Analyzing Collected Data

I organized and prepared the data for analysis by fully transcribing the interviews, studying my notes at each interview for additional context and observations at the time, then listened to each interview again and reading the transcript, before developing codes for analyzing the data (Creswell, 2009, Bryman and Bell, 2007). In the findings section of this research, statements of interviewees are illustrated in a form that filler words are left out to let the reader focus on the meaning of collected data and not be interrupted by grammar mistakes.

Also, I did not develop the codes before conducting the interviews or transcribing them. I used the approach that understands coding not as a precise measurement but as interpretation of data (Sadlana, 2009).

Coding the collected data is a way of seeing meaning in the information gained during the

interviews. Coding is a process of tagging or labeling units of meaning in text compiled

during a study (Miles and Huberman, 1994). Coding is accomplished manually and according to Sadlana, the researcher has more control over the qualitative data and more ownership over the work of the coding process by using paper and pencil, as it were, instead of using a computer at this point (Sadlana, 2009). My research required the analytical reflection that comes with lived experience of social interaction. After I had completed the coding, I used the software program Nvivo

¹

, which offered me a tool for qualitative data analysis. I used the Nvivo software more as a technical tool than as an analytical tool, mainly to filter and categorize the collected data.

Overall, I developed eleven codes, and all information given by participants was captured in those codes, categorized for meaning. The eleven codes I developed are the following:

innovation, strategy, sustainability, slow fashion, Copenhagen/Denmark, business model, tendencies, transparency, craftsmanship, obstacles and education. The codes can be viewed in Appendix A1.

Thereafter, I analyzed the codes again and grouped them into the same meaning categories.

Codes that share the same meaning were grouped together and by doing so I categorized them into overarching themes. The overarching themes that emerged help when trying to generate theory from the collected data. This way of organizing and analyzing data is influenced by Saldana’s “streamlined codes to theory model” (Saldana, 2009, p.12). The model below illustrates the organization of data analysis in the “Codes to Theme Model”.

“Codes to Theme Model”

Theming the data is an outcome of coding and categorization which results of human analytical reflection in this research (Sadlana, 2009 p. 13). Theming the data helps to understand a phenomenon and aids to comprehend the data and codes that have been

1 “Nvivo is qualitative data analysis software for researchers working on Windows and Mac operating systems” (qsrinternational.com, 2016)

Theme

Code

Code

Code

collected (Sadlana, 2009; Creswell, 2009). Using the coding process to generate themes for the analysis can build an additional layer for the complex analysis (Creswell, 2009).

This research uses Nvivo as a program to structure and organize the transcriptions and code the collected data. As mentioned before, Nvivo is thus not used as an analytical tool, but only as a practical tool. Nvivo is used because the researcher believes that the program can mark and filter the data in a more reliable and faster way, than the researcher would be able to do by hand. In other words, the extent of the interviews and transcription is so large that this program is assisting when organizing the material (King and Horrocks, 2010).

All the coded content can be found in what are called “nodes”. When clicking on the node the researcher can view all content that has been collected under one specific code. For example, one code in the analysis is labeled “sustainability”. All the content from all the interview transcriptions, coded as “sustainability“, will be found in that specific node (qsrinternational.com, 2016 (see example in Appendix A4)). The software program allows the researcher to see the original source of any piece of content collected within a particular node. The software lets the researcher analyze the material within a node to see developing patterns and themes in the collected data.

Saldana states that theming the data is as intensive as coding and part of a strategic choice and requires reflection on participant meaning and outcomes (2009). Further, this research leans on Rubin and Rubin’s idea that (1995) themes are statements of ideas “presented by participants during interviews, or conceptual topics developed by the researcher during a review of the data”. This is part of the analysis when reviewing the data (Sadlana, 2009).

Themes are furthermore used to support the process of interpretation (Sadlana, 2009).

Thus, three major themes I developed in the process can be seen below:

•

Innovative Strategy

•

Slow fashion

•

Denmark

After the analysis and critical reflection upon the eleven codes developed, I created three overarching themes, which made it easier for me to be more focused on the “reality” of data.

One can say those themes are like headings that frame the analyzed data. According to Saldana, “when the major categories are compared with each other and consolidated in various ways, you begin to transcend the “reality” of your data and progress toward the thematic conceptual and theoretical” (Saldana, 2009, p. 10). As mentioned before, overarching themes emerge during the analysis and “our ability to show how these themes and concepts systematically interrelate lead toward the development of theory” (Corbin and Strauss, 2008, p.55 in Saldana 2009, p.11). Throughout the analysis a reflection on theory and concepts helped when exploring similarities between earlier research and new findings, which serve as a foundation for the development of new theory. There are some disadvantages to consider when carrying out an inductive approach, as the point of departure is taken in the data, and not in theory and there is a risk that no relation with theory can be identified in the end (Saunders et al., 2009). When writing the findings, the content of coded and themed data that is relevant for this research will be illustrated by quotes. This way, when elaborating on the findings, the reader gets a clearer idea of the analyzed content.

I approached my research using some features of grounded theory, starting off with general research questions, conducted relevant data collection, coded the data and then categorized and themed the results (Bryman and Bell, 2007). And while I could proceed with developing a theory from the collection and analysis of the data that was related to the grounded theory approach, I considered it important to remain sensitive to already existing concepts. And due to practical difficulties like a tight deadline and because I questioned whether a new theory could be developed, the method of the grounded theory was not fully applicable to my research (Saunders et al., 2009 and Bryman and Bell, 2007).

Interview Questions

At this point it is important to mention the two dominant research questions again.

1. What are the tendencies regarding slow fashion in Denmark?

2. How can Denmark-based fashion companies identifying with slow fashion, use slow fashion as an innovative and competitive strategy?

For answering the research questions, the interview questions where of substantial

importance. This will be investigated below. Please see appendix A2 for more details.

In document SLOW: A new take on the fashion industry Denmark as a pioneer (Sider 35-39)