In this chapter we provide the reader with insight to the motives for the steps chosen in the process of our study. We give a logical reasoning for choosing a qualitative research design and strategy. Thereafter we describe our data collection process and the method for our data analysis. We dedicate the final sections of this chapter to concerns regarding the trustworthiness and the quality of the research as well as the ethical aspects considered.
The research philosophy
According to Saunders, Lewis and Thornhill (2016) it is important to be aware of research philosophy and one’s own standpoint in the matter before deciding upon the research design and method. Since we believe that the truth is relative and dependent on the situation and its circumstances, thus what is true and works for one organization may not be true or appropriate for another, our philosophical assumptions align with the relativistic ontology and the constructionist epistemology approach (Easterby-Smith, Thorpe, & Jackson, 2015). From a position of constructionism one believes that truths are created by people and people have different viewpoints, hence the ‘truth’ of something is reached through discussion and agreement between the different viewpoints (Easterby-Smith, et al., 2015). The philosophical position that we now have taken is strongly linked to the research approach that we choose to apply for this thesis.
The research approach
Saunders et al. (2016) present the deductive approach that has its origins in natural science research. They also present the inductive approach that arose as an alternative for social science research when it emerged in the twentieth century. Induction is the more suitable approach when trying to understand how humans interpret their social world and gaining such knowledge is the true strength of an inductive approach. Through an inductive approach the researcher creates a theory or build on an existing one from what he/she finds in the data collected. Hence, induction means moving from observation of the particular case to broader generalization (Saunders, et al., 2016). Since our literature review generated little, close to no, previous research on the how the application of HK occurs in Sweden, we claim that an inductive approach to our study was justified in order to create a theory. According to Saunders et al. (2016), a decision regarding the research approach certainly affects the choice of research design.
The research design and purpose
According to Saunders et al. (2016) researchers that apply induction are more likely to use qualitative data. However, this is not reason enough to go for a qualitative research design. The second part of the purpose of this study is to investigate how Japanese subsidiaries based in Sweden have implemented Hoshin Kanri. Two of the research questions in our thesis; RQ
How do Japanese subsidiaries based in Sweden implement HK?, RQ 3: How does the implementation of HK differ between the companies?, are questions of how something works or seems. The research design that best served us in getting an answer to these questions was the qualitative research design. Qualitative data can be defined by on one hand their non-numeric form and on the other by the way in which they are created, which is through an interactive and interpretive process (Easterby-Smith, et al., 2015). In terms of research purpose, our research possesses the characteristic of an exploratory study (Saunders, et al., 2016; Yin, 2013). This can be explained by the purpose of our research, which is to investigate the application of HK in the Swedish setting. Exploratory studies are especially useful when wanting to clarify the understanding of an issue or phenomenon whose exact nature is unclear. Since we, after the literature review, perceived the picture of HK as indistinct and somewhat confused it further enhances our conviction that we are conducting an exploratory study. Moreover, the research questions that guide our study are, as mentioned earlier, questions that begin with ‘How’ which is also characteristic for exploratory studies. The way in which one conducts an exploratory research is through search of the literature, interviewing experts in the subject and through conducting in-depth individual interviews, three methods that we have embraced and completed. In exploratory studies the interview questions that are asked in order to explore the issue or phenomena often start with ‘What’ and ‘How’ (Yin, 2013). The interview questions asked for the purpose of our study (see appendix 5) are questions that indeed mostly start with ‘What’ and ‘How’. In conclusion, we chose to apply a qualitative research design with an exploratory nature because it enabled us to obtain a more in-depth knowledge about the implementation of HK, which was necessary in order to answer our RQ:s and fulfill the purpose of our study.
The research strategy and format
There are many different strategies associated with qualitative research. Some of these are; action research, ethnography, Grounded Theory, narrative research and case study research (Saunders et al., 2016). Of the strategies just presented, action research and narrative inquiry are the two alternatives that could serve well in conducting our study. According to Altrichter, Kemmis, McTaggart and Zuber-Skerritt (2002) action research is hard to define, or rather, the definitions that emerge do not stick because the nature of action research is so diverse and hence likewise is the perception of it. However, Altrichter et al. (2002, p. 128) provide a definition consisting of three points: ”(1) action research is about people reflecting upon and improving their own practice;(2) by tightly inter-linking their reflection and action; and making their experiences public to other people concerned by and interested in the respective practice.” Action research could serve the aim of our study well because the purpose of action research is to “promote organizational learning and produce practical outcomes through identifying issues, planning action, taking action and evaluating action”
(Saunders, et al., 2016, p. 190). Moreover, action research incorporates different types of knowledge, namely both abstract theoretical knowledge and practitioners’ knowledge such as experimental knowledge and knowledge that comes from practical application (Saunders et al., 2016). This aspect of action research matches well with the type of knowledge and information that we aim to collect for the purpose of our study. Finally, action research has implications beyond the research project in terms of providing information that can be used as a basis for change in the organization where the action research took place. Hence, this type of research strategy involved implications for practitioners (Saunders et al., 2016), which is something that we aim for with our study. However, action research was not the research strategy that we chose for our study because of the demand on collaboration with the company in question and above all its longitudinal nature, which requires more time than we have at hand.
Narrative research was another research strategy that we had to consider. Narrative research has its origins in people’s stories with the objective to explore, understand and concretize them to capture their experiences (Josselson, 2010). This particular research strategy would serve the purpose of our research well because it is based on participants providing complete stories of their experiences instead of having the researcher collect data in the form of pieces of these experiences from interviews. With the narrative research strategy the researcher gets to analyze a complete story rather than fragmented bits of data, which provides a clearer picture of the situation and deeper contextual understanding. A core aspect of the narrative research strategy is the aiming of preserving the chronological connection and the specific sequence in which events occurred as told by the participant (Saunders et al., 2016). For the purpose of our study, the chronological connection was not of importance. This important aspect of narrative research, together with the intensive and time-consuming characteristics of this particular research strategy, lead to the dismissal of this research strategy as a possible strategy for our study.
However, what indeed was of interest to our study was (1) some level of context, (2) getting access to in-depth information about the application of one specific phenomena (HK) and (3) preferably getting this kind of information from multiple sources in order to be able to do some kind of comparison between the cases and hopefully also generate or build on theory. In order to be able to this we chose to apply the case study strategy, and to be more specific – the collective case study research – to our study. A collective case study involves multiple (no less than three) cases within which the same research questions are examined, using the same methods for data collection and analysis for each case (Goddard, 2010). With the help of this research strategy researchers can undertake close studies of different cases that (must) share a link, either through different similarities, or as in our case, through applying the same strategic management system. Through an in-depth approach to each case the researcher can develop an understanding of each individual case, but also make comparisons of all the cases. Later on, in the phase of analysis, the different cases can, due to the link they share, be considered as a collective whole (Goddard, 2010). In this study the case is that of HK in Japanese owned companies in Sweden. Stake (1995) describes collective case study research as conducting multiple instrumental case studies. Cousin (2005) explains the difference between instrumental and intrinsic case studies: “Whereas an intrinsic case study aims to generalize within, instrumental case study attempts to generalize from a case study.” (p. 422). Explaining collective case studies, the author further states that these studies extend the attempt of instrumental case studies, namely to generalize from the case. According to Stake (1995) an instrumental case study is used, not for understanding a particular company or person, but rather to understand the specific situation, issue or phenomena that the company or person has insight into or experience of.
Hence, in the case of our study the Japanese owned companies are instruments for us as researchers to get to the true interest and focus of this study, namely HK. Based on figure 6 below, presented by Yin (2013), the type of case study that we conduct can also be referred to as an embedded single-case design (type 2 in the figure). For that particular type of case study we have a context, Swedish companies with Japanese ownership, within which we have the case, HK, and finally we have multiple units of analysis (or what Goddard (2010) and Cousin (2005) refers to as ‘cases’ and Stake (1995) refers to as ‘instruments’) embedded in that particular case, represented by three different companies for the purpose of this research. Hence, in order to avoid confusion we refer to HK as the Case (the focus of our interest) and to the companies as instruments.
According to (Goddard, 2010) there are limitations to this research strategy, which we, just as Stake (1995) chose to refer to as collective case study research. First, this strategy can be both financially straining and time consuming, moreover the time spent at each site or with each respondent should be more or less the same. Adding on, the researchers conducting the study need to be equally knowledgeable in order to ensure the quality during the data collection, but also in order to ensure validity for comparative purposes later on in the process. In our thesis the financial aspect has not been an issue. Regarding the time consumption it did not pose any problem since we knew the time restraint from the beginning and could plan according to that. Concerning the level of knowledge we were both novice in the subject of HK and have together conducted a literature review and structured the HK theory to gain an equal base of knowledge. However, there are benefits to the collective case research strategy as well. This particular research strategy, when properly applied, offers rich data that holds the potential to gain deep understanding about a phenomenon in its context (Goddard, 2010). For the purpose of our research the collective case study strategy is the most fitting one, and hence that is the one we chose. Going back to what was mentioned above as being of interest to our study, the chosen strategy helps us take these into consideration. With the help of a collective case study strategy we can consider (1) the context of the case, (2) the in-depth information both from individual sources and the collective and (3) multiple sources and by that come to a conclusion in the form of comparisons between participants and theory building.
The literature review
We started our research project by a systematic literature review (see appendix 1) followed by snowballing with focus on the most dominant scholars from the systematic review, combined with recommendations from an expert in the field of strategic management and HK. The literature review started at Google Scholar in order to get to know the literature and which keywords to use. When the keywords were found we moved on to Web of Science and conducted a search on our keywords, this resulted in 128 articles that after the refinement process went down to 19 articles. We used these 19 articles plus those received from the snowball process to create a knowledgebase to build our theoretical framework on. When we reviewed the literature we discovered that the picture of HK is a bit ambiguous which led us to our first RQ. We also discovered that studies had been made about HK in different countries but not in Sweden, which led us to RQ 2-4.
The data collection
For the data collection of our thesis we were inspired by the master-servant design, which is a mixed, quantitative and qualitative, methods design. This design implies that one method is used in order to serve the other (Easterby-Smith et al., 2015). In one variant of this design the quantitative method comes before and serves the, main and more dominant, qualitative method. In this variant a survey can be used in order to identify a smaller number of particularly interesting cases for a more in-depth investigation (Easterby-Smith et al., 2015). This is, more specifically, the variant of the master-servant design that inspired our data collection. We used a quantitative method, a survey, in order to map the field and serve the main, qualitative method, which were in-depth interviews. According to Easterby-Smith et al. (2015), the servant part of the method is to the most part not acknowledged in the results of the thesis. However, in this thesis we are going to present some of the survey results and briefly discuss it. Nevertheless the main reason for the survey is to find suitable organizations for in-depth interviews. Following comes an account of the respondents to the survey and the interviews, as well as an explanation of the methods applied and a description of how we practically proceeded with the research.
The selection and number of respondents
The aim of our survey was to map the application of HK in Japanese subsidiaries in Sweden and to generate candidates for in-depth interviews. It was a short (less than 5 min), online survey (created in Qualtrics, see appendix 6) that in the end gave us an indication of which companies that use HK. The initial questions had to do with the recognition and knowledge of HK. If the respondents answered that they had not or did not know if they had heard about HK or applied the system in their organization they were automatically sent to the end of the survey where we thanked them for their contribution. If the respondent had heard of HK before and where applying it in his/her organization than the survey continued. The following questions could be seen as the main questions, where we asked about HK and which steps in the process the person recognized. We ended the key part of the survey with the question; “Could you consider to participate in a more in-depth interview as part of our continuing work on Hoshin Kanri in Sweden?” and we used that question in order to detect possible companies for further analysis in the form of in-depth interviews.
Since the list over Japanese owned companies in Sweden, 158 companies, (see appendix 7) was just over one and a half year old we double-checked all the companies in the list in order to find out which of them that were still in business. There were 18 companies in the list that were out of business by March 15th of 2016 and another 14 companies which did not have any contact information (other than a box number) and/or had an unclear ownership or structure. At this point we were left with 126 companies that still were in business. We started to contact these companies by phone in order to get contact information to a person that could answer the survey either by mail or by phone. During this contact round, another 20 companies fell away due to that they were either out of business or that they did not have any working contact information. We were now left with 106 companies, which received the survey. One company answered the survey by phone and in that case we filled in the respondents answers in the online survey afterwards because of practical reasons. The remaining 105 companies wanted us to send the survey via e-mail. 14 companies answered the survey during the following ten days and when that time had passed we sent out a reminder to the remaining 91 companies. The reminder generated eleven new answers during the following week. We now had 26 companies that had answered the survey, out of these 26, only seven used HK and out of these only three were positive to continue with a more in-depth interview. We contacted the three companies that were positive to the interview and booked these, one by phone and two face-to-face, the last on turned out to not having any time for the interview, so it was cancelled. Since we wanted more interviews we also contacted those three companies that used HK but did not want to participate in an interview, this resulted in zero interviews. The number of interviews was still low so we decided to call every company that had not answered the survey again in order to get some more responses. This generated another eleven survey answers, six online and five via phone, were we filed their answered in the same way as after the first survey that were answered by phone. Out of these eleven only one company used HK and were positive to do an interview. So to conclude we contacted the companies at least four times, two times by phone and two times via e-mail. Some companies/persons were contacted more times due to some issues about whom that should answer the survey etc. This generated 37 survey answers and 3 interviews.
The master-aspect of the master-servant data collection design that we were inspired by is represented by a qualitative method for data collection. The choice of data collection technique for our research is strongly linked to the particular strategy that we apply. As mentioned earlier, in order to follow the guidelines when conducting a collective case study research we needed to gather in-depth information and knowledge about the application of HK in the Swedish setting. Therefore, we chose to conduct in-depth qualitative interviews, which have the aim of gaining an understanding of the respondent’s perspective (Easterby-Smith et al., 2015), with the participants in our study. As we proceeded with our data collection we kept in mind some reflections expressed by Brinkmann and Kvale (2005): Today many qualitative interview researches seems to use a lot of interviews as a way to reach a higher scientific level. This seems to be result of a misunderstood qualitative presupposition, namely that more interviews is equated with a more scientific research (Brinkmann & Kvale, 2015). The scholars argues that this is a defensive overreaction and that there is a general impression regarding currant research based on interview inquiry that these researches would be better off with fewer but better prepared and analyzed interviews.
The interviews that we conducted had predetermined questions in a questionnaire that were followed. This aspect implies that our interviews are of the structured kind. However, during our interviews the respondents talked quite freely and the questions asked at each interview where not exactly the same but rather dependent on the respondents’ answers and the situation at hand. Moreover, we did not record all the interviews conducted (because one respondent preferred not to do it) nor did we have a standardized schedule with pre-coded answers for the response. These are components that are important for a structured interview. Hence, despite the somewhat structured appearance of our questionnaire we argue that the interviews that we conducted where of semi-structured nature. Since, as mentioned earlier, the purpose of our research is to explore the theory of HK and moreover its’ application in practice, a semi-structured interview is quite fitting (Saunders et al., 2016). Semi-structured interviews are characterized by having some themes and key questions. The order of the questions may vary dependent on the participant and the flow of the conversation and additional questions may be needed in order to establish an understanding of the participants’ viewpoint. Semi-structured interviews often have an interview schedule, guiding the opening of the discussion but also to provide some closing comments. Interviews of this structure can either be audio-recorded or documented through taking notes (Brinkmann & Kvale, 2015).
Regarding the interviews that we conducted we had the same themes and key questions in the questionnaire mentioned earlier. We did switch the order of the questions depending on the participant and in some cases we also added some questions to clarify the answers we got. We conducted the interviews in Swedish since that seemed most appropriate and natural as Swedish is the native language of all three of our participants. We used audio-recording (except of in one interview) and made sure to take notes during the interview in order to, later on, properly capture and reflect on what the participant said. When setting up the interviews we first reached out to the respondent in order to agree upon a date and place for conducting the interview. When the date and place was set the respondents were contacted by email prior to the interview with the questions so that they could prepare themselves for the interviews. In one of our cases we set up time and date for a phone interview since the participant felt like that would be most convenient for him/her. Interviews like this over the phone or email are called remote interviewing. Their strengths are in that they offer a high flexibility in terms of time and location since the interviewer and interviewee do not need to be at the same location. However, the flexibility also implies a lack of immediate contextualization, depth and non-verbal communication that face-to-face interviews offer (Easterby-Smith et al., 2015; Brinkmann & Kvale, 2015).
Table of Contents
1.1 The Problem
1.2 The purpose and the research questions of the research
1.3 Delimitation of the study
2 Theoretical Framework
2.1 Historical background to Hoshin Kanri
2.2 The Hoshin Kanri literature
2.3 Hoshin Kanri and TQM
2.4 The Hoshin Kanri process in the literature
2.5 The steps of Hoshin Kanri
2.6 Catchball and the PDCA-Cycle
3. Research Method
3.1 The research philosophy
3.2 The research approach
3.3 The research design and purpose
3.4 The research strategy and format
3.5 The literature review
3.6 The data collection
3.7 The data analysis
3.8 Concerns regarding quality and trustworthiness for case study research .
3.9 Ethical dimensions
4 Empirical Data
4.1 Survey results
4.2 Interview results
5.1 RQ 1: What are the variations of HK in the literature?
5.2 RQ 2: How do Japanese subsidiaries based in Sweden implement HK?
5.3 RQ 3: How does the implementation of HK differ between the companies?
5.4 RQ 4: If there are variations in the implementation of HK, why do they exist?
6.1 Contribution and practical implications
6.2 Challenges and limitations
6.3 Further research
6.4 Connection to guiding principles
GET THE COMPLETE PROJECT
Hoshin Kanri – the Japanese way of piloting