The twenty in-depth interviews were analysed using thematic analysis. The reason for using this method was that it allows for qualitative research methods to be interpreted in a way that brings meaningful results (Nowell, Noris, White, & Moules, 2017). Data analysis is crucial for the whole research, forming the most complex part (Thorne, 2000), therefore, a basic overview of the method used will be now provided.
2.5.1 Introduction to thematic analysis
Braun and Clarke (2006, p. 79) define thematic analysis as “a method for identifying, analszing, organising, describing, and reporting themes found within a data set.” The authors also state
54 that thematic analysis offers an answer to a question, even when the question may become apparent only in the analysis itself. Besides this, thematic analysis provides core skills for conducting also other types of qualitative research.
There are two approaches to the analysis – either it is possible to focus on obvious patterns or to go beyond the obvious and try to find patterns that are not noticeable at first glance. In this thesis, both approaches are combined. Also, inductive and deductive analysis can be distinguished. Inductive analysis focuses mostly on coding the data, whereas deductive is based on the knowledge of the theoretical base of this topic.
As any method, also thematic analysis poses both advantages and disadvantages. However, for analysing this type of data, advantages prevail. First of all, it offers a more accessible type of analysis, as it does not require any special theoretical or technological knowledge. Very importantly for the interviews with D&I coaches and experts, thematic analysis is seen as a very suitable method for examining the various views of the interviewees and finding and comparing the differences and similarities (Braun & Clarke, 2006; King, 2004). One disadvantage that may arise from thematic analysis is that despite its useful flexibility, it can also lead to inconsistency when identifying themes (Holloway & Todres, 2003).
One of the most important things to reach in the thematic analysis is trustworthiness, which can be further divided into several criteria that help to reach it. Those criteria are credibility, transferability, dependability, and confirmability (Lincoln & Guba, 1985). Credibility is a representation of a coherence between the views of the respondents and how they are then provided by the researcher (Guba & Lincoln, 1989). Transferability means how the results can be generelised and transferred. Dependability is about researchers making sure that the whole process is clearly documented and that it is intelligible. Lastly, confirmability is the attempt to ensure that all the conclusions of the researcher are directly derived from the data (Tobin &
Begley, 2004), and that it can be achieved when credibility, transferability and dependability are reached (Guba & Lincoln, 1989).
2.5.2 Process of thematic analysis
In the following chapter, I will illustrate the process of the thematic analysis, based on the elements of trustworthiness described above. Both theoretical framework for the process will be provided, as well as practical insights from the interview analysis with D&I coaches woven into this chapter.
55 Thematic analysis is composed of 6 basic steps; Familarising with the Data, Generating Initial Codes, Searching for Themes, Reviewing Potential Themes, Defining and Naming Themes, and Producing the Report (Braun & Clarke, 2012).
The first step is to familiarise with the data. If one is engaged in the interactive data collection, there is a high likelihood that the data analysis phase has already been approached with some knowledge of the analysed data, together with some initial thoughts about the interview process.
When such thoughts are documented during the process of gathering the data, here for interviews, the actual data analysis may begin during the collection phase, as some initial hints and questions may arise.
In the case of the interviews with D&I coaches and experts, some topics such as opinions on measuring and evaluating the outcome of the sessions, or on whether the person should look diverse on the outside if they want to focus on D&I, arose often during the interviews and were therefore discussed further, although the interviewees were not asked about some of those topics initially.
This step of the analysis then involves actively reading the data and searching for patterns. In fact, Braun and Clarke (2006) advise to look at all the data before starting with the coding part, as it might largely contribute to the identification of the mentioned patterns. Besides that, it is also recommended to be transparent in the whole data collection process and to write down the reflective ideas about the researched topic as well as some initial ideas for coding (Lincoln &
Guba, 1985; Sandelowski, 1995).
Secondly, the phase on generating initial codes follows. According to Savage (2000, p. 1496),
“coding is a process of reflection and a way of interacting with and thinking about data.” This phase allows to shift the attention towards any particular characteristics.
During this stage, all 20 interviews were properly reviewed and codes for specific topics were identified. This allowed for a better understanding of repeating patterns and served as a basis for themes identification.
As a third step follows searching for themes. In this phase, themes are produced rather than passively discovered (Braun & Clarke, 2021). According to Braun and Clarke (2006, p. 86), a theme “captures something important about the data in relation to the research question, and represents some level of patterned response or meaning within the data set.”
In practice, this part took place via reading through the previously coded data and searching for larger themes and then combining codes that show similarities or might even overlap in some cases.
56 Fourthly, the thematic analysis is about reviewing potential themes, especially from a quality point of view, and discovering whether the previously identified themes are related to the data.
In this step, some of the themes were merged together, as tight correlations were found. Also, others were deleted and some codes were relocated to another theme, as they were found to better fit there, and as more contributing to the research question and goal of the thesis. Lastly, some themes were identified as codes instead and therefore, no need for another separate theme creation was seen. On the other hand, some of the themes were too broad and therefore split in more themes. In the end, themes that capture the most important parts of the interviews and contribute to the answer of the research question were kept.
The fifth step in the process is to find suitable names for the themes and define each of them.
The name of the themes should not only provide information about the theme, but also be eye-catching. In this step, it is particularly important to be able to demonstrate why a specific theme was chosen and what within the quote should draw attention. (Braun & Clarke, 2012)
In practice, quotes from the interviews were selected to demonstrate the story behind each of the themes, and to explain their importance within the context of the whole research topic. Also, some sub-themes were identified. Each quote was chosen to clearly demonstrate the purpose of the theme. Additionally, some of the selected quotes were kept more extensive, in order to demonstrate the point of the interviewee more vividly. Attention was also paid to selecting quotes from multiple interviewees, to show that the themes were present across the whole dataset.
Importantly, where applicable, also theoretical frameworks and previous research tied to the identified theme are provided within the Research Results & Discussion section. Please note that in some cases, previous research is not included, as the topic of D&I coaching is very recent, and therefore, does not offer much in the way of previous research focusing on this topic. As a result, this thesis should bring new and useful findings in the area of diversity and inclusion coaching.
The final sixth step of thematic analysis is to amalgamate all of the themes into a written report.
In this step, all previous analyses is combined and put into writing. Here, it is crucial to organise the themes in a logical order, which allows the latter themes to be built on the previous ones.
In this last step, themes were put into writing and organised in a way that they make sense for the reader.
57
3 Research Results & Discussion
In this chapter, results from the interviews will be discussed. To keep a logical structure when going through the interview results with the use of thematic analysis, results will be presented in the following way. There will always be a section of the interview (see section Interview Structure for more details), under which themes will be introduced, further commented on and quotes from the interviews will be provided.
In the entire dataset consisting of 20 interviews with experienced D&I coaches and professionals, nine themes were identified. Those identified themes are as follows: Experience of exclusion/difference as the main career motivation; Coaching for D&I does not mean having a “D&I Coach” title; No unified D&I Coaching process?; Voluntary or Mandatory? That is the question; Resistance, the fear of the privileged group; Trust as a corner stone; Building the business case for diversity; We want to measure, but we can’t; and One-on-one. Each of the themes will be described below in greater detail with illustrative examples from the interviews.
On top of the identified themes, also best practices and common mistakes perceived by each coach will be provided, as well as one main recommendation from each interviewee. These three sub-chapters do not have any common theme, but do however, bring many valuable insights from 20 different perspectives of both D&I coaches and experts operating in various industries and being based in different parts of the world.
In the graph below, you can see a basic overview of themes identified and codes used for each of those themes.
58 Figure 8 - Overview of Sections, Themes & Codes
Source: Author
In the following chapters, each of the themes will be described in greater detail.