Financial and Managerial Knowledge Transfers to a Unicultural Recipient Group: The Case of the Lerd Program

(1)

1

Master Thesis for:

MSc International Business & Management – International Financial Manage ment MSc Business & Economics

Financial and Managerial Knowledge Transfers to

a Unicultural Recipient Group:

The Case of the Lerd Program

Author: Vladimir de Poel

Student numbe r: 1478834

Email: vladimirdepoel@gmail.com

Thesis Supe rvisor: Dr. B. J. W. Pennink Date/Place: March 2008, Groningen

University of Groningen

(2)

2

Abstract

This study examines the effects of source context, knowledge context, and relational context on the process of knowledge transfer in a unicultural student group. Specifically, I test the impact of capability, credibility, and communication on the process of knowledge transfer. Analyses revealed the main predicted effects and suggests that professors in higher educational institutions who have high task related capabilities and are perceived by their students as highly credible are able to transfer significant amounts of knowledge. Additionally, professors with high managerial capabilities transfer significant amounts of managerial knowledge and those with high levels of credibility transfer significant amounts of financial knowledge. No interaction effects were found, validating the interpretation of the independent variables in the multiple regression analysis. Interestingly, communication levels did not affect knowledge transfer.

(5)

5

Introduction

Since knowledge is becoming more and more a vital resource in obtaining and retaining the competitive advantage of a firm, and has emerged as the most strategically significant resource (Simonin, 1999), acquiring such knowledge is a primary strategic objective for many firms. Value can only be created by using this knowledge, and therefore it is imperative that this knowledge, whether financial or managerial be transferred to the relevant parties successfully (Joshi & Sarker, 2004). Various researchers in the field of organizational theory identify organizational learning as one of the most important research subjects and we can gain greater understanding of this topic by studying organizational knowledge flows (Gupta & Govindarajan, 2000). In order for firms to create a sustainable competitive advantage, organizations must be capable of creating and utilizing knowledge (Nonaka & Toyama, 2003) and knowledge transfer mechanisms must be fully understood and managed effectively.

A firm’s knowledge base has a greater ability, than any other resource a firm possesses, to serve as a source of sustainable differentiation, which leads to competitive advantage (Ibid). If competitive advantage is based on knowledge as the underlying primary resource, then ability of the owner of this knowledge to earn rents is determined by the transferability of this knowledge (Spencer & Grant, 1996). Hence when markets are “efficient” it is difficult to obtain a competitive advantage due to the widespread availability of relevant knowledge.

(6)

6

In this paper I will examine the extent of knowledge transfers from professors (hereafter referred to as source) at the University of Groningen, to a group of students (hereafter referred to as recipients) from Indonesia during a three-week local economic rural development course (Lerd). In particular, I will examine the extent to which financial and managerial knowledge and skills are transferred to this subject group, and identify the inhibiting or enabling factors. I am studying individuals because the focal starting point of a multinational organizations’ knowledge transfers, is the individual. Once the individual has gained a cognitive grasp of this knowledge, it can be transferred to the organization. It is therefore vital that organizations manage transfer routines and provide strategic direction and structure in order to facilitate acceptance and implementation of knowledge as well as stimulate the necessary attitudes the individual needs to be a successful conduit of these transfers.

The Lerd course as mentioned in the previous paragraph is a local eco nomic rural development course designed to help local actors in a community by teaching them basic economic, financial, and managerial skills. The goals of the Lerd programs are to strengthen the local economic capacity of an area, to improve the investment climate, to increase the productivity and competitiveness of local businesses, entrepreneurs, and workers, to improve the quality of life in communities, and to create new opportunities to fight poverty. The Lerd programs are an initiative of the University of Groningen and are given in not only various regions of Indonesia such as West Borneo, South-East Sulawesi, and West Sumatra, but also in various regions in Africa in cooperation with the Local Economic Development Network of Africa (Ledna)[1].

Although much has been written on the topic of knowledge transfers, there is very little empirical research on knowledge transfers in higher education. Furthermore, the specific case of knowledge transfers from a Dutch university to an Indonesian student group, has not been researched to the best of my knowledge. Since these individual knowledge transfers are the focal starting point of organizational knowledge transfers, it is interesting for professors, students, researchers, and Dutch multinationals with subsidiaries and/or operations in Indonesia, to find the inhibitors and enablers of such knowledge transfers.

(7)

7

capability, credibility, communication, and culture. These characteristics of the source of the knowledge transfer affect the extent to which knowledge transfers are hindered or enabled. This framework was later also successfully applied to the context in which knowledge transfers occurred between individuals from a western country (U.S.A.) to individuals from a east-Asian country (Thailand) by Sarker (2005). It must be noted that when studying a

unicultural group of students such as in Joshi et al. (2004), the 4th construct, culture is unable

to be quantitatively researched and is hence eliminated from the original model, resulting in a 3C model. Since my recipient group is unicultural I will follow the lines of previous research and use the 3C model as the basic theoretical framework for my research.

This research will be a formal study of how these three factors affect knowledge transfers from the source to the participants in the Local Economic Rural Development (Lerd) course. Thus the main research question is: To what extent do the 3Cs as proposed

in the literature explain financial and managerial knowledge transfers from the source to the recipient group during the Lerd course.

Additionally, I will try to discover whether the 3C model is more adept in explaining either financial or managerial knowledge transfers for this sample group. I will do this by separately analyzing the effect of the 3Cs on financial knowledge transfers and analyzing the effect of the 3Cs on managerial knowledge transfers. This results in my secondary and tertiary research questions: To what extent are the 3Cs adept at explaining solely financial

knowledge transfers. To what extent are the 3Cs adept at explaining solely managerial knowledge transfers.

This paper is structured into two main parts, a literature review and an empirical analysis section on knowledge transfers. The literature review examines existing literature on knowledge, data and information, and the various factors that inhibit or enable knowledge transfers. This section is followed by the development of my theoretical model, and the research methodology presents a more detailed description and explanation of the design of the Lerd course.

(8)

8

I. Literature Review

Knowledge can be described as a state of knowing, or a condition of understanding in which the focus should be on facilitating the expansion of personal knowledge in individuals, and applying it to the needs of the organization. Knowledge can also be viewed as an object that can be stored and manipulated, in which case knowledge management should focus on managing knowledge stocks. However, if knowledge is seen as a process of acting and knowing simultaneously, then the focus should be on creating, sharing, and distributing knowledge (Alavi & Leidner, 2001). This means that the specific knowledge management strategy must be based on the accepted view of knowledge in an organization.

A. Data, Information, and Knowledge

We must initially make a clear distinction between the concepts of data, information, and knowledge in order to avoid confusion. Defining these concepts is difficult and we can only distinguish between them through external means or a user’s perspective (Bhatt, 2001). The generally accepted view is that informatio n is seen as an organized set of data, and knowledge is seen as meaningful information that can be used to make predictions when put into context. Hence data are descriptions or observations of past, present, or future phenomenon. Data can also be simple isolated facts, that can become information when put into a structured context. When the information is interpreted and given meaning it becomes knowledge, which is a context-dependent product of human reflection and experience. Furthermore, knowledge leads to increased decision making capacity and action to achieve some purpose (De Long & Fahey, 2000). This knowledge can, in turn, be used by the human mind to make inferences, or choose between alternatives and become intelligence (Tuomi, 1999).

B. The Three Types and Three Dimensions of Knowledge

(9)

9

We analyze these types of knowledge in terms of three dimensions. According to Garud & Nayyar (1994) the three dimensions of knowledge are simple vs. complex, explicit vs. tacit, and independent vs. systemic, and the three types of knowledge can have a certain positioning along each of the three dimensions.

The first dimension simple vs. complex is relevant during cross-border knowledge transactions. The amount of information needed to accurately convey complex knowledge is far greater than what would be needed to convey simple knowledge due to complex knowledge evoking more causal uncertainties (Bhagat et al. 2002). The second dimension explicit vs. tacit knowledge concerns how well articulated the knowledge is, and this knowledge is often highly specialized. Tacit knowledge requires richer context and is usually embedded in an individual’s cognitive processes. Because tacit knowledge cannot be easily reduced to a list of how it must be implemented and replicated, the level of causal ambiguity is high. Due to the fact that the bulk of knowledge exists in tacit form, it is non-tradeable and as such renders external markets ineffective mechanisms for knowledge transfers (Gupta & Govindarajan, 2000). Explicit knowledge differs from tacit knowledge in this sense because it can be easily codified and transferred. The third dimension is independent vs. systemic and refers to the extent that knowledge is embedded in the organizational context. Independent knowledge can be expressed on its own, but systemic knowledge must be described in the context of the transferring organization (Bhagat et al. 2002).

Figure 1: The Three Types and Dimensions of Knowledge

(10)

10

The three types and dimensions of knowledge are adeptly portrayed in the diagram as adapted from Bhagat et al. (2002) in figure 1 (p.9). Garud & Nayyar (1994) state that the amount of information needed to describe the knowledge and the amount of effort needed to transfer the knowledge depends on where it is positioned along each of the three dimensions. This is because each type of knowledge can exist in a particular form in all three dimensions. They further note that if the type of knowledge that is being transferred (whether human, social, or structured) is tacit, complex, and systemic, it is more difficult to transfer and absorb. This figure helps us visualize how the knowledge we are confronted with on a daily basis is structured and of which combinations of type and dimension it is composed.

C. Resource-Based Vie w

In organizational competition models such as Porter’s (1980) firms competing in an industry fight over control of the same strategic resources. This view does not consider competitive advantage as being sustainable because of resource mobility and similar organizational strategies. This is especially true when competing producers purchase their inputs from the same suppliers. These conditions would make it rather difficult for the manufacturer to create a competitive advantage (Dyer & Hatch, 2006). The price paid for a resource acquired through competitive markets should reflect its value to the organization, therefore the focus should be on resources made valuable inside an organization. In addition, the source of competitive advantage is not from resources held by many competitors, but resources that are difficult for competitors to imitate (Argote & Ingram 2000). Thus the resource-based view of knowledge considers every firm as constituting a bundle of knowledge and focuses on an organization’s capabilities, competencies, and intangible elements as being the factors that facilitate the competitive advantage due to the fact that they are difficult to replicate or imitate. Of these resources, the organization’s knowledge base has the greatest ability to provide sustainable differentiation and thus, competitive advantage (Gupta & Govindarajan, 2000).

D. Knowledge-Based View

(11)

11

because these knowledge-based resources are complex and difficult to imitate, these assets can produce long term sustainable competitive advantage. This can be further facilitated when an organization simultaneously creates new knowledge and utilizes existing knowledge.

E. The Process of Knowledge Creation

Nonaka, Toyama, and Konno (2000) propose a knowledge creation model consisting of three components. Firstly the SECI process which is the conversion from tacit to explicit knowledge, secondly the ba which is the shared context, and thirdly knowledge assets which are the inputs, outputs, and moderators of the knowledge creating process.

The SECI process consists of four modes of knowledge conversion which is defined as the interaction between explicit and tacit knowledge. This process which allows explic it and tacit knowledge to expand in quality and quantity consists of socialization, externalization, combination, and internalization.

Socialization uses shared experiences to convert new tacit knowledge. These shared experiences such as living in the same environment, ease the difficulties related with the transfer of tacit knowledge. An apprenticeship is a typical setting in which socialization occurs, however, informal social meetings also form an applicable venue for such transfers.

Externalization concerns the articulation of tacit knowledge into explicit knowledge. During this process knowledge is given a more definite form, such as images or written documents, and can thereafter be shared. Dialogue is an effective method of externalization through which contradictions among tacit knowledge of individuals are made explicit.

Combination is the conversion of explicit knowledge into more complex forms of explicit knowledge through editing and processing. Databases and communication networks are facilitators of this mode of knowledge conversion. Combining information from throughout the organization into one context and the breaking down of concepts into operationalized concepts creates explicit knowledge.

(12)

12

internalized as shared mental models in the minds of the individuals, and leads to further knowledge creation through socialization (Nonaka et al. 2000; Nonaka & Toyama, 2003).

The shared context ba is a “place” where knowledge is created, shared and utilized and where information is interpreted by individuals to become knowledge. Ba allows participants to share time and space for the integration of applied knowledge. Knowledge assets, in turn, are firm-specific resources, inputs, outputs and moderating factors that are absolutely vital to firm value creation. Knowledge assets can be divided up into four types, experiential, conceptual, systemic, and routine knowledge assets. Efficient management of these knowledge assets leads to a sustainable competitive advantage for the firm (Nonaka et al. 2000).

F. Knowledge Transfer in Higher Education

Knowledge transfers that occur in higher education institutions are increasingly codified. The knowledge spiral model developed by Nonaka, Toyama, & Konno (2000) is related to knowledge transfers in higher education. During the teaching process the source’s tacit knowledge is encoded into explicit knowledge through the process of externalization. This explicit knowledge is transferred through various mediums such as lectures and textbooks to the recipients. The recipient then decodes and understands this information through the process of internalization. Once this step is complete the recipient can engage in socialization which is the process of converting their tacit knowledge to others’ tacit knowledge, demonstrating completion of the knowledge transfer from source to recipient (Ju, et al. 2008b).

Research has shown that effective higher education improves a person’s career opportunities and performance, and within an organization this can lead to a competitive advantage. In the educational setting the main goal of higher education is to improve the abilities and skills, in this case financial and managerial skills, of the students. This means that the interactions between professor and student are of great importance. An advantage that higher education institutions often have, are extensive knowledge management resources such as libraries, digital learning, and networks (Ju et al. 2008a).

(13)

13

relationship, motivation-related factors; intrinsic motivation, and communication-related factors; communication encoding competence, and source credibility. Of these factors, capability, credibility, and communication are researched in this study following the constructs of the 3C model.

In the field of university to industry knowledge transfers various channels are used through which this transfer takes place. The most important of these channels that affect industrial R&D are through published papers and reports (Cohen et al. 2002). However, most benefits to firms are through formal collaboration with the university. In addition, the employment of university researchers by the firm is also an effective way of facilitating university to firm knowledge transfers (Bekkers & Freitas, 2008). The presence of “smart people” is hereby a critical factor that enables the transfer of knowledge in an organization. This means that a person who is more knowledgeable or “smarter” has a greater ability to facilitate the knowledge transfer process (Sarker et al. 2003).

G. Knowledge Transfer and the Firm

Since the 1960s world trade has greatly liberalized and trade barriers have been lowered allowing for a global flow of factors of production. In addition, policy regarding restrictions on knowledge transfers concerning importers and exporters have been relaxed. This liberalization not only facilitates knowledge transfer but has led to an expansion of what’s tradeable. The development of new financial markets for intermediate products such as swaps, index futures, puttable bonds, and Eurobonds has made knowledge increasingly relevant as a differentiator which is a source of a sustainable competitive advantage for the firm. This competitive advantage flows from successful deployment of non tradeable assets such as knowledge assets. These assets are grounded in the expertise and experience of the individuals and when combined with the firms physical resources, become competencies (Teece, 1998).

H. Knowledge Transfer Mechanisms

(14)

14

management literature on knowledge transfers show that firms that have effective methods for the internal transfer of knowledge, and prevent knowledge spillovers to external parties, are more successful than firms that lack effective knowledge management (Argote et al. 2000). This is because external knowledge spillovers are due to the inability of a firm to establish property rights over knowledge. This in turn leads to a decrease in willingness to invest in research and development, and this creates an opportunity for free riders to benefit from the inventive knowledge leaks across firms (Almeida & Kogut, 1999). Effectiveness in cross-border knowledge transfers in organizations is a critical component of retaining competitive advantage as competition among MNCs intensifies (Bhagat et al. 2002). This effectiveness, in turn, depends on how capable both the transferring and the recipient organizations are in using institutional mechanisms to accomplish these transfers.

According to Alavi (2000) knowledge generation coupled with successful knowledge transfers are necessary for superior organizational performance. Transferring knowledge can be a time consuming endeavor, and this is even more so when transferring knowledge across

cultures. Culture affects knowledge transfers in MNCs when their subsidiaries operate in

heterogeneous national cultures, and the compatibility of these different cultures determines the level of inter-organizational cooperation. A recent theory states that differences in beliefs between perception and reality determine how strongly culture affects inter-organizational relationships (Lucas, 2006).

Knowledge can be transferred thro ugh a variety of mechanisms which include communication, training, personnel movement, technology transfers, patents, observation, the reverse engineering of products, scientific publications, and interorganizational relationships such as mergers, acquisitions, strategic alliances and joint ventures (Argote et al. 2000). According to Simonin (1999), strategic alliances constitute the most adequate vehicle for internalizing competencies from a knowledge holder to a knowledge seeker in the field of knowledge imitability. Nevertheless, knowledge is fairly immobile and is also referred to as being “inert”. This can be attributed to its ambiguity, context embeddedness, and res istance to clear communication.

(15)

15

movement of engineers in the semiconductor industry in Silicon Valley. The transfer of knowledge through personnel movement allows for the exchange of both tacit and explicit knowledge. This knowledge is much more difficult to transfer across more conventional means such as documents and/or presentations. In part due to this reason, many researchers accept personnel rotation as a knowledge transfer mechanism Kane et al. 2005).

Social identity is an important factor which positively influences knowledge transfers between groups. The degree to which persons feel a sense of belonging to a social aggregate is their social identity which is a part of their personal identity. The degree of social identification in such groups leads the individuals to feel more comfortable in sharing knowledge with ingroup members rather than outgroup members Kane et al. 2005).

There are a number of different components that determine knowledge transfers. According to Joshi et al. (2004) the knowledge transfer process is determined by five contextual components; source, recipient, knowledge, relational, and situational. Here source context and recipient context refer to the attributes that can facilitate or inhibit the knowledge transfer process. The knowledge context refers to the type of knowledge being transferred. Of the other contextual components, relational context refers to the relation between the source and the recipient, and situational context refers to the immediate enviro nment in which the transfer takes place, such as physical proximity (Joshi et al. 2004). The three contextual components source, knowledge, and relational is where this research will place its focus.

I. Barrie rs to Knowledge Transfer

According to research, knowledge transfers can be relatively hard to achieve and are even referred to by some researchers as time consuming, laborious, and difficult which makes it important to understand what the impediments to these knowledge transfers are (Sarker et al. 2003). Inter-organizational knowledge transfers from one unit to another may fail due to a number of reason such as the quality of the relationship between the donor and recipient groups, or due to the specific characteristics of the knowledge that is being trans ferred (Kane et al. 2005).

(16)

16

transfer (Gupta & Govindarajan, 2000; Szulanski, 1996). Furthermore, buyers who wish to create competitive advantage by transferring knowledge to their suppliers run the risk of knowledge spillovers to their competitors who also use the same supplier. In order for the buyer to create sustainable competitive advantage, the knowledge asset must be relationship specific. If this is not the case, competitors can tap into these network resources via members of the same network (Dyer & Hatch, 2006).

Additional barriers to knowledge transfer are categorized by Szulanski (1996) as causal ambiguity, lack of source motivation, lack of recipient motivation, lack of source credibility, lack of recipient absorptive and retentive capacity, and an arduous relationship between source and recipient. Causal ambiguity stems from the fact that knowledge is complex and gives rise to ambiguity concerning how the factors of production interact and how to measure their marginal contribution. This relates back to the notion that much knowledge exists in tacit form such as human skills, and is difficult to transfer. Additionally causal ambiguity could result from an imperfect understanding of the new context in which knowledge is put to use.

Lack of motivation hinders knowledge transfers when the source is reluctant to share knowledge. This could be attributed to the source’s fear of losing ownership of this knowledge or fear of losing superiority. Additionally recipient lack of motivation can be a result of his/her reluctance to accept knowledge from outside, also known as the “not invented here” (NIH) syndrome. This lack of recipient motivation can manifest itself as passivity, feigned acceptance, or the rejection of implementation.

(17)

17

Previous research has shown that the national culture of the source also influences knowledge transfers. Culture creates a context in which values and knowledge are exchanged and these stem from the norms, values and assumptions of the culture (Joshi et al. 2005). This “programming” of the mind is manifested in beliefs, behaviors, and actions and distinguishes people from each other (Sarker et al. 2003). The cultural dimension used to examine the influence of the knowledge source’s culture on the transfer, and deemed most relevant for the 4C framework is the individualism/collectivism dimension in Hofstede’s (2001) study on cultural distance. Part of the individualistic behavior is the belief that withholding information is the key to success, thus individuals from collectivist cultures should theoretically transfer more knowledge. Empirical research on this topic, however, has shown ambiguous results. In research by Joshi et al. (2003) a significant positive relation was found between culture and knowledge transfers ratifying their hypothesis that individuals from more collectivist cultures transfer more knowledge, while in research by Joshi et al. (2005) and Sarker (2005) a significant negative relation was found between culture and knowledge transfers meaning that individuals from more individualistic cultures transfer more knowledge. Further research on culture is needed in order to identify its affect on knowledge transfers.

Studies have shown that competition is another factor that can adversely influence knowledge transfers. In a study by Connelly et al. (2009) the researchers found that competition in both team and individual context, significantly affected knowledge transfers. Similarly, this research has shown that during competitive interactions, such as tests or exams, time constraints could also play a part in influencing the transfer of knowledge between individuals.

Organizations must be clinically aware of the barriers to knowledge transfers and know how to mold their organizational processes and strategies accordingly, as this will increase the extent of knowledge transfers and foster competitive advantage.

(18)

18

II. Theoretical Framework

A. Theoretical Model

The 3C model created by Joshi et al. (2004) that I utilize for this research is based on the 4C model created by Sarker et al. (2003) to measure knowledge transfers. The independent variables capability (C1), credibility (C2), and communication (C3) are used in both models to predict the dependent variable knowledge transfer. The only difference is the omission of culture in the 3C model due to the unicultural nature of the recipient sample group. As mentioned before, the 3C model is based on the source context of knowledge transfers, but also includes the relational and knowledge contexts as methods of measuring knowledge transfers. The following diagram (Figure 2, p.18) of the theoretical model shows the assumed relationship between the variables. The source, in an environment of institutional engagement transfers knowledge to a recipient group. Various factors influence this transfer in a variety of ways, and thus the main focus in this case is how capability, credibility, and communication affect these knowledge transfers during the Lerd course.

Figure 2: Theoretical 3C Model

Here the causal relationship between the dependent and independent variables is shown. Stemming from the knowledge source, in this case the university professor, are the

Source Lerd Capability Credibility (C2) Communication (C3) Engagement Source

Lerd fi nancial capability &

Lerd managerial ca pability

Lerd fi nancial capability & Lerd mana gerial capability

(19)

19

constructs of capability, credibility, and communication. To derive the independent variable capability, the difference between the financial and managerial capabilities of the source and the recipients is calculated. In this study and based on previous research (Joshi et al. 2004; Joshi et al. 2005; Sarker, 2005) I assume that for all these independent variables a positive relation exists with the dependent variable. I will now give a more detailed explanation of the variables in this theoretical model.

B. Variables

B.1. Dependent Variable: Knowledge Transfer

Knowledge transfer is a process of diffusion and travels from source to recipient, and for this study is defined as the amount of knowledge that is perceived to be transferred to the recipients by the source during the Lerd course.

Researchers have used a variety of methods to define knowledge transfer as a dependent variable. Some frequently used methods are the number of transfers during a period of time, the degree of difficulty, the degree to which knowledge is recreated in the recipient, and the knowledge internalization approach (Joshi et al. 2004). Following the lines of previous research, for this study I will use the knowledge internalization approach. This approach assesses the outcome of the knowledge transfer in terms of learning experiences, or how much knowledge the recipients have gained from the source (Joshi et al. 2005), through measuring changes in the level of recipient knowledge (Argote & Ingram, 2000).

B.2. Independent Variables and Hypotheses

The most important factors affecting knowledge transfers are source, recipient, situational, relational, and knowledge (Joshi et al. 2005). Since recipient knowledge transfers have received much attention in previous knowledge transfer research, I will focus on the source, relational, and knowledge contexts. In the 3C framework, the independent variables that measure these three contexts are capability, credibility, and communication. I will now further explain each of the independent variables and formulate my research question into more specific hypotheses.

B.2.1. Knowledge Context: Capability and Knowledge Transfer

(20)

20

to a greater understanding of relevant knowledge (Joshi et al. 2004), as well as the ability to transfer knowledge more effectively (Joshi et al. 2005).

Capability refers to the “smart people” or knowledgeable persons that possess the necessary expertise needed to accomplish the tasks at hand. This is logical because a person or source with greater knowledge is able to transfer more knowledge to a recipient than a source with limited knowledge. Additionally, due to rec ipients’ absorptive capacity, experts are more likely to influence recipient behavior than others (Sarker et al. 2003). Furthermore research has shown that individuals that are more competent and knowledgeable than recipients are more likely to transfer knowledge to the recipients (Levin et al. 2004). Thus the amount of knowledge transferred is hypothesized to be proportional to the difference in their respective knowledge levels.

Successful completion and execution of the Lerd course requires recipients to obtain and internalize certain amounts of knowledge. I have classified this knowledge into the two categories of financial knowledge and managerial knowledge. This classification is further explained in section II. Research Methodology (p.25). I make the argument that sources with higher levels of expertise will transfer more knowledge to the recipients. This brings me to my first hypothesis.

Hypothesis 1: The greater the difference in financial and managerial capabilities

between the source and the recipients, the greater the extent of knowledge transferred by the source.

B.2.2. Source Context: Credibility and Knowledge Transfer

(21)

21

the knowledge (Joshi et al. 2005; Szulanski, 1996) and this may, in some cases cause recipients to challenge and resist the advice offered by the source (Sarker et al. 2003). This brings me to my second hypothesis.

Hypothesis 2: The extent of knowledge transferred by the source is positively

related to his/her credibility, as perceived by the recipients.

B.2.3. Relational Context: Communication and Knowledge Transfer

Communication influences knowledge transfers in that it is the main venue by which colleagues share know-how with each other. In addition research has shown that frequent communication facilitates interaction and reduces anxiety caused by misinformation amongst individuals. Extended discussions allow an individual’s views and beliefs to be made available to others and this helps to create a shared meaning through which knowledge can be transferred (Szulanski, 1996; Sarker et al. 2003). The internalization of knowledge is facilitated by interactive communication (Joshi et al. 2005) and this is the primary venue through which individuals discover what they know and share this knowledge with their colleagues.

Past literature has brought up various arguments such as member distraction, creativity blocking, and decreased productivity, that advocate a negative relation between communication and knowledge transfer (Baron, 1986). Nevertheless, more recent studies have shown that frequent communication and meetings are vital predictors of knowledge transfers (Joshi et al. 2004). In addition Szulanski (1996) notes that an arduous relationship impedes knowledge transfers, and that this can be resolved by increased levels of communication which lead to socialization which, in turn, nurtures group-values that cultivate ardent relationships and lead to a shared context between source and recipient. Communication is an important way for a recipient to internalize knowledge (Nonaka et al. 2000) and has an auxiliary function in building and maintaining the embedded social capital of a group (Joshi et al. 2004). Thus I hypothesize that high volumes of interactive communication will transfer more knowledge to the recipients. This brings me to my third hypothesis:

Hypothesis 3: High volumes of communication between the source and the

(22)

22

C. Differentiated Models for Knowledge Transfers

Based on my initial theoretical 3C model (figure 2; p.18) of knowledge transfers I will differentiate between financial and managerial knowledge transfers. The main way the differentiated models will deviate from the original is in how the dependent variable

knowledge transfer (KT) is defined, and how the independent variable capability (C1) is

defined. I will make this distinction in order to identify the individual effects of these constructs as compared to their collective effects on the theoretical model.

C.1. Theoretical Model for Financial Knowledge Transfer

In the theoretical model for financial knowledge transfer t he 3C model will, in

essence, remain unchanged. However, the way KT and C1 are defined will change. The

dependent variable KT will no longer represent both acquired financial and managerial skills but only represent acquired financial skills and will be called Financial Knowledge Transfer

(KTfin). In addition the independent variable C1 will no longer measure the difference in

initial levels of both financial and managerial capabilities of the source and recipient, but only the difference in initial levels of financial capabilities(C1f) of the source and recipient. The model for financial knowledge transfer is depicted in figure 3 (p.22).

Figure 3: Theoretical Model for Financial Knowledge Transfer

(23)

23

For the theoretical model for financial knowledge transfer I hypothesize that the

causal effects of the independent variables C1f, C2, and C3, on the dependent variable KTfin

follow previous research and follow the same causal relationships as in my original 3C model, thus leading to the following hypotheses:

Hypothesis 1a: The greater the difference in financial capabilities between the source

and the recipients, the greater the extent of financial knowledge transferred by the source.

Hypothesis 2a: The extent of financial knowledge transferred by the source is

positively related to his/her credibility, as perceived by the recipients.

Hypothesis 3a: High volumes of communication between the source and the

recipients will positively affect the amount of financial knowledge transferred to the recipients.

C.2. Theoretical Model for Managerial Knowledge Transfer

In the theoretical model for managerial knowledge transfer the 3C model will also

remain largely intact. However, the way KT and C1 are defined will change. The dependent

variable KT will no longer represent both acquired financial and managerial skills but only represent acquired managerial skills and will be called Managerial Knowledge Transfer

(KTman). Additionally, the independent variable C1 will no longer measure the difference in

initial levels of both financial and managerial capabilities of the source and recipient, but

only the difference in initial levels of managerial capabilities (C1m) of the source and

(24)

24

Figure 4: Theoretical Model for Managerial Knowledge Transfer

For this differentiated model I hypothesize that the causal effects of the independent variables C1m, C2, and C3, on the dependent variable KTman follow the lines of previous research and follow the same relationships as in my original 3C model, and differentiated financial 3C model, thus leading to the following hypotheses:

Hypothesis 1b: The greater the difference in managerial capabilities between the

source and the recipients, the greater the extent of managerial knowledge transferred by the source.

Hypothesis 2b: The extent of managerial knowledge transferred by the source is

positively related to his/her credibility, as perceived by the recipients.

Hypothesis 3b: High volumes of communication between the source and the

(25)

25

III. Research Methodology

A. Design

The Lerd program is designed to strengthen the local economic capacity of a specific area. This includes improving the investment climate, increasing productivity, increasing the competitiveness of local businesses, and stimulating entrepreneurship. This is brought about firstly by a theoretical five stage strategic planning process, followed by field trips to various institutions, and finally writing specific business plans. The Lerd “engine” is portrayed in the following diagram, figure 5 (p.25).

Figure 5: The Lerd Engine

(26)

26

completion of the course in The Netherlands the students retur n to Indonesia to present the new economic development plans to their stakeholders. Together with the stakeholders the development projects are initiated with the short term goal of sector development, and the long term goal of regional development. The entire Lerd program is an ongoing process that can take up to several years before the long term goals are achieved.

The focal point of this study is the training stage of the program (see figure 5; p.25). The participants should acquire certain skills and capabilities upon completion of the Lerd course (given in the University of Groningen, The Netherlands) and I have divided these into two categories, financial issues, and managerial issues.

A.1. Financial Issues

The financial issues mentioned here are the financial skills taught in this Lerd course and are aimed at providing the recipients with the basic financial knowledge and skills that are needed to be able to finance operations in their home country, instill awareness concerning basic finance paradigms, and competitively produce goods. These issues range from basic costing and pricing to understanding financial accounting.

The broad range of financial issues that are taught during this development course are basic financial issues such as; pricing, cost analysis, accounting, microfinance, budgeting, exchange rate volatility, taxation, and investment strategies. The extent to which the recipients and source are competent in these financial areas is measured in detail before the start of the course. In addition, the extent to which the recipients have acquired and learned these skills will be measured upon completion of the course (see appendix section C; p.59).

A.2. Managerial Issues

These are the knowledge and skills needed to be able to cultivate an ove rall competitive advantage regionally, nationally, and internationally. This is done by using business theories, management strategies, and marketing strategies to gain a comparative advantage over their competitors. The following managerial competencies are taught during the Lerd course and the extent to which the following managerial skills are learned will be measured:

1) Learn how to set up a research project/action plan

(27)

27

3) Become familiar with entrepreneurship and small business development policy

in The Netherlands

4) Gain improved insight into the role of leadership and teamwork in small

businesses

5) Learn how to create competitive advantage and develop SWOT analysis

6) Become familiar with marketing strategies, distribution and export marketing

problems of SMEs

7) Learn about the importance of networking in SMEs

8) Learn how to deal with cross-cultural negotiations and manage change in

regional development

(see appendix section C; p.59).

This case study on knowledge transfers during the Lerd course will evaluate the financial and managerial knowledge and skills of the recipients and the sources in detail before the start of the course, forming the basis of how the independent variable capability is measured. In addition, the recipients will be evaluated after completion of the Lerd course concerning the extent of acquired financial and managerial knowledge. This is how the dependent variable knowledge transfer is measured.

(28)

28

IV. Data & Research Structure

A. Data Collection

Primary data will be collected from a sample of 25 Indonesian students that are following the Local Economic Rural Development (Lerd) course given by staff members of the University of Groningen in The Netherlands. The time dimension of this study is in essence a cross-sectional study, however the financial and managerial Lerd skills acquired by the recipients over the duration of the course will be measured longitudinally. This will be done through hard copy questionnaires aimed at assessing the capacity of these stude nts before the beginning of the course. Secondly, a second questionnaire that gauges the total knowledge transferred after completion of the course, coupled with the add itional elements of credibility and communication, will be given to the same sample of students after they have completed the Lerd course.

Student subjects have often received criticism due to supposed lack of generalizability, however, researchers consistently advocate their similarity to working professionals (Dipboye & Flanagan, 1979), and students are thus appropriate subjects for the investigation of social processes (Joshi et al. 2004).

The questionnaires will comprise a combination of close-ended and open-ended questions. The close-ended questions will be based on established questionnaires and likert rating scales ranging from 1 to 7, that have been used in previous research on knowledge transfer (Joshi et al. 2004; Joshi et al. 2005; Sarker et al. 2003; Sarker, 2005). This will allow for statistical comparison between the levels of competence as well as the source characteristics perceived by the participants. The open ended questions will serve to increase the scope and provide additional insight into personal experiences, cultural inferences, and other information relevant to the study.

B. Research Structure

(29)

29

Gauging engagement of the university is important in that it shows evidence that the university in question has the ability and the capacity to engage with institutions, communities, and constituencies, thus facilitating knowledge transfers. According to the German Center for Higher Education Development (CHE), the University of Groningen is a member of the “Excellence group” and is in the top 25 European universities; hence, for the purpose of this study we will adhere to the assumption that the level of institutional engagement of the University of Groningen is sufficient for the transfer of knowledge.

(30)

30

V. Measures

A. Dependent Variable

Extent of knowledge transferred (KT): The dependent variable in this study is the extent of knowledge transferred by the source, and is measured by the average of two items; the extent to which the recipients acquired financial Lerd skills, and the extent to which the recipients acquired managerial Lerd skills from the source. The questionnaire will use likert scales ranging from 1 (not at all) to 7 (to a great extent). KT is assessed after completion of the Lerd course.

B. Independent Variables

Capability (C1): The independent variable capability is measured as the difference in

capability between the source and the recipients. This is gauged by the self-reported capabilities in the questionnaire regarding Lerd financial and managerial skills. The difference in financial skills between the source and the recipient is calculated as well as the difference in managerial skills between the source and the recipient. The average of these calculated differences constitutes the individual recipient’s independent variable capability. C1 is measured using likert scales from 1 (not at all) to 7 (to a great extent), and is assessed before the recipients start the Lerd course.

Credibility (C2): The independent variable credibility is based on the recipients’

perception of the source’s credibility based on its two co mponents, trust and performance. The average of the scores given by the course recipients for both trust and level of performance is calculated as credibility. C2 is measured using likert scales of 1 (low) to 7 (high), and is assessed after completion of the Lerd course.

Communication (C3): The independent variable communication measures the extent

of communication that the recipients have with the source during the course, based on the recipients’ perceived level. C3 is measured on a likert scale of 1 (low) to 7 (high), and is assessed after completion of the Lerd course.

C. Control Variable

Age (A4): Of the various variables that could be used to control the sample, t he sole

(31)

31

(32)

32

VI. Measurement constructs

A. Reliability

Reliability analysis of the data collected is measured using Cronbach’s Alpha, which is the mean of all the possible split- half coefficients that are computed using the Rulon method (Crocker & Algina, 1986) using formula 1 (p.32).

(1)

Where N is the number of items in the scale, S2yi is the variance of the item scores i,

and S2x is the variance of the scale. Cronbach’s alpha was developed by Cronbach (1951) as a

way of measuring the internal consistency of a multi item scale. In order for a scale to be valid, it must be reliable, reliable in this sense meaning the degree to which measures yield consistent results and are free from error (Peterson, 2004). The Cronbach coefficient alpha is a measurement scale of 0 to 1; the reliability of the scale is generally accepted at levels above 0.6 although many researchers advise for 0.7 or higher (Nunnally, 1978).

The questions in the questionnaire de termined the measure of the variables. The answers to the sub-questions in each section were grouped to form the summated scale used to measure the variable in question. The reliability of each scale used to measure the objective was assessed using Cronbach’s alpha to measure internal consistency.

Table I (p.33) shows Cronbach’s coefficient alpha values for the dependent variable

knowledge transfer (KT) and for the independent variables capability (C1) and credibility

(C2). The third independent variable communication (C3) consists of a single item scale, and

as such, the reliability is not able to be computed.

The dependent variable KT which measures the extent of financial and managerial

knowledge transfers has an alpha value of 0.753. C1 was measured using the average of 18

financial measures and 16 managerial measures. The internal consiste ncy of C1 is measured

(33)

33

Table I: Cronbach’s Coefficient Alpha values

Knowledge construct Cronbach’s alpha

Knowledge transfer (KT) .753

Capability (C1) .779

Credibility (C2) .755

B. Validity

Validity is concerned with measuring whether the scale accurately measures the construct and should be supported by previous research. This research study is based on past

empirical studies which show relationships between the independent variables C1, C2, and C3,

and the dependent variable KT. In table II (p.38) Spearman’s Rank Order (rho) correlations

are presented and show the relationship between the variables. The variables C1 (rho = 0.575,

N = 25, p<0.01) and C2 (rho = 0.584, N = 25, p<0.01) show validity through their significant

correlation with KT, however C3 does not. Correlations with p<0.05 are significant enough to

confirm past research, and correlations with p<0.01 indicate that the strength of the relationship is considered to be statistically correct in 99% of the cases.

It is important to examine whether the measures which should not be related, are not related and that the measures discriminate between each other. In table II (p.38) we see that none of the independent variables are correlated to each other and this suggests that the variables exhibit discriminant validity.

C. Multiple Regression

(34)

34

I will use multiple regression analysis as the method of determining the relationship between the dependent and the independent variables using equation 2 (p.34).

KT = β0 + β1C1 + β2C2 + β3C3 + β4A4 + ε (2) Where KT is the dependent variable extent of knowledge transfer, the independent

variables C1, C2, and C3 are capability, credibility, and communication respectively. A4 is the

control variable age, the betas (β) are the regression coefficients, and ε is the error.

I will also carry out multiple regression without the control variable age to determine whether the 3C model has a better ability to predict the dependent variable wit hout it. This will serve to show whether the model has a better fit and whether a greater percentage of the variance in the dependent variable can be explained by the independent variables. This regression equation is formulated in equation 3 (p.34).

KT = β0 + β1C1 + β2C2 + β3C3 + ε (3)

C.1. Interaction Effects

Next I will check for interaction effects, which is the case when the effect of two or more variables is not simply additive, for example when the impact of an independent variable depends on the level of another independent variable. I will do this check because when interaction effects are present, interpretation of the individual variables, such as in linear regression, can be incomplete or misleading. According to Aiken & West (1991) centering the predictor variables leads to an increased interpretability of the interactions. Failure to use centered variables may cause multicolinearity problems with the data meaning that the product of the independent variables may have a high correlation with the original independent variable.

I will compute centered variables from the independent variables by subtracting the mean from the observation and then computing the difference of the two. I will denote the centered independent variables with a lower case “c”. In order to check for interaction effects the must take the product of the centered variables and add it to the regression equation as an independent variable. Since I have three independent variables I will need to compute Cc1Cc2, Cc1Cc3, Cc2Cc3, and Cc1Cc2Cc3 for interaction effects. For my 3C model the regression equation will be as posted in equation 4 (p.34).

(35)

35

Where Cc1 is the centered variable capability, Cc2 is the centered variable credibility, Cc3 is the centered variable communication, and Cc1Cc2, Cc1Cc3, Cc2Cc3, and Cc1Cc2Cc3 are the interaction variables.

C.2. Regression in the Differentiated 3C Models

In order to more closely examine how the independent variables capability, credibility, and communication affect financial and managerial knowledge transfers I will separate the dependent variable knowledge transfer (KT) into the two respective dependent variables financial knowledge transfer (KTfin) and managerial knowledge transfer (KTman)

and conduct regression analysis on both. Since the independent variable capability (C1) is

also measured by both financial and managerial skills, this predictor will also be separated into its two respective components to create two independent variables, financial capability

(C1f) and managerial capability (C1m). For measuring financial knowledge transfers I will use

the differentiated regression equation 5 (p.35).

KTfin = β0 + β1C1f + β2C2 + β3C3 + β4A4 + ε (5)

Similarly for measuring managerial knowledge transfers I will use the differentiated regression equation 6 (p.35).

(36)

36 D. Data Assumptions

D.1. Normality Plot

Figure 6: Normality Plot of Variables Against KT

A normality plot or normal probability plot is a way of testing for normality. Figure 6 (page 36) shows the normality plot of the dependent and the independent variables. In figure 6 the normality plot of the residuals follow the straight line along the diagonal. Since there are no significant deviations we can assume that the model has a good fit and the assumption of linear relationship is not violated.

I have also run partial regression plots for the independent variables to show the nature of the relationship. Partial regression plots attempt to show the affect of adding an additional variable to the model, and shows whether there are failures or violations of the underlying assumptions of the model. The partial regression plots graphs are posted in the

appendix subsection A (p.57), and show a strong positive linear relation with KT against C1, a

strong positive linear relation with KT against C2, and only a moderate positive linear relation

with KT against C3. These relationships are reflected in the regression results in table 3 (p.39).

D.2. Data Distribution

(37)

37

Figure 7: Histogram of Frequencies and Standardized Residual

Figure 7 shows a slightly positive distribution which is a small violation of the normal distribution, and may only weaken the analysis slightly but will not invalidate it.

D.3. Outliers

Next I will check for outliers in the data sample by using a scatter plot, which is often used to identify the strength of the relationship between the variables. Figure 8 (p.37) shows no extreme outliers exceeding ± 2 of the standardized variable value.

Figure 8: Scatter Plot for Outliers

(38)

38

VII. Analysis of Results

A. Introduction

The data collected for this case stud y is analyzed using a number of parametric and non-parametric tests. Since the data in this research is categorical and measured with ordinal scales, non-parametric tests are ideal. Multiple regression is used even though it is a parametric test and this is because the results produced by either parametric or non-parametric tests are usually similar or identical.

B. Original 3C Model

B.1. Exploring Relationships Between Variables

All of the ordinal variables are analyzed using Spearman’s rank correlation (rho) in order to identify the relationship between the dependent and independent variables. Analysis of the results of this test follow the guidelines of Green, Salkind & Akey (2000) which state that for behavioral sciences weak correlations are rho = 0.3, moderate correlations are in the range rho = 0.3 to rho = 0.5, and strong correlations are greater than rho = 0.5, where p = <0.05. Table II (p.38) shows the descriptive statistics for the variables and correlations.

Table II: Means, Standard Deviations, and Spearman’s Rank Order Correlations.

**correlation is significant at the 0.01 leve l (t wo-tailed)

I will discuss the statistically significant correlations with the dependent variable KT first. C1 is significantly correlated with KT (rho = 0.575, N = 25, p<0.01), C2 is strongly correlated with KT (rho = 0.584, N = 25, p<0.01). Besides these two independent variables that show strong positive correlations to the dependent variable, we see no other significant

Variables Mean St. dev. 1 2 3 4

(39)

39

correlations among the variables. These relationships are analyzed further in the regression analysis in subsection B.2. (p.39).

B.2. Multiple Regression Results 3C Model

The results from the regression are shown in table III (p.39). Regression equation 7 shows the relation between the dependent variable, the predictors and control variable in formula form.

KT = -0.865 + 0.509C1 + 0.609C2 + 0.134C3 + 0.012A4 (7)

Table III: Multiple Regression Results 3C Model

*significant at the 0.05 leve l; (R2=.576)

The squared multiple correlation R2 is 0.576 which shows that 57.6% of the variance

in the dependent variable KT is explained by the independent variab les.

The results of the regression analysis show a significant positive (0.509) relation

between the independent variable C1 and the dependent variable KT (sig. = 0.016). Where

significance is assumed at confidence levels of 95% or higher. This is in line with previous research by Joshi et al. (2005) who found that capability (measured in technical and managerial ability) was positively related to knowledge transfers. Additionally we see a

significant positive (0.609) relation between the independent variable C2 and the dependent

variable KT (sig. = 0.048). This is in line with previous research by Joshi et al. (2004); Sarker (2005) who found that credibility was positively related to knowledge transfers in their research.

For the independent variable C3 we see an insignificant positive relation to the

dependent variable KT. This means that for this study the extent of communication between

MODEL B Std.

Error

(40)

40

the source and the recipient has not been able to predict the extent of the knowledge transfers, although most previous studies using the 4C model (Joshi et al. 2003; Sarker, 2005) and 3C model (Joshi et al. 2004) have found a significant positive relation between communication and knowledge transfers.

The relation between the control variable A4 and the dependent variable KT is a very

slightly positive, but nevertheless insignificant. This means that age did not have any significant effect on the extent of knowledge transfers. Previous research has also not been able to present consistent findings in this area. While some research shows that the extent of knowledge transfers is reduced as the age of the recipients increase due to loss of productivity, skills, and critical knowledge (Strack et al. 2008), this has yet to be supported by a larger more comprehensive body of empirical research.

B.3. Multiple Regression Results Omitting the Control Variable

Due to the insignificant effect of age on knowledge transfer I will check how the omission of the control variable age will affect the regression results and if it will result in a

better model fit and higher R2. The results of the regression without the control variable age

are posted in table IV (p.41). Regression equation 8 shows the relation between the dependent and independent variables.

KT = -0.089 + 0.542C1 + 0.532C2 + 0.143C3 (8)

Table IV: Multiple Regression Results Omitting the Control Variable

**=significant at the 0.01 level; *=significant at the 0.05 level; (R2=.568)

The squared multiple correlation R2 is 0.568 which shows that 56.8% of the variance

in the dependent variable KT is explained by the independent variables. This is a slightly

lower R2 than in the multiple regression results including the control variable age. The

regression results are, nevertheless similar with a significant positive (0.542) relation between

MODEL B Std.

Error

Financial and Managerial Knowledge Transfers to a Unicultural Recipient Group: The Case of the Lerd Program

Financial and Managerial Knowledge Transfers to

a Unicultural Recipient Group:

The Case of the Lerd Program

Table of Contents

Abstract

Introduction

I.

Literature Review

II.

Theoretical Framework

III. Research Methodology

IV.

Data & Research Structure

V.

Measures

VI.

Measurement constructs

VII. Analysis of Results