Nonaka & Takeuchi’s SECI model of knowledge conversion

Get Complete Project Material File(s) Now! »

Methods

In this chapter, the method is presented to illustrate the most appropriate research approach chosen to exe-cute the purpose of the paper. The demonstration is aiming to explain in detail how the study is carried out, followed by the explanation of the research strategy and data collection methods used.

Methodological approach

The philosophy behind the decision to undertake a certain research strategy should be explained, in order to conceptualize and undertake a study. It influences how the study is carried out and understood by the re-searchers (Johnson et al., 2006).
Since knowledge as a topic is reasonably manifold and thus hard to measure, the research questions can be best approached from an abductive standpoint. Dupois & Gadde (2002) argue that the abductive approach can be understood as a mixture of deductive and induc-tive approaches. It is especially beneficial approach to be taken if the objective of the study is to explore something new. It provides possibilities to take an advantage not only of the empirical world, but also of the theoretical models. Instead of going directly from the theo-ry to finding data, or moving from specific data to theory, an abductive approach allows to begin with a set of observations and to continue to the most possible explanation for these observatios.
To best understand knowledge transfer, the study is carried out on an exploratory basis. Ex-ploratory research can be characterized by a flexible and evolving approach to understand phenomena that are inherently difficult to measure (Malhotra & Birks, 2006). It is ideal for seeking insights from the knowledge transfer process and pose questions in order to assess the object of interest in a new light and clarify the problem at stake. As new insight ap-pears, exploratory research allows a change in direction if needed.
Two case studies were conducted and analyzed using a qualitative approach to explore the knowledge transfer practices and the challenges associated with. In order to gain a compre-hensive picture, the cases are first described individually and then later analyzed against rel-evant theories.

Method approach

Two different research approaches can be taken: quantitative and qualitative. According to Silverman (2010), the decision rather depends on the posed research questions and the matter of practicality. To better understand the challenges stemming from knowledge transfer within organizations, the topic can be best approached from a qualitative point of view. Choosing a qualitative approach does not only provide a good basis for understand-ing, but also helps integrate the research questions, the data, and the data analysis (Richards & Morse, 2007).
The focus of the qualitative approach is on expressive descriptions, observations and stories that produce data that is mainly interested in words, in a form that knowledge can be best observed as well. The benefit of choosing the qualitative approach is that it helps make sense of the world of the particular, helping provide elaborate interpretations of phenome-na that can hardly be identified in quantitative terms (Richards & Morse, 2007; Zikmund, 2000). A qualitative approach tends to be more flexible, allowing a greater adaptation to different contexts and more complex phenomenon. Thus, it tends to work with a relatively small number of cases (Crossan, 2003).
On the other hand, the quantitative approach aims to determine the quantity or extent of some phenomenon that can be expressed in the form of numbers. However, although this has been historically the “norm” among many social science departments (Silverman, 2010), the topic of knowledge can hardly be approached from this perspective. Thus, com-pared to the qualitative approach, quantitative methods often rely on large amount of ana-lytical data that is not needed in this case. The type of data quantitative approach produces is often lacking in flexibility, which is not the most ideal way to gain in-depth understand-ing of the issue at stake.

Research strategy

Case study

Conducting a case study was perceived to be the best practice to fulfill the purpose of this thesis. A case study is a research strategy based on the study of contemporary in a real life context (Silverman, 2010; R. Yin, 2007; Eisenhardt, 1989). ‘The term ‘case’ refers to an ap-proach and a focus.’ (Byrne & Raging, 2009) According to Welman, Kruger and Mitch-ell’s (2005) definition, the term pertains to the fact that a limited number of analyses are studied intensively, being more than simply conducting research on a single individual or situation (Baxter & Jack, 2008).
According to Yin (2007), the essence of a case study is that it attempts to illuminate events or a set of events in a natural setting. Accordingly, this is especially vital in gaining data about knowledge transfer. It is also beneficial that case studies have the potential to deal with simple as well as complex situations (Baxter & Jack, 2008). While taking into consider-ation how the focus of the study, process, or phenomenon is influenced by the context in which it is situated, case studies enable one to answer “how” and “why” type questions (Yin, 2007).

The case study design – a multiple case study

Yin (2007) concludes that the term “case study” can stand for either a single- or multiple-case study, where more than one case is investigated. It may even be the preferred option compared to a single case since the results are considered to be more absorbing (Yin, 2007). In the boundaries of the proposed research questions, a multiple amount of cases al-so allows researchers to conduct analyses within each setting and across settings, which helps strengthen the findings from the entire study (Baxter & Jack, 2008). The cases might have been chosen as replications of each other, deliberate and contrasting comparisons, or hypothesized variations (Yin, 2007).
Case studies have traditionally been underestimated among different science departments. Yin (2007) and Gummesson (2000) mention one of the critiques being concerned with the lack of statistical reliability and validity, which will be discussed later in the chapter. Sec-ondly, case studies have been claimed to be able to generate hypotheses but incompetent to test them. Thirdly, case studies are said to give a poor basis to generalizations.
The case study method was chosen as the main research strategy due to its consistency with the nature of the study. Gummesson (2000) argues that it is the most ideal strategy when aiming at studying processes, such as knowledge transfer in-depth. It can be best applied when research addresses descriptive, explanatory, and exploratory questions and aims to produce a first-hand understanding of people and events. Also, it can be assumed that the challenges stemming from knowledge transfer practices are highly influenced by the con-text. In this kind of situation the best way to explore the issues can be tackled through a re-al life case. By forming a multiple-case study of companies, we were aiming to gain a deep-er understanding about the challenges that the companies are facing.

Sampling design

Cases representing the best “fit” with the study design should be chosen. This is due to the fact that selected cases have a fundamental effect on the ultimate quality of the study and on the process of building conclusions from the case studies (Eisenhardt, 1989; Couche, 1997). The selected cases should illustrate some particular features or processes in which the researchers are interested in (Silverman, 2010) – cases where the events and processes are most likely to occur. Yin (2007) argues that the cases either a) predict similar results (a literal replication) or b) predict contrasting results but for predictable reasons (a theoretical replication). The amount of cases selected is determined by the study research objectives and the characteristics of the “population.” The three most common selection methods are purposive sampling, quota sampling, and snowball sampling. For this particular study, pur-posive sampling methods serves best. The cases chosen were intentionally representing lit-eral replications and selected using the purposive sampling method in order to find the best answer to the posed research questions.

Purposive sampling

Purposive sampling is a non-probability sampling method in which the decision about the most appropriate samples are done according to preselected criteria relevant to a particular research question (Morse, 1991; cited in Couche, 1997; Marshall, 1996; Saunders, 2007; Sil-verman, 2010). The probability sampling techniques, such as random sampling used for quantitative studies, are rarely appropriate when conducting a qualitative research (Mar-shall, 1997; Silverman, 2010).
Logically, cases with the most information should be chosen to be studied (Couche, 1997). The decision should also be based on convenience and the purpose that guided the selec-tion of KPMG and Grant Thornton. The selection criteria was based on of having knowledge-intensive companies operating within consulting that are accessible in terms of time and location. Personal connection to these companies had an impact in the selection process as well, which was also seen as an advantage of getting more information-rich data. Other companies that filled the criteria, but in the end were excluded from the study for consistency reasons, but were used as pilot studies. These pilot studies assisted in conduct-ing interview questions and directed the study.

Validity

The concept of validity assesses the truthfulness of the study findings. The concept can be further divided into internal validity and external validity, in which internal validity is more ap-plicable to explanatory studies and therefore not elaborated further (Yin, 2007). External validity is concerned with the extent to which the findings of the research can be general-ized, which has been a rather problematic topic in case study research. Case studies have been considered to give a poor basis for generalization, but according to Gummesson (2000) and Silverman (2010), this does not hold true any longer. They concur that there is no “golden key” to validity in qualitative research, but it is rather a matter of how you gen-eralize. For instance, instead of finding traditional generalizations to “populations,” it should be emphasized at what extent others are interested in applying the findings and conclusions in later stages (Yin & Heald, 1975; cited in Gomm et al., 2000).
Gummesson (2000) argues that the generalizability of qualitative case study findings are based on analyses to identify certain phenomena. Later, these findings can also be used as a comparison to previously developed theory (Yin, 2007). However, the generalizability is strongly dependent on the experiences of the reader that Gomm et al. (2000) describe as “naturalistic generalization.” The essence of this concept is that the readers will gain insight to the details and description presented in the case study, thus recognizing similarities to cases of interest to them. However, demonstrating the similarities and differences across a number of settings through the comparative case study approach can tackle the questions of generalizability.
The aim of this research is not to form generalizations to “populations,” but as the re-search question indicates, to gain understanding about phenomena in a specific context – consultancy companies. Since the study is purposively conducted as having knowledge-intensive consultancy companies in mind, it is realistic to expect that the findings can be generalized only up until limited extent. Accordingly, the research is seeking the “fit,” which makes the generalizability of the results be dependent on the readers’ understanding and capability to apply the findings to similar cases. From this perspective, one can argue for the generalizability of case studies.

Reliability

The goal of reliability is to indicate that the research, if repeated by later researchers, yields consistent results (Yin, 2007). By attempting to minimize the errors and biases of the re-searches, Gummesson (2000) claim that the research can be replicated by others and the re-liability increased. Since this study is especially interested in knowledge-intensive consultan-cy companies, it is doubtful that the results would be fully applicable in other kinds of companies, as indicated before. It is still highly expected that the same finding can be at-tained again. However, this possibility is dependent on the companies selected, the context, and how the companies are being studied. For instance, misunderstanding an interview question, or understanding but lacking in justification, may lower the reliability of the re-search. To avoid this, pilot studies were conducted.

Pilot studies

Pilot studies are often used in the context of exploratory research (Zikmund, 2000). The essen-tial benefit of conducting a pilot study prior to the larger study is that it can serve as a guide and give insight to the strengths and weaknesses of the procedures that the researchers are aiming to undertake. It disregards rigorous research and sampling standards and helps re-fine and formulate e.g. precise questions, testable hypotheses, and data collection plans with respect to both the content of the data and the procedures to be followed (Yin, 2007; Gummesson, 2000).
In the beginning, five service companies in total were selected. Conducting an interview with each of them clarified the interest of the study and increased the understanding of the best approaches and research methods to be undertaken, such as interview methods and questions. For instance, lack of prior contact with the interviewee often gave less in-depth answers and thus lacked in understanding the questions, especially via e-mail. This indicat-ed that the interview questions needed reformulation and clarification, and that a prior con-tact was essential. Secondly, interviews conducted face-to-face or over the phone gave the best results. The pilot studies helped discover new insights and ideas about issues that the researchers did not come across before. The understanding gained from the pilot studies served as a basis when formulating the official interview questions, and clarifying the re-search question.

Data collection

Primary and secondary data

The term “case study” is often taken to carry implications for the kind of data that are collected, and ana-lyzed (Gomm, Hammersley & Foster, 2000).
In order to conduct a comprehensive study and reach the research objectives, both primary and secondary data were used. Primary data can be characterized as new data collected particu-larly for the research being undertaken. The strength of using primary data lies in its ability to give fairly unbiased and direct first-hand information about the primary objectives of the research. In contrast, the collection procedure can be time consuming, providing a large amount of raw data that needs to be analyzed. Primary data was used in order to gain first-hand information. In general, primary data can be collected, for instance, through in -depth interviews, focus groups, surveys, and observations. The main way to attain this infor-mation in this study was through semi-structured interviews conducted with the consultan-cy companies. The semi-structured interview design gave the researchers more flexibility to ask probing questions and elaborate if needed. This was an important aspect to attain the richest information and also avoid misunderstandings of the questions.
Secondary data allows to build a comprehensive “body” of knowledge before starting primary research. It is rich information that already exists in the form of publications collected by other researchers (Easterby-Smith et al., 2008). This was beneficial since secondary data is already assembled data, and it does not re-quire access to respondents or subjects (Zikmund, 2000). The major advantages of attaining secondary data are that it can be obtained rapidly and with low cost. Howev-er, the fact that the data was originally de-signed to serve other needs can be problem-atic. Firstly, the data may be outdated, which requires accuracy from the researcher. Sec-ondly, variation in terms used can be ex-pected since in principle, each researcher can has the right to define the terms or concepts under study. Thirdly, different units of measurements may not be identical to re-searchers’ needs. Lastly, the data could be in-accurate (Zikmund, 2000). Secondary data such as scientific journals, academic handbooks, and models were used to construct a com-prehensive base knowledge for our study. The main sources of secondary data have been the Jönköping University’s library and online databases, such as Jönköping University’s database, Scopus, and Google Scholar.

Interviews

When the nature of the research is exploratory, it is likely to include qualitative research interviews (Cooper & Schindler 2008; cited in Saunders, 2007).
An interview is probably the most popular and important technique to obtain qualitative data. Saunders (2007) and Yin (2007) describe the situation as a guided discussion between the interviewer and interviewee. The essence of an interview is to gather valuable and relia-ble data that is relevant in attaining the research objectives. Interviewing as a data collection method is flexible in terms of time and content; an interview can be conducted face-to-face, over a phone, or online, and it can be tailored according to the research question.
An interview can be highly formal or informal (Saunders, 2007), in which the informal ap-proach was applied. This is because informal interviews are non-standardized and more flexi-ble in nature, and thus more appropriate for qualitative purposes. Non-standardized inter-views can be further divided into in-depth and semi-structured interviews, which are often re-ferred to as qualitative research interviews (King, 2004; cited in Saunders, 2007). Although in-depth interviews are optimal in collecting data about sensitive and hardly detectable phenomenon, a semi-structured interview was more appropriate since it is more guided with detailed topics, which was vital to the research question. In-depth interviews are usual-ly conducted in a few brief topics (Lee & Lings, 2008), which would have assumingly been too brief to get detailed information about specific matters.
The detailed topics and sub-questions are usually conducted in advance, still allowing the interviewer to “probe” when needed. The possibility of probing was one of the main de-terminants of choosing an informal interview approach and the most important tool in at-tempting to explore as complex a topic as knowledge. It also allowed the interviewees to talk more freely, and to reveal aspects that the interviewers had not come across before, therefore the decision was made to keep the interview semi-structured.
According to Saunders (2007), the role of semi-structured interviews is not only to reveal and understand the “what” and the “how” but also to explore the “why.” Therefore the benefit of choosing to use the semi-structured interview design is that the answers are also usually more complex than just “yes” or “no.” Another benefit of using informal interview-ing techniques is that the questions asked also depends on the flow of the interview and personal interaction. More open-ended questions may lead the discussion into areas that the interviewers did not previously consider.
The empirical data was collected during one- or two-time sessions with a representative from each company, having six in-depth interviews is total. The interviews endured for half an hour up to one and a half hours, depending on the comprehensiveness of the data. The interviews were recorded to avoid losing any valuable information, and to be returned to if needed. Thus, all the interviews were conducted in the interviewees’ terms, such as time and location, which also had implications to the amount of interviewees and interviews conducted. The limitations of chosen methods were basically focused on attaining primary data. As noticed in the pilot study section, the best way to conduct and interview was face-to-face. However, this was not always possible in terms of time and location. Therefore, some of the interviews were held by phone, which still allowed the interviewers to probe and elaborate. The lack in human interaction made the situation more formal, which argua-bly may had prohibited asking more sensitive questions. The sensitive and intimate nature of the data complicated the data collection process to some extent and may have excluded some valuable information. Firstly, it was paid attention to that there will be information that the interviewees may not want to elaborate too much in-depth. Thus, if they had, this information was considered to be rather intimate, and therefore excluded.

Empirical findings

In this section, the cases included in the study are presented through the company background followed by the interview conducted with the interviewees.

KPMG

Background

The case is focused on KPMG Jönköping and KPMG AB in particular, the latter being the Swedish member of KPMG International, the Swiss cooperative. In brief, KPMG is one of the world’s largest network of professional firms providing tax, advisory, and auditing ser-vices. In addition, the firm is considered to be one of the four biggest audit firms. The KPMG network was founded in 1987 as a result of a merger of Peat Marwick International (PMI) and Klynveld Main Goerdeler (KMG) along with their respective member firms (KPMG, 2014), whose history goes all the way back to the 19th century.
Although the head office of KPMG is located in Amstelveen in the Netherlands, each na-tional unit is seen as an independent legal entity and part of KPMG International Coopera-tive. The main purpose of this independence is to limit each members’ liability. KPMG is currently operating in 155 countries worldwide, and employing over 155 000 professionals across a range of disciplines. The main customers of KPMG are both national and interna-tional corporations, from small- to medium -size businesses, as well as the public sector, nonprofit and for-profit organizations. Their main focus is to help their respective custom-ers facilitate risks and spot opportunities.

KPMG AB Sweden

What is today called KPMG AB came into being when a company founded by Lars-Ture Bohlin in 1923, merged with KPMG in 1989. KPMG AB is represented in 60 different lo-cations employing 1600 employees altogether, and thus plays a central role in the Swedish market as an objective advisor (KPMG AB, 2014). As an addition to the values implement-ed by KPMG International, KPMG AB’s vision is to build and maintain a reputation as the best partner by developing their employees, their customer, and their society to its full po-tential (KPMG AB, 2014).
KPMG’s culture is rooted in their values. They are building trust and collaboration through the policy of open and honest communication, while their diversity and flexibility is aiming to reinforce a culture in which people can share knowledge freely, bringing out the best of each other. Their professional ethics, loyalty, and approachability are seen as major deter-minants among the customers when choosing to work with KPMG. Simply, their values are said to define what the firm stands for and how things are done there (KPMG, 2014). These values are aiming at creating a shared identity within the international organization. As an employer, the firm strives to imbue a global vision to be recognized as the “employer of choice”, aiming at recruiting, retaining, and developing the best professionals.
There are approximately 40 employees at KPMG in Jönköping, two of which are tax con-sultants, one working within advisory, and a few administrators. KPMG Jönköping also has one of the biggest offices in Sweden after Stockholm, Gothenburg, and Malmö.

Knowledge management at KPMG

KPMG is a knowledge intensive company that has recognized the significance of knowledge management, which results in it being one of the leading consultancy compa-nies applying knowledge management in its operations. It has internalized knowledge man-agement systems and practices, which are built around the values of the company (Kwiat-kowski & Stowe, 2001). Moreover, its knowledge-sharing culture is summarized by ‘turning knowledge into value for the benefit of its clients, its people and its community.’ (Kwiat-kowski & Stowe, 2001)
KPMG AB is centrally managing knowledge through an officially established knowledge department which is located in the company’s headquarter in Stockholm. According to their own definition, knowledge management comprises ‘the set of policies, procedures and systems associated with the creation, collection, safeguarding and dissemination of the firm’s intellectual capital.’ (Kwiatkowski & Stowe, 2001) With centrally established knowledge management, their main objective is to affiliate all firm employees and enhance collaboration among the network members in different geographical locations.
This particular study is based on interviews with the Chief Knowledge Officer of KPMG, Eva Winter, who is currently based in Stockholm and has been working for the company since 1982. In our research, Eva Winter represents KPMG’s knowledge management de-partment of 60 branches and 1600 employees across the country. She has been working with auditing and IT auditing among other things at KPMG. We have also conducted in-terviews with Anna Lexell who is working as an auditor at KPMG in Jönköping. She has been working at the company for 3,5 years and came to KPMG as a newly graduated stu-dent from Jönköping University. Anna Lexell is working with auditing in small, medium, and large companies in various industries. In addition, she is responsible for the coopera-tion with Jönköping University and local marketing.

Table of Contents
Acknowledgements
Abstract
1 Introduction
1.1 Background
1.2 Problem discussion and purpose
1.3 Research question
1.4 Perspective
1.5 Delimitation
1.6 Definitions
2 Theoretical framework
2.1 Nonaka & Takeuchi’s SECI model of knowledge conversion
2.2 Stages of the transfer process by Szulanski
2.3 Integrated knowledge transfer model
3 Methods
3.1 Methodological approach
3.2 Method approach
3.3 Research strategy
3.4 Data collection
4 Empirical findings
4.1 KPMG
4.2 Grant Thornton
5 Analysis
5.1 Challenges on an organizational level: Nonaka & Takeuchi
5.2 Challenges on an individual level: Szulanski
6 Conclusion
7 Discussion
7.1 Implications
7.2 Proposed solutions to prepare for and overcome challenges
References
GET THE COMPLETE PROJECT