning as a social process of knowledge building, which includes both the aspect of exploitation (existing knowledge) as well as exploration (new knowledge) between two or more employees within an organization (Van den Bossche et al., 2006).
2.3 Environmental Management Practices
The past three decades, after the Brundtland report of 1987 on sustainable development, Corporate Social Responsibility (CSR) has started to emerge on the policy agenda of an increasing amount of companies in order to achieve sustainable production. CSR is
15
and environmental concerns into the day-to-day business decisions of an organization (Porter & Kramer, 2006; Aguinis & Glavas, 2012). According to Elkington (1998), CSR is closely related to realizing economic value, as businesses can generate profit while addressing social and environmental issues at the same time. This is also referred to as the Triple Bottom Line (TBL), consisting of the people, planet and profit, and urges businesses to strive for
sustainable development. CSR can thus be seen as the management practices that aim to meet the TBL, by prioritizing the issues involved in order to establish new working routines
(Burke & Logsdon, 1996; Porter & Kramer, 2011). Of interest in the current study is the planet dimension, which involves the environmental concerns that a firm could/should address.
One way sustainable environmental development can be achieved, is by implementing an EMS as a new working routine (Coglianese & Nash, 2001). This involves the management practices that seek to reduce the company’s environmental footprint according to
international standards. These standards are based upon the ISO 14001 certification, consisting of guidelines for organizational environmental activities. Hence the aim of an EMS is to continuously review, evaluate and improve the organization’s environmental impact in order to regulate its environmental performance in the best possible way (Darnall & Edwards, 2006). According to the Environmental Protection Agency (2017), an EMS consists of the following basic elements of practice: analyzing the environmental impact, setting environmental objectives, monitoring and measuring progress and ensuring employees’ environmental awareness. Environmental management practices are accordingly “the
techniques, policies and procedures a firm uses that are specifically aimed at monitoring and controlling the impact of its operations on the natural environment” (Montabon, Sroufe & Narasimhan, 2007. p. 998). Such practices can either be defensive (e.g. end-of-pipe
16
prevention technologies) (Christmann, 2000). The present study looks at the sum of various environmental management practices rather than simply the ISO certification, as research shows that companies with green management not always possess the certification and vice versa (Montabon et al., 2007). Thus, by focusing on the management practices that aim to change the organization along a more sustainable path, a comprehensive understanding of environmental management is gained. Especially given the fact that greenwashing has been found to be closely associated by discourses of certifications (Boiral & Gendron, 2011).
17
chapter three Theoretical Framework
This chapter presents the theoretical arguments found in earlier research that support the eight hypotheses of the current study.
3.1 Hypothesis One
Many scholars found that ensuring innovation (and thereby competitive advantage) within an organization is highly dependent upon the availability of knowledge (Grant, 1996; Spender, 1996; Cabrera & Cabrera, 2002; Bollinger & Smith, 2001). Innovation is established through the linkage between knowledge and creativity, whereby creativity empowers the
transformation of knowledge X to knowledge Y in order to create new ideas (Amabile, 1988; Carneiro, 2000). Knowledge X therefore can be seen as the building block of a creative process and Knowledge Y the starting point of a new innovation (Weisberg, 1999; Subramaniam and Youndt, 2005; Chen & Huang, 2009). Thus, given that organizational knowledge is a critical factor affecting an organization’s potential to ensure innovation, it means that knowledge can be considered a strategic asset that could create competitive advantage (Spender, 1996; Bollinger & Smith, 2001; Volberda, Foss & Lyles, 2010).
According to Barney (1991), for an asset to be considered strategic it must be valuable, rare, inimitable and non-substitutable (also referred to as the VRIN framework). Organizational knowledge meets those characteristics in the following ways. Knowledge can be considered valuable as it reflects the internal and external interactions adaption of an organization, which eventually have resulted or will result in improved products, processes and services and is thus “intrinsically valuable for its activity” (Nelson & Winter, 1982; Cabrera & Cabrera, 2002, p. 690). Secondly, organizational knowledge can be considered rare due to the fact that it is person- and path-dependent. This implies that knowledge resides
18
within employees and that no two organizations will ever possess the same amount and type of knowledge, simply because they consist of different employees and different ways of working. This applies to both the past and the present and thus the way they evolved over time (Spender, 1996; Volberda et al., 2010). Thirdly, organizational knowledge can be considered inimitable since knowledge is based on employees’ own interpretations and norms, whether it is written down or spoken about (Cabrera & Cabrera, 2002). This implies that organizational knowledge is causally ambiguous, which means that it is embedded in a network of interorganizational and interpersonal relationships. As a result, a certain piece of information might not be important to a manager from organization A, while manager from organization B finds it crucial data in accordance with the organizational beliefs (Sanchez & Heene, 1997). In a similar vein, the non-substitutable aspect refers to the replication of a specific asset. As organizational knowledge equals the sum of all employees’ knowledge, it follows that a similar synergy of all experiences, insights, manuals or routines within an organization is very hard to either create or replicate. Thus, given the highly individual nature of organizational knowledge, it means that it has no strategic equivalents that can offer the same potential (Meso & Smith, 2000; Alavi & Leidner, 2001; Bollinger & Smith, 2001). Important to mention is that applies to private knowledge only, as public knowledge is available for and accessible to anyone and thus not valuable, rare, inimitable or
non-substitutable (Matusik & Hill, 1998). Based on these four characteristics, knowledge can be considered a strategic asset and thus has the potential to ensure innovation (and eventually competitive advantage) (Volberda et al., 2010). The way this applies to environmental management practices is the following.
The adoption of environmental management practices is essentially a transformation from ‘regular’ practices towards (more) sustainable ones, by integrating environmental management activities into the current business model (Darnall & Edwards, 2006; Hahn,
19
Pinske, Preuss & Figge, 2015). This can thus be seen as a form of process innovation,
whereby the organization seeks to revolutionary change the way it is designed and structured along a more sustainable path (Davenport, 1993; Carneiro, 2000). The starting point hereby is knowledge X (regular practices), which are transformed into knowledge Y (environmental management practices) in order to improve the organization for the better (process
innovation). This has been highlighted by many environmental scholars, too. As the adoption of an EMS is a highly dynamic innovation process, a constant stream of information on environmental issues that is required to ensure the adoption of an EMS (Bloodgood & Salisbury, 2001; Zobel & Burman, 2002; Cash et al., 2003; Darnall & Edwards, 2006). This is mainly due to the complex and ambiguous nature wherein many environmental issues reside, which increases the dependence upon knowledge in order for a firm to ensure
successful change (Lichtenthaler, 2009). Hence knowledge as strategic asset leveraged in its most effective way positively affects the adoption of environmental management practices as innovation process (Teece, 1986; Christmann, 2000).
This implies that tangible knowledge (explicit knowledge) is needed to measure environmental impacts, make statistical analyses on waste audits, get information on
pollution procedures, control environmental inventory and manage sustainable procurement. However, in order to inform, teach, train, raise awareness, establish environmental ethics and propagate the firm’s environmental policy intangible knowledge (tacit knowledge) is needed in the form of employee’s experiences and insights (Boiral, 2002; Reed, 2008; Raymond, Fazey, Reed, Stringer, Robinson & Evely, 2010). As such, the following first set of hypotheses is proposed.
H1a: Tacit Knowledge is positively related to Environmental Management Practices. H1b: Explicit Knowledge is positively related to Environmental Management Practices.
20
3.2 Hypothesis Two
In addition to the direct relationship between organizational knowledge (tacit and explicit), a mediating role for collaborative learning is proposed. When it comes to interorganizational collaboration, it has been found that collaborative learning as a behavioural skill explains how organizations use their existing knowledge in a more efficient way in order to come up with synergistic solutions (Hardy, Philips & Lawrence, 2003). The way this happens is as follows. According to Bower and Hilgard (1981), any type of organizational learning can only take place in combination with accumulated prior knowledge, as one simply cannot recognize or assimilate new knowledge without having a foundation of knowledge in place. Hence the more existing knowledge residing within a company, the more collaborative learning within that company. As a result of the knowledge mobilization by means of
collaborative learning, a more interdependent amount of knowledge is created, which helps to address and solve issues in a holistic way (Choi & Lee, 2002). This is done by exploitation of old practices (refinement) and exploration of new possibilities (change) in a continuous and collaborative manner, so that organizational knowledge can be used to its full potential (March, 1991; Gibson & Birkinshaw, 2004). Cohen and Levinthal (2000) even argue that an organization’s ability to recognize, assimilate and apply the value of existing and new information is inherent to its ability to ensure innovation.
The way this applies to the adoption of an EMS is that old practices equal the regular structure of the organization and new possibilities equal the environmental management practices. This implies that collaborative learning plays a crucial role in an organization’s aim to adopt an innovative initiative, in this case being green management. Especially given the environmental context wherein these new possibilities reside. Several scholars found that individuals throughout the world start to have difficulties in identifying wasteful processes as the number of environmental challenges is growing and accordingly the ways they are linked
21
to each other. Environmental issues thus become meta-problems (i.e. interconnected problems) and environmental management becomes increasingly ambiguous and complex (Allen, Bosch, Kilvington, Oliver & Gilbert, 2001; Reed, 2008). It implies that for knowledge to be more comprehensible and environmental management to be more successful, a dynamic and context-dependent capability in the form of collaborative learning is required
(Lichtenthaler, 2009). The way this happens is that organizational knowledge, when shared and reflected on among employees from various departments, results in a more integrated knowledge set is created that fosters the assessing, reporting, solving and improving of environmental practices and the complexity thereof (Cohen & Levinthal, 2000; Allen et al., 2001; Raymond et al., 2010).
Furthermore, as described earlier, collaborative learning was initially developed as a way to engage students more deeply in various academic activities (Bruffee, 1984). However, the same logic applies to employees and business practices (Argyris, 1995; Schein, 1999). It has been found that stakeholder management is crucial in the adoption of a proactive environmental strategy. This applies specifically to primary stakeholder groups, which have a significant impact on determining the success or failure of adopting green management practices (Buysse & Verbeke, 2003). Of the primary stakeholder group, it is argued that employees could yield the most benefits from satisfactory environmental management practices, as they provide the resources (knowledge) that are intrinsically tied to the firm. This means that their engagement and participation are central to the adoption of an EMS (Hart, 1995; Sharma & Vredenburg, 1998; Reed, 2008). Henriques and Sadorsky (1999) even argue that employees are the source of a company’s environmental success. The way this takes place is that employees showing collaborative learning behaviour discuss and reflect on current organizational practices, which in turn foster the adoption of environmental
22
explains socially inclusive and environmentally sustainable decision making (Kapoor, 2001). This leads to the second set of hypotheses.
H2a: Collaborative Learning mediates the relationship between Tacit Knowledge and Environmental Management Practices.
H2b: Collaborative Learning mediates the relationship between Explicit Knowledge and Environmental Management Practices.
3.2 Hypothesis Three
The current study introduces the concept of managerial systems, consisting of technological infrastructure and organizational structure, as moderating variable affecting the strength of the relationship between organizational knowledge and collaborative learning. The
technological infrastructure is hereby closely related to self-organization and the organizational structure to decentralization.
According to Gold et al. (2001), managerial systems first and foremost affect this relationship by investing in information technology (IT) support and accordingly assuring its implementation. The way this happens is that by providing the organization with a
technological infrastructure (also known as an intranet), employees are better able to access, store, combine and coordinate previously disintegrated data and pieces of information (Lee & Choi, 2003; Kim & Lee, 2006). Accordingly, this is extremely important to enable
collaborative learning, as a functional IT system allows employees to both save their existing knowledge (exploit) and generate new knowledge online (explore) (March, 1991; Gold et al., 2001). Furthermore, the availability of technological communication systems implies easier communication and increased feedback loops by eliminating geographical barriers that may have initially withheld or disabled employees from storing and/or sharing their knowledge in the first place (Hendriks, 1999). Technology can thus be seen as a facilitator of collaborative
23
learning, as it allows employees to work together and collaborate intensively, regardless of time and place. This applies both to tacit knowledge (tools to facilitate communication) as well as explicit knowledge (online library with a hierarchy of databases) (Alavi & Leidner, 2001). However, researchers stress the importance of technological infrastructure on the exploitation and exploration of explicit knowledge. The reason being is that explicit
knowledge is in general predominantly transferred and stored by means of technology, which makes the relationship between explicit knowledge and collaborative learning more
dependent upon the availability of a technological infrastructure (Dhanaraj, Lyles, Steensma & Tihanyi, 2004). This in comparison to tacit knowledge, whereby generation and transfer partly happen through technology and mostly through body language and physical
demonstration (Leonard & Sensiper, 1998; Lam, 2000). However, with recent innovations such as whiteboarding and video conferencing, IT is increasingly supporting the process of creating and sharing tacit knowledge (Lee & Choi, 2003). This implies that both are related to a well-functioning technological infrastructure (Smith, 2001).
In order to leverage the technological architecture of a firm, organizational structure is required to enable employees to leverage their and other’s knowledge (Hendriks, 1999; Gold et al., 2001). Organizational structure is commonly defined as the formal way of allocating and coordinating tasks, which are controlled and integrated by administrative mechanisms. Although organizational structures mostly have the intention to support different business units and functions within the organization, it often occurs that centralized structural elements unconsciously hinder or inhibit interorganizational collaboration. On the other hand,
decentralized organizational structures, with as little structural elements as possible, can positively contribute to interorganizational collaboration (Gold et al., 2001; Lee & Choi, 2003). For instance, a decentralized structure has found to have a positive effect on the interorganizational sharing and communication among employees (Child, 1972; Kim & Lee,
24
2006). The reason for this is that a smaller number of corporate layers supports spontaneity and experimentation, which are critical to knowledge creation and mobilization (Nonaka, 2000). A decentralized structure simultaneously accelerates the process of storing and sharing knowledge, as information has to go through very little to none bureaucratic levels in order to eventually reach either a database or a colleague (Lee & Choi, 2003; Ismail Al-Alawi, Yousif Al-Marzooqi & Fraidoon Mohammed, 2007). In a similar vein is research by Ipe (2003), who argues that the relationship with the recipient determines the process of sharing. This implies that a close relationship (which mostly happens between employees of similar status, i.e. decentralized structure) reduces the amount of information screening beforehand and
increases the amount of knowledge being shared. This indicates that a decentralized structure has a non-hierarchical and self-organizing design, which fosters the creation, storing and sharing of knowledge (Tsai, 2002; Lee & Choi, 2003).
Whereas this applies to both types of knowledge, tacit knowledge specifically requires a high degree of flexibility due to its intangibility and complexity. Hence this can be heavily affected by a hierarchical structure and high degree of centralization, for the reason that the lack of openness negatively influences trust which hampers sharing and learning (Grant, 1999; Smith, 2001). Both are nonetheless positively affected by an open and decentralized organizational structure (as measured in this study). This leads to the third set of hypotheses.
H3a: Managerial Systems have a positive moderating effect on the relationship between Tacit Knowledge and Collaborative Learning, so that this relationship is stronger for higher levels of Managerial Systems.
H3b: Managerial Systems have a positive moderating effect on the relationship between Explicit Knowledge and Collaborative Learning, so that this relationship is stronger for higher levels of Managerial Systems.
25
3.3 Hypothesis Four
The current study introduces the concept of values + norms as moderating variable affecting the strength of the relationship between organizational knowledge and collaborative learning. It has been found by numerous scholars that perhaps the most important facilitator of
interorganizational collaboration is the organizational culture, as it is central to an
organization’s ability to manage the knowledge residing within employees (Leonard-Barton, 1992; Nonaka and Takeuchi, 1995; Alavi & Leidner, 2001; Gold et al., 2001; Ipe, 2003). According to Schein (1984), organizational culture refers to the shared values and norms.
Research indicates several ways by which organizational culture affects the process of exploitation and exploration of knowledge. First, culture shapes assumptions regarding knowledge and its importance. Thus, culture implies what knowledge to generate and/or communicate and what not depending on what the organization considers valuable and meaningful. For instance, hotels may tend to give priority to knowledge that revolves around customer experience, while supermarkets might stress the importance of supply chain
knowledge (De Long & Fahey, 2000; Ipe, 2003). It follows that when an organization underlines the importance and encourages all types of knowledge to meet specific goals, it positively contributes to the process of generating, assimilating and levering knowledge (Gold et al., 2001; Lee & Choi, 2003). Secondly, organizational culture revolves around trust. Research indicates that trust influences both the generation of, sharing of and learning from knowledge, by mitigating the fear or risk of losing one’s individual value that comes with information creation and exchange. This means that a high degree of trust, which is formed through the encouragement of experimentation and participation, empowers employees to create knowledge and share this with other employees, rather than keep it the individual level (Nonaka, 2000). Important hereby is that trust simultaneously allows employees to make mistakes, which fosters the learning aspect (Jarvenpaa & Staples, 2001; Lee & Choi, 2003).
26
Furthermore, culture creates a context for the social interaction around knowledge (De Long & Fahey, 2000). Following from this, organizational culture is created and maintained through socialization among employees from various departments and levels. This implies that when managers take the time to create and share their knowledge, they not only
contribute to the overall amount of holistic knowledge, but also set an example to value the role of (constructive) knowledge as a crucial and highly appreciated part of the organizational performance (Gold et al., 2001). This type of managerial involvement and dedication will encourage and empower employee interaction throughout the entire company to mobilize their knowledge, in order to create an integrated set of information (Ipe, 2003; Cabrera & Cabrera, 2005). Culture thus shapes the entire process of creating, sharing and applying knowledge by having and actively exhibiting a corporate vision and objectives (De Long and Fahey, 2000; Gold et al., 2001; Lee & Choi, 2003). Employees are accordingly significantly more likely to share and give feedback in an open culture with a high degree of trust, support and involvement, where the role of knowledge is valued and appreciated throughout the entire organization (Connelly & Kelloway, 2003; Cabrera & Cabrera, 2005).
The effect of culture on the degree of sharing applies both to tacit as well as explicit knowledge. However, it is argued that tacit knowledge is more strongly connected to the transformational leadership element of organizational culture. The reason being is that
creating, assimilating and sharing tacit knowledge rarely happens without active participation and coordination of the beholder (Nonaka and Takeuchi, 1995; Lam, 2000). As such, a leader that openly supports dialogue and interaction thereby positively affects the role of tacit knowledge within the organization. However, explicit knowledge is mostly linked to the objectives element. When an organization stressed the importance of all types of knowledge, especially in combination with the mentioning of specific goals or objectives, it positively affects the mobilization of explicit knowledge (Smith, 2001). This implies that the
27
relationship between tacit/explicit knowledge and collaborative learning are both fostered by the culture residing within the organization (Gold et al., 2001; Ipe, 2003; Alavi et al., 2005). This leads to the fourth and last set of hypotheses.
H4a: Values + Norms have a positive moderating effect on the relationship between Tacit Knowledge and Collaborative Learning, so that this relationship is stronger for higher levels of Values + Norms.
H4b: Values + Norms have a positive moderating effect on the relationship between Explicit Knowledge and Collaborative Learning, so that this relationship is stronger for higher levels of Values + Norms.
3.4 Conceptual Model
28
chapter four Research Context
In case of not having sufficient ingredients (resources) or cooking expertise (capabilities) for the earlier mentioned lasagne dish, individual X could be redirected to one of the many Italian restaurants.
Responsible for any meal that is consumed out of home is the food service industry, consisting of the following six subsets: full-service restaurants, fast food restaurants,
cafes/bars, street stalls/kiosks (also referred to as ‘on the move’ sales points), cafeterias and 100% home delivery/takeaway. Below the total amount of food service sales (Table 1) as well as the sales per subsector over the period 2011 - 2016 (Table 2) can be found.
Table 1: Food service sales in The Netherlands, in million EUR. Source: FAS, 2017
2011 2012 2013 2014 2015 2016
Total food service sales (€) 9.521 9.649 9.830 10.041 10.470 10.850
Table 2: Food service sales in The Netherlands per subsector, in million EUR. Source: FAS, 2017
2011 2012 2013 2014 2015 2016
Full-service restaurants (€) 3.725 3.791 3.858 3.949 4.143 4.337
Fast food restaurants (€) 2.672 2.698 2.766 2.830 2.927 3.002
Cafes/bars (€) 2.088 2.093 2.090 2.102 2.168 2.281 Street stalls/kiosks (€) 0.199 0.204 0.206 0.207 0.211 0.217 Cafeterias (€) 0.558 0.572 0.600 0.615 0.630 0.574 100% home delivery/takeaway (€) 0.281 0.294 0.311 0.339 0.393 0.441
Annual growth rates of all six establishments were between + 3 - 5%, except for cafeterias (- 9% in 2016) and home delivery/takeaway (+ 12% in 2016). Similar growth figures are expected for the upcoming years, with an even stronger rise in home delivery/takeaway due to increased customer demand for convenience and flexibility (Foreign Agricultural Service (FAS), 2017; Euromonitor, 2018; Foodstep, 2018). The food service industry is, in turn, a
29
subset of the food and beverage industry and accounts for approximately one-third of the total food turnover in The Netherlands. The remaining two-third can be attributed to the retail sector, involving supermarkets and smaller chains (Berkhout, 2017; Foodstep, 2018).
Whereas growth figures in food service industry are seeing a significant rise, so is its footprint. As a result of Kyoto Protocol in 2005, aiming to fight global warming by means of making sustainable development part of daily business operations, the food service industry is gaining increased attention with regards to its contribution on the environmental burden. According to the National Institute for Public Health and the Environment (Rijksinstituut voor Volksgezondheid en Milieu (RIVM), 2016), the Dutch food service and beverages industry negatively affects the environmental degradation by emitting a significant amount of greenhouse gases and thereby reducing biodiversity. This result follows from an analysis of six environmental issues, being: climate change, land use, water use, acidification,
eutrophication and soil depletion. Furthermore, the biggest determinant of the environmental burden is the wastage of food, resulting in a negative impact on the use of water and land, energy consumption and CO2 emissions. This is in line with a study of the Pacific Gas & Electric’s Food Service Technology Center (2016) on the footprint of the food service industry. It states that “restaurants are the retail world’s largest energy user. They use almost five times more energy per square foot than any other type of commercial building (...)”. Following from this, it is becoming clear that sustainable development in this industry could make a significant positive contribution to the environment, as it is currently the least sustainable industry in the world (Maloni & Brown, 2006; Parfitt, Barthel & Macnaughton, 2010; Wang, Chen, Lee & Tsai, 2013; RIVM, 2016). Hence the Dutch food service industry forms the research domain of this study.
30
chapter five Research Design
In this chapter, the empirical setting of the study is discussed. This includes the method, the data collection and the measurement constructs of the various variables.
5.1 Method
For this study a quantitative - rather than qualitative - research method has been chosen in order to find empirical evidence for the research question and hypotheses, by using a cross-sectional survey method. This survey was explanatory and descriptive in nature, aiming to analyze and clarify the relationships between the different variables represented in the conceptual model. Hence a deductive approach is maintained to test the relationships based on existing theory. The survey consisted of standardized measures of the independent, dependent, mediator and moderator variables verified and corroborated by various scholars. However, some constructs were shortened in order to keep the questionnaire proportional and achieve a higher response rate (Baruch & Holtom, 2008). Details of the shortening can be found in chapter six. The questionnaire was available in Dutch as well as in English and translated by means of direct translation, done by the researcher. The Dutch translation was then checked and slightly altered by an English-Dutch native, in order to optimize validity of the Dutch version. The survey can be found in Appendix 1. All questions were to be answered by means of a Likert scale from 1 = firmly disagree to 5 = firmly agree (Jamieson, 2004). Questions regarding the company size, employee’s gender, age, tenure, (see paragraph 5.7), type of company, and whether the participants have managerial responsibilities consisted of either a multiple-choice or drop-down answer field.
31
5.2 Data Collection and Sample
In order to test the hypotheses mentioned in chapter three, an online survey questionnaire has been generated and distributed. The advantages of this research method revolve around both cost efficiency as well as time efficiency, accessibility, scalability and anonymity. The drawbacks, however, come with the limit possibility of answers due to the Likert scale and the chance of misinterpreting the pre-formulated questions (Saunders & Lewis, 2004). The questionnaire was designed by use of Qualtrics, an online survey software provided by the University of Amsterdam to all its students. Hence all acquired data could easily be imported into SPSS, a software package for statistical analysis.
As the research domain of the current study involved the Dutch food service industry, the sample included anyone who is currently working in either a full-service restaurant, fast food restaurant, cafe/bar, street stall/kiosk or 100% home delivery service/takeaway. The distribution of the survey followed a probability sampling technique, by sending the survey link to all 935 food service establishments in and around Amsterdam, of which a list was supplied by the municipality. After listing all available email addresses of the establishments still in operation, a total of 807 surveys was successfully sent out. The aim of the study was to get a response rate of approximately 20%, which is based on earlier research analyzing
surveys distributed by email (Fox, Crask & Kim, 1988). However, the response rate accounted for approximately 15% (n = 118).
5.3 Independent Variables Tacit and Explicit Knowledge
The current study considered tacit and explicit knowledge, both subsets of organizational knowledge, as independent variables. Tacit knowledge refers to “know-how” information, consisting of insights and experiences. Explicit knowledge, on the other hand, refers to
32
“know-what” information and consists of information that can be codified (Smith, 2001). Both concepts were measured according to constructs designed by Lee and Choi (2003), who did several empirical studies on (the enablers of) knowledge management, knowledge
creation, corporate performance and the relationship between.
The reliability of their measurement instruments was tested by means of Cronbach’s alpha, whereby the value for both tacit knowledge (ɑ = 0.794) as well as explicit knowledge (ɑ = 0.829) exceeded the 0.7 threshold (Gliem & Gliem, 2003). Both constructs were adapted from existing literature and pilot-tested. The 4-item construct for tacit knowledge measured the acquisition of knowledge through co-workers and mentoring, face-to-face advice from experts and the use of informal dialogues. The construct for explicit knowledge measured the acquisition of knowledge through formal documents, the documentation of knowledge and the codification of knowledge. However, as one of the items of measuring explicit knowledge revolved around sharing (i.e. mediator in the current study), it was deleted from the scale. To guarantee an equally large scale for both independent variables, an extra construct was developed based on well-cited research from Nonaka, Toyama and Konno (2000) and an empirical study from Dhanaraj, Lyles, Steensma and Tihanyi (2004) (ɑ = 0.899).
Accordingly, it states the acquiring of knowledge by analyzing published literature and procedural documents. A detailed description of all eight items can be found in the appendix.
5.4 Dependent Variable
Environmental Management Practices
Environmental management is commonly defined as a way to meet and tackle the challenges in the ever-accelerating world of resource depletion (Coglianese and Nash, 2001). In order to test this, most empirical studies looked at corporate environmental attitudes and
33
an ISO 140001 certification. This consists of practical tools to operationalize corporate environmental behaviour (EPA, 2017). This study, however, took a different approach by drawing upon work from Montabon et al. (2007) who exemplified environmental
management as the sum of various environmental management practices. Hence, they define environmental management as an evolution of organizational behaviour rather than simply the implementation of pre-defined tools. However, as their construct exists of 48 items, the 12 items that are most closely linked to an EMS as designed by EPA are used in order to for the questionnaire to maintain proportionally and ensure a higher response rate (Baruch & Holtom, 2008). Examples of environmental management practices include developing an environmental mission statement, recycling, reducing waste and packaging. The reliability of the construct was measured according to Cronbach’s alpha, whereby the value showed ɑ = 0.77. A detailed description of the construct (12 items) can be found in the appendix.
5.5 Mediator Variable Collaborative Learning
The concept of collaborative learning has up until today mainly been researched in a qualitative manner. Collaborative learning refers to the process of “defining, analyzing, negotiating, organizing, implementing and reflecting on activities” together (Keen et al., 2005, p. 105). However, as the current study follows a quantitative approach, empirical studies regarding related concepts were considered in order to develop an appropriate measurement scale. These consist of the following: knowledge sharing, interorganizational communication, organizational learning and absorptive capacity. Hence, by combining constructs, the continuous loop of sharing and learning together (i.e. exploiting and exploring knowledge - collaborative learning) was measured and the multidimensionality of the concept was captured. These studies included Goh and Richards (1997) and Lichtenthaler (2009), of
34
which respectively four and six items were used. The corresponding values for the
Cronbach’s alphas were ɑ = 0.9 and ɑ = 0.88. A detailed description of the construct and the source of every item (10 in total) can be found in the appendix.
5.6 Moderator Variables Managerial Systems
The first moderating variable in the current study is managerial systems, which refers to the internal design and administration of an organization. According to Gold et al. (2001), this is dependent upon three types of infrastructure, being technology, structure and culture. Given the crucial and infused role of culture, this element will be treated as a separated moderator (Leonard-Barton, 1992). The constructs used to measure the technological and structural infrastructure of the firm are designed by Lee & Choi (2003), who tested technology based on IT support (five items) and structure based on decentralization (five items).
Examples of IT support include online communication and simulation tools. Examples of decentralization include the freedom to make own decisions and the absence of hierarchy. Reliability was tested by means of Cronbach’s alpha, being ɑ = 0.86 and ɑ = 0.85
respectively. A detailed description of the construct can be found in the appendix.
Values + Norms
The second and last moderating variable of the study consists of values + norms, which is commonly referred to as organizational culture. According to Schein (1984), organizational culture is defined as the beliefs that are embodied within the organization and accordingly shape the environment of the organization. This is measured through a construct designed by Gold et al. (2003), consisting of 12 items. Examples include understanding the importance of knowledge, the level of participation in capturing knowledge and the encouraging of
35
exploring and experimenting. The original construct existed of 13 items, of which one was deleted as it revolved around sharing outside of the industry. Cronbach’s alpha showed ɑ = 0.83.
5.7 Control Variables
Several studies have recognized the difference between small and large firms in the process of adopting sustainable development, or more specifically - environmental management practices. For instance, it has been found that a positive relationship exists between firm size and the degree of sustainable innovation (Hockerts & Wüstenhagen, 2009). Furthermore, large firms are more likely to engage in environmental activities and to implement environmental training programs for its employees. The reason behind this lies in the environmental budget an organization possesses as well as the stakeholder pressure it perceives (Aragón-Correa, 1998; Darnall, Henriques & Sadorsky, 2009). The first control variable in the current study is thus company size, measured by the number of employees working within the organization.
The second control variable revolves around knowledge sharing and its determinants. Scholars have recognized the fact that several personal characteristics of the individual (in this study the manager of the food service establishment) have the potential to influence the process and perspective of knowledge sharing. The reason behind this lies in the fact that people tend to communicate with people who are closely alike. Furthermore, different communication styles have been found among females and males. When it comes to age, research shows that older employees generally have more knowledge and thus more to share. However, in the case of a technological communication system present, older employees appear to share less in comparison to their younger colleagues. Tenure affects sharing and learning in a way that new employees transfer more and more frequently than employees who
36
have been with the company for a longer period of time (Zenger & Lawrence, 1989; Connelly & Kelloway, 2003). Gender, age and tenure are accordingly controlled for.
5.8 Method of Analysis
In order to test the proposed hypotheses, three different methods of analysis were used. Firstly, to determine the relationships between all constructs within the study, a bivariate correlation analysis was employed. This was done to get an overall view of the interaction between the various constructs, including the control variables. Secondly, a multiple hierarchical regression was applied to predict the value of the dependent variable
(environmental management practices) based on the value of the independent variables (tacit and explicit knowledge) (Field, 2009). Finally, the mediating variable (collaborative learning) was measured according to the PROCESS model. The two moderating variables (managerial systems and values + norms) were analyzed using two separate moderation analysis
37
chapter six Data Analysis and Results
In this chapter, the results of the data analysis are presented. First, several measurement tests are discussed. These are followed by the descriptive statistics, after which the results of the regression, mediator and moderator analyses are disclosed.
6.1 Factor Analysis
An exploratory factor analysis was applied to identify groups or clusters of variables and thereby the similarities among the various questionnaire items, which are called factors (Field, 2009). In order to verify the sampling adequacy for the analysis, the Kaiser-Meyer-Olkin (KMO) test was applied. Bartlett’s test of sphericity was used to analyze the
correlations between the different measurement items. Lastly, by running an initial analysis, Kaiser’s criterion of eigenvalues >1 was employed to determine how many factors to extract. Tacit knowledge equaled KMO = .73, thereby exceeding the 0.6 threshold. Bartlett’s test of sphericity χ² (6) = 85.83, p <.001, indicated a high correlation between the different items. Explicit knowledge measured KMO = .684. Bartlett’s test of sphericity χ² (6) = 133.07, p <.001, again indicated a high correlation between the different items. For collaborative learning, it showed that KMO = .90 and Bartlett’s test of sphericity χ² (45) = 526.91 and p <.001. For managerial systems KMO = .81, with Bartlett’s test of sphericity χ² (45) = 613.04 and p <.001. However, both Kaiser’s test and the screeplot showed that two components had an eigenvalue over 1 and together explained 59.93% of the total variance. As managerial systems consisted of organizational structure (OS) items as well as items revolving around information technology (IT), the two factors show that managerial systems was measured as it was supposed to. See Table 3 below. Values and norms exhibited KMO = .806, with Bartlett’s test of sphericity χ² (66) = 622.64 and p <.001. Lastly, environmental management
38
practices showed KMO = .89. Bartlett’s test of sphericity χ² (68) = 822.43, p <.001, again indicated a high correlation between the different items.
Table 3: Factor Loadings MS
Item IT OS
Can take action without a supervisor -.22 .72 Are encouraged to make their own decisions .04 .59 Do not need to refer to someone .04 .71 Do not need to ask their supervisor before action .16 .79 Can make decisions without approval .08 .70 Provides IT support for collaborative works
regardless of time and place .83 -.01 Provides IT support for communication among
organization members .81 .07
Provides IT support for searching for and
accessing necessary information .86 .04 Provides IT support for simulation and
prediction .73 .03
Provides IT support for systematic storing .87 -.07
6.2 Reliability Analysis
In order to guarantee that measurement error is kept to a minimum and is consistent over time, a reliability analysis was conducted. As such, it measures whether the construct yields the same results under the same conditions. It is being done by means of Cronbach’s alpha, whereby a threshold of ɑ >.70 should be secured in order to ensure that the results are reliable (Field, 2009). For tacit knowledge, Cronbach’s alpha equaled ɑ = .71. Deleting the last item would result in .01 increase of ɑ. As that difference is rather insignificant, the measurement construct remained the same (Saunders & Lewis, 2014). Explicit knowledge showed ɑ = .78, with no increase if one item were to be deleted. For collaborative learning ɑ = .90.
Managerial system equaled ɑ = .846, whereas for values + norms ɑ = .89. Lastly,
environmental management practices resulted in ɑ = .93. This concludes that all measurement constructs exceeded the ɑ > .70 threshold and can thus be considered reliable.
39
6.3 Check For Normality
The degree to whether all variables follow a normal distribution was assessed by means of a normality check. According to George (2011), a normal univariate distribution is ensured when skewness and kurtosis are between 1 and 1. For tacit knowledge, skewness showed -.57 and kurtosis .02. Explicit knowledge equaled a skewness of -.07 and kurtosis of -.64. For collaborative learning skewness was -.63 and kurtosis .53. Managerial systems exhibited a skewness of -.12 and kurtosis of .32. Values + norms came with a skewness of -.57 and kurtosis of -.09. Finally, for environmental management practices, it indicated skewness of .03 and kurtosis of -.73. This means that all variables were within the -1 to 1 threshold.
6.4 Normality of Residuals
Another assumption of doing a regression analysis is that the residuals are normally distributed with a mean of zero (Field, 2009). In order to test the normality of residuals, a normal P-P plot and histogram were created. As can be seen in Appendix 2 and Appendix 3, the residuals approach normality. Furthermore, normality was checked by means of the Shapiro-Wilk test. Hereby it showed p >.005, which means that the null hypothesis of the residuals following a normal distribution was accepted. Lastly, the Durbin-Watson test to check for absence of autocorrelation between residuals showed 2.01. As this is between the 1.5 < DW > 2.2 threshold, the assumption was met.
6.5 Homoscedasticity
In order to test whether the variance of the residuals is homogenous, a scatter plot was constructed to get a visual representation of the predicted scores and the error of predictors (Field, 2009). In Appendix 4 the scatterplot, which shows that the variance is indeed homogenous as the distribution remains rather consistent, can be found.
40
6.6 Descriptive Statistics
Of the 807 questionnaires that were successfully sent out 118 have been finished, indicating a response rate of approximately 15%. The largest part of the respondents was female (66%), whereas three respondents either preferred not to disclose their gender. The average age of the respondents was 32.93 years. The average size of the company was between 11-50 employees and the average tenure of the respondents amounted to 3 years. With regards to the type of food service establishment: 71 respondents mentioned they are working in a full-service restaurant, 13 in a fast food restaurant, 30 in a cafe/bar, 3 in a street stall/kiosk and 4 at a 100% home delivery service. Of the 118 respondents, 70% stated they had managerial responsibilities within the company.
6.7 Correlations
To examine the relationships between all constructs, a bivariate correlation analysis was conducted. See Table 4 below for an overview.
Table 4: Means, Standard Deviations and Correlations
Variables M SD 1 2 3 4 5 6 7 8 9 10 1. Gender 1.66 0.55 - 2. Age 32.93 1.154 -.32** - 3. Size 2.34 125 .00 -.08 - 4. Tenure 5.30 2.35 -.35** .65** .06 - 5. Tacit Knowledge 3.63 0.82 .07 .16 .05 -.06 (.71) 6. Explicit Knowledge 3.04 0.95 -.02 .19 .28** .03 .45** (.78) 7. Collaborative Learning 3.54 0.78 -.04 .27** .10 .01 .78** .53** (.90) 8. Managerial Systems 3.27 0.73 -0.09 0.21 .16 -.04 .36** .35** .45** (.85) 9. Values + Norms 3.87 0.71 -.02 .27** -.11 -.08 .61** .55** .75** .48** (.93) 10. Environmental Management Practices 2.98 0.99 0.60 0.18 .15 .06 .40** .45** .45** .44** .47** (.93) Correlation is significant at the 0.01 level (2-tailed)**
