Supplier relationship

Get Complete Project Material File(s) Now! »


In this chapter there will be a description of how this research was conducted and why. The major parts in-cluded are research approach, research design, selection of cases, method of data collection as well as method of data analysis. Lastly, there will be a discussion regarding trustworthiness.

Research approach

In order to achieve the purpose, a qualitative approach was taken, since it provides a deep-er understanding for the topic studied (Holme & Solvang, 1997). The qualitative and the quantitative are the most widely used approaches (Saunders, Lewis & Thornhill, 2009). One of the main differences between the two approaches is that a qualitative study is based on opinions and meanings motivated by words, whereas a quantitative study is based on num-bers and statistics. Furthermore, quantitative studies result in numerical and statistical data, are based on many respondents, while qualitative studies require classification into catego-ries, and are based on a relatively low number of respondents (Saunders, Lewis & Thorn-hill, 2009).
Another type of decision that needs to be considered is whether to use the deductive, in-ductive or abductive approach (Alvesson & Sköldberg, 2008). The approach devoted in this thesis is the abductive, which is commonly used in case studies. An abductive approach was chosen as it allowed us to collect theory and empirical data simultaneously, which im-plies a learning loop (Pedrosa, Näslund & Jasmand, 2012). Thus, we were able to change and add theory depending on the empirical data we collected. As can be seen in Figure 3.1 the abductive approach is influenced by both the inductive and deductive, but unlike the other approaches, the abductive approach has the advantage to imply understanding (Al-vesson & Sköldberg, 2008).

Research process

The research started with the first version of a theoretical framework, which was used in designing the interview questions. Then interviews were carried out, and after each inter-view we reconsidered and updated the theoretical framework if something new were found that we had not included before, hence new theory were added. Thus, the interviews and theory matching were done simultaneously. When all the interviews were executed and the theory was matched, we analyzed the empirical findings together with the theory. That re-sulted in suggested new theory and conclusions. At last, the conclusions were applied on the companies interviewed. The process is shown in Figure 3.1.

Research design

When choosing research strategies, there are a number of factors that need to be taken into consideration, which are presented in Table 3.1.
To begin with, the research questions that are stated in order to achieve our purpose should be paid attention to. As this thesis aims to answer the two research questions focus-ing on “how”, only experiment, history or case study appear to be appropriate. Since we neither have the attention to control the behavior of our interviewees nor to study a con-temporary event, we chose to conduct a case study. This is also in coherence with Christo-pher et al. (2011), who argue for the case study as being the most appropriate choice when conducting research within logistics.
In the beginning of the study, three cases were included. Since that appeared not being enough, other cases where added gradually until data saturation was considered reached. This is in accordance with the abductive approach, since adding new cases are necessary in order to test and develop theories (Danermark, Ekström, Jakobsen, & Karlsson, 2003).

Selection of cases and respondents

In order to select cases we used a few criteria. Since the purpose of this study is to study small companies, they had to fit with the definition of a small company (European Com-mission, n.d.). Furthermore, we chose companies within different industries, manufactur-ing, wholesaling and fashion, with the intention to get a broader view. The website was used in order to screen for companies located relatively close to Jönköping. On the one hand, it was practically impossible to cover the entire Sweden within the study, and on the other hand, there is no guarantee that the result would have been bet-ter. Therefore, we argue for only using companies in the region of Jönköping. However, this will have a negative impact on the generalizability of the study to other companies in other parts of the country. After the screening, we called the CEO of the companies we found. Since our purpose is to study global sourcing risk, we only proceeded with the com-panies that had at least one supplier located internationally, which they purchase from on a relatively regular basis.
The selection of respondents was made together with the CEO of the companies. Our in-tention was to speak to the CEO as they usually have a holistic view of the company, the purchasing manager, as they presumably work within purchasing daily and the logistics manager, as we believed that they are familiar with our topic. However, since we are focus-ing on small companies, not all of them had an employee for all the requested positions. A limitation of this study is that in three out of the seven companies, only one respondent was interviewed. The reason for not including more respondents was simply that there were no more employees involved in the purchasing and logistics processes. This can have an impact on the study, since the respondents can have biased the answers. The respond-ents and companies are all listed in Table 3.2. Since some of the companies and respond-ents would like to be anonymous in the study, the respondents will be referred to as (Com-pany A, B, C etc.) in the empirical findings chapter. This means that the reader will not be able to link the label, neither to the companies nor to the respondents. Consequently, two respondents from the same company will be referred to by the same label, and thus not able to distinguish. We do not believe that this could have any major influence on the study, since it is of little importance to distinguish between the different respondents. It is of greater interest to be able to differentiate the companies.

Data collection

The data collection was done by using face-to-face semi-structured interviews as well as one interview via email. According to Yin (2009) interviews are one of the most important sources when it comes to case studies. We chose to conduct interviews as it is possible to focus directly on the research questions and, in addition, it provides insights since infer-ences and explanations can be given. By conducting semi-structured interviews, infor-mation can be obtained in a flexible way as questions can be added to provide a deeper explanation. In addition to that, certain questions can be excluded if the interviewee does not feel comfortable answering the specific question or does not think it will provide input to the study (Bailey, 2007). One of the drawbacks with interviews is that the formulation of questions can lead to a poor result (Yin, 2007). To reduce this risk, we based our interview questions (see Appendix 1) on questions used in previous studies within the area (Black-hurst et al., 2005; Christopher et al., 2011). Two other disadvantages are the variations in how the different interviewees answer the questions and that they answer the questions in the way they think the interviewer would like, instead of being totally honest (Yin, 2007). However, these drawbacks are harder to prevent. As we only did face-to-face semi-structured interviews and one interview via email, we only collected primary data (Saunders et al., 2009). The interviews were carried out in Swedish, meaning that the authors of this thesis translated everything to English. For example, this means that, the quotes were not translated literally. This might have influenced the empirical findings as well as the analysis and conclusion. However, the empirical part was sent to the case companies in order for them to check for possible misunderstandings. The interviews contained mostly open-ended questions, which enabled the respondents to answer them in more detail and crea-tively. Furthermore, the questions were phrased to stimulate unbiased, truthful and unaf-fected answers.
Interim summaries were done during the collection of the data in accordance with Saun-ders et al. (2009). Due to the summaries, we could see what had been found so far, if we had to add something to the theoretical framework and it also enabled us to change the questions for the rest of the interviews if we found it necessary. As we got the opportunity to audio record the interviews, both of the researchers could fully concentrate on the re-spondent and the interview. It also allowed us to use direct quotes for the empirical data (Saunders et al., 2009).

READ  Data plane offloading on a high-speed parallel processing architecture 

Data analysis

A case study-based research requires analysis of the data, and it often involves categoriza-tion, interpretation and inferences (Pedrosa et al., 2012). The data analysis for this study was carried out by following a model in Roulston (2014), analyzing and representing inter-view data: practical steps. The first step was to reduce data to locate and examine phenom-ena of interest, which was made by eliminating repetitive statements and data irrelevant to the topic. In order to carry out this step, we colored the parts in the transcripts with differ-ent colors depending on what answers that belonged to which part in the theoretical framework, for example, risk identification, risk assessment and risk mitigation. Thus, the uncolored text was reduced. Thereafter, we proceeded to the second step, which includes reorganizing, classifying and categorizing data. This was done by re-coloring the important information by respondent instead, so every respondent was tied to a specific color, and that was done in order for us to be able to see which information that was received from which respondent/company. Then we printed the colored information and cut it into small pieces and placed them in categories where they fitted, having the theoretical framework as a base. This was done for the mitigation part of the thesis as we realized that it was the part where we got the most information. The categories were structured in the same way as the mitigation part in the theoretical framework. Then in the third step, interpreting and writ-ing up findings, the researchers should consider assertions and propositions in light of pri-or research and theory in order to develop arguments. This was done by comparing the prior research with the empirical findings.


In order to measure the quality of research, some criteria can be evaluated. We chose to base mainly this part on Pedrosa et al., (2012), who analyzed 134 case study- based articles in the field of logistics and supply chain management in order to assess and evaluate the quality of the articles. Thus, we found it relevant to use this article as a base for the quality discussion of this thesis. Trustworthiness is divided into three parts; transferability, truth-value and traceability.
Transferability refers to “the extent to which a study’s findings apply to other contexts” (Pedrosa et al., 2012, p. 278). It is similar to external validity, which refers to generalization, however, transferability acknowledges that time can change context and people, which is a limit to the opportunity to generalize findings. Pedrosa et al. (2012) present some actions that could be taken in order to enhance the transferability. In this thesis, Pedrosa et al. (2012) were followed by explaining how and why the cases were selected, which is neces-sary to provide the reader with insight into the suitability of the cases for the research pur-pose, questions and the context studied.
Truth-value refers to “the match between informants’ constructed “realities” in their par-ticular context and those represented by the researcher” (Pedrosa et al., 2012, p. 278). It emphasizes the importance of respondents correcting and/or confirming the researcher’s interpretation of their answers. In order to enhance the truth-value of this thesis we sent a draft of the empirical findings to the respondents, which allowed them to confirm that the information was true and that we were allowed to use it in the thesis. We also described the data analysis process, which allows assessment of how the data were processed to generate the findings, and therefore to judge the truth-value. Both of the methods used are in line with what is presented in Pedrosa et al. (2012).
Traceability relates “to the documentation of the research process and data sources” (Ped-rosa et al., 2012, p. 279), and it includes for example dependability. In order to enhance the traceability, a description of the data collection technique and process, and an explanation of how and why the respondents were selected are important (Pedrosa et al., 2012). Those recommendations were all followed in this thesis.

Ethical considerations

In order to comply with general ethical considerations, the respondents had the opportuni-ty to be anonymous if they wanted to. Moreover, they could choose not to answer the questions and could end the interview anytime they wanted. All the aforementioned meth-ods are in line with the recommendations in Saunders et al. (2009).

Empirical findings

The empirical findings are divided into three parts. To begin with, there will be a discussion regarding risk identification, where tables of both supply risk and environmental risk are presented. Then, findings related to risk assessment and risk mitigation will be discussed. In other words, the empirical findings will follow the same structure as the theoretical framework.

Company background

In Table 4.1 a short description of the case companies is provided. As presented before, the respondents and the companies will be referred to as Company A, B, C, D, E, F and G and the labels were added in a random order.

Risk identification

Most of the case companies lack a process for risk identification (Company A; B; C; E; F). According to Company A, they do not identify risks as they never experienced any issues related to supply, whereas Company B claims that the reason for not identifying risks is that they have not searched for any new suppliers lately. However, this is likely to happen in the future. Furthermore, it is stressed that lack of time is a reason for not identifying risks (Company B). The only company that identifies risks is Company D, which stresses that they are aware of existing risks.

Supply risk

The conception of the existing risks, both in terms of supply risk and environmental risk differs between the companies in the study. In Table 4.2 the supply risks that are men-tioned are listed.
The risks listed in Table 4.2 are emphasized with different severity and with different fre-quency among the respondents. Payment in advance appears to be rather common practice when purchasing internationally. According to all the respondents, payment in advance is considered being one of the most significant supply risks. Longer lead time, quality of products and currency fluctuations are other risks several respondents mention (Company A; C; D), whereas the remaining risks in the table are only referred to by either of the com-panies (Company A; B; C; D; E; F). Several of the companies stress communication as a potential risk (Company D; G), however the only company considering it being an issue is Company G. It claims that problems related to payment are more complex and time con-suming to solve if the supplier is located far away. Furthermore, Company D stresses that a problem might occur if the contact person is replaced, at least until the new contact person gets familiar with the companies operations. Another aspect that all the companies have in common is that they solely purchase from foreign suppliers that are larger or significantly larger than they are (Company A; B; C; D; E; F; G), which could be a risk. In addition, it is mentioned that it happens occasionally that their orders are not prioritized, but rather de-layed because of the small quantities ordered (Company G). Similarly, Company F men-tions that if they orders too small quantities they are not considered being a customer, which their suppliers are interested in working with.

Table of Contents
1 Introduction 
1.1 Background
1.2 Problem statement
1.3 Purpose
1.4 Definition of key terms
1.5 Disposition of the thesis
2 Theoretical framework
2.1 Risk identification
2.2 Risk assessment
2.3 Risk mitigation
2.4 Supplier relationship
2.4.4 Trust
3 Method 
3.1 Research approach
3.2 Research process
3.3 Research design
3.4 Data collection
3.5 Data analysis
3.6 Trustworthiness
3.7 Ethical considerations
4 Empirical findings 
4.1 Company background
4.2 Risk identification
4.3 Risk assessment
4.4 Risk mitigation
5 Analysis
5.1 Global sourcing and international purchasing
5.2 Risk identification
5.3 Risk assessment
5.4 Risk mitigation
5.5 Risk handling in small companies
6 Conclusions 
6.1 Theoretical implications
6.2 Managerial implications
6.3 Limitations and further research
C ompanies’ action s towards global sourcing risk A multiple case study on small companies in Sweden

Related Posts