• Ei tuloksia

states that stories must be in coherence with the data and, at the same time, emphasise an order and meaningfulness that are not apparent in the data itself.

During constructing my stories, I needed to rewrite and shorten them several times. After every round, I cut something that was not relevant to my research questions, but also kept some in order to maintain the attractiveness of the stories. As total, I found this very challenging. However, I discovered that after every round the stories became better and more fluent. To sum up, I finally turned the data of eleven interviews into four descriptive stories.

3.4 Evaluation of the methodology

Finally, after analysing the data, I am evaluating the methodology and methods that I applied.

Thereafter, I also refer to research ethic in terms of the interviewees’ privacy. First of all, I have chosen to accomplish a qualitative study since I can hopefully discover and describe better the multiple connections and lack of connections between branding, COO and the quality schemes through it, than by testing quantitatively some predefined relationships. Furthermore, since the connections and lack of connections were not previously discovered, it could have been hard to determine some predefined factors.

Secondly, the narrative study strategy and narrative data analysis that I applied are open to various interpretations. In this respect, Riessman (1993, 64) and Eriksson and Kovalainen (2008, 223) state that the classical methods of the research evaluations, such as verification and procedures for establishing validity based on the realist assumptions, are not so suitable for narrative research. According to Riessman (2004, 708), narratives are not mirrors of the past, but they refract it. Furthermore, she highlights that it depends on the imagination and strategic interests of the narrator how the events are connected and the meaning is generated. Similarly, Eriksson and Kovalainen (2008, 223) state that narratives offer only one version of truth by highlighting someone’s point of view.



During analysing my data narratively, I aimed to follow Riessman’s (1993, 64) evaluation criteria of coherence and pragmatic use of narrative analysis. Accordingly, I attempted to show that my interpretation of data is more than ad hoc. Secondly, I provided information that allows other to determine my work. In this respect, I have strived to describe the steps of my thematic and narrative analyses in a way that also other researchers could conduct the similar kind of study.

After all, why have I chosen a narrative strategy for conducting my research? I trusted and still trust in the idea that through narrative analysis the entrepreneurs’ voices are gaining the best hearing. Moreover, together with ProAgria (the cooperative partner of this study) we have found stories to be an easy access to identify with the results. Finally, I must also admit classically, ‘I love stories’.

Of course, three languages (Italian, Finnish and English), which were used to conduct and interpret the interviews, made the things more complicated. However, since I constructed the stories from the interview data by using narrative analysis it was possible anyway. However, if analysis of narratives had been applied, in which the focus would have been in the collected stories (Polkinghorne 1995) told, for instance, by interviewees, it would have been more complicated.

In any case, in order to add ‘trustworthiness’ (as Riessman outlines it 1993, 65) of the data, my native Italian co-researcher translated, held and interpreted the Italian interviews with me. In this way, I could have been surer that the interviewees understood and answered the questions without the language barriers. For the same reason, I accomplished the Finnish entrepreneurs’

interviews in Finnish.

In terms of choosing what to include my stories from the data, I considered Polkinghorne’s (1995) ideas of applying the expression of judging the credibility of a story. He explains this with a distinction between the accuracy of the data and the plausibility of the plot. He continues by saying that the researcher is in charge that the events really happened. In this respect, he suggests that triangulation methods can be used to verify the data. This means that several independent reports of the event are hunted for. However, even if the event is proved to happen by many



interviewees, the meanings of the answers can vary between the times that they give the answer to the interview and when the events have actually happened.

Therefore, during organising my data, I noticed that some themes occurred from the data only once and the others many times, e.g. both Finnish entrepreneurs described that they do not have enough time for the quality schemes. Therefore, I will illustrate all the findings based on themes and narrative analysis story by story later (See Chapter 4) in the result table, in which I have marked if the finding exists only in one story or whether it was described in some of them or in all of them. In this way, I will attempt to increase the transparency of my research.

However, I needed to group the interview data in order to fit it into four stories. In this respect, Riessman (2004, 706) argues that when many interviews are grouped into a similar thematic category, everyone in that group means the same and the ambiguities responses are not totally concerned. I think that it is perhaps true and that gave me a reason to generate all four stories instead of two for instance. I admit that it is a big amount, but in this way, I have tried to cover at least some of the ambiguities responses as well.

As next, it is good to take into consideration that I created the plots of the stories, as I have carefully explained above. Hence, it is also affecting ‘the credibility’ of the constructed stories and finally to the results. In this respect, Polkinghorne (1995) talks about judging of credibility of a story in a sense of the plausibility of the plot. He says that experiences are organised and tightened into a meaningful form, while conducting narrative analysis, even if they stand on their own also before. Moreover, the plots can vary according to interpretations of different researchers. Therefore, he states that since the researchers construct configurative analyses, it is not suitable even to ask if the story is ‘real’ or ‘true’. Instead, he ponders that value of the story depends on its capacity to offer understanding and insight into its reader.

In my case, the constructed stories (see the next chapter) do not represent a single entrepreneur’s opinion since the interview results of eleven participants are grouped into four stories, as I have explained already. In this respect, I want to highlight the modified facts of my stories that I did in order to secure the interviewees’ privacy. Firstly, the turnover and number of employees in my



constructed stories represent the average turnover and staff headcount of the interviewees’

companies involved in that specific story. Secondly, I have changed the years presented in the stories by adding or taking away a few years. In addition, I have not used the business related words, such as ‘added value’, in my stories if the entrepreneurs have not used them during the interviews. As contrast, when they have used these words, I have also kept them in my narrative analysis, and further in the stories, which I will present next.



4 ANALYSES AND RESULTS

As next, I will present four stories based on narrative analysis explained in the previous methodology chapter. After the stories, I will answer my main research question at first: How do the Finnish and Italian entrepreneurs describe the connections and lack of connections between branding, COO and the quality schemes? Then, I will answer my first sub-question: How do the entrepreneurs describe their branding?

Thereafter, I will summarise and illustrate the results for those two questions by means of the table. Lastly, I will focus on answering my second sub-question: How do the descriptions differ between the Finnish and Italian entrepreneurs?