Data Analysis

Qualitative analysis does not consist of a single approach but includes various methods that share a similar goal in that they seek to arrive at an understanding of a particular phenomenon from the perspective of those experiencing it (Al-Busaidi, 2008). In the present research, thematic analysis was used for identifying, analysing and reporting patterns or themes within the data. This technique is a tool for the organisation and description of the source information as well as for the interpretation of the various aspects of the research findings (Braun and Clarke, 2006). It constitutes a foundational form of qualitative analysis and it includes core elements which are widely used in almost every other method of qualitative research. Indeed, Holloway and Todres (2003: 347) identify the process of ‘thematising meanings’ as one of a few shared generic skills across qualitative analysis. With this method, it is possible to classify words, phrases and sentences into content-related categories (Sandelowski, 1995).

There are some terms which are related to this method. The ‘unit of analysis’ commonly refers to a whole interview or an observational protocol. Graneheim and Lundman (2004) recognise the following terms of gradual coding as more data are accumulated in each grade. A ‘content unit’ is a constellation of statements which relate to the same meaning. Furthermore, the parts of the text which deal with a specific topic are referred as a ‘content area’. A ‘category’ is a group of contents that share a commonality. Categories must be complete and exclusive. This means that no data related to the purpose of the study should be excluded due to a lack of a suitable category and no data should fall between two categories or fit into more than one category. Moreover, a ‘theme’ is defined as threads of meaning that describe a certain aspect of a

phenomenon. Of course, ‘sub-categories’ and ‘sub-themes’ are also described for pieces of data and include an intermediate content. In this way, thematic analysis processes narrative materials by breaking the text into relatively small units of content (Sparker, 2005). The goal of the above process is to filter the textual data to reveal the ‘essence’ of the phenomenon which derives from the text (Sandelowski, 1993b).

Thematic analysis processes data in great detail as it allows the researcher to understand a phenomenon from its elemental components up to its associations with other situations within the context they appear (Thomas and Harden, 2008). This process gives an opportunity to understand the potential of any issue more widely (Marks and Yardley, 2004). Thematic analysis is capable of detecting and identifying participants’ actions and thoughts and constitutes a source for explanations of phenomena (Alhojailan, 2012). In addition, it allows the management of data, starting from a precise content and then moving to broader conclusions and ultimately to theories (Alhojailan, 2012). Due to the flexibility of this method, researchers are free to make their subjective assumptions (Bristowe, et al, 2015) provided that they are clear about what they are doing, how and why (Attride-Stirling, 2001). Finally, Vaismoradi and colleagues (2013) argue that this approach is robust enough to be used for conducting an introductory study on a novel phenomenon, for which the quality of its data depends on the amount of energy and time the researcher spends on the process of data collection and analysis.

The raw data of the present research consisted of the interview transcripts. As DeSantis and Ugarriza (2000) suggest, the thematic analysis in the present study involved the search for, and identification of, common threads that extended across a set of interviews. The process of analysis started with the initial search for, and identification of, patterns of meanings and issues of potential interest in the data and was completed through reporting established themes and providing relevant interpretations. The present study was mainly concerned with producing knowledge based on existing theory. Therefore, the data analysis – that is, the search for, and formation of, themes – was guided by the theoretical framework presented in Chapter 6. However, the researcher was alert during the analysis to notice and analyse other novel patterns or themes.

The analytical procedure in the present study mirrored the phases identified by Braun and Clarke (2006) and Burnard (1996) and is summarised in Table 7.2 below:

Table 7.2: Phases of the Thematic Analysis in the Present Research

Phase Brief description of the process

1. Familiarising with the data

Transcribing data, reading and re-reading the material, noting down initial ideas.

2. Generating initial codes

Coding interesting features of the data in a systematic fashion across the entire set of information, collating data relevant to each code.

3. Searching for and creating themes

Collating codes into potential themes, gathering all data relevant to each potential theme.

4. Reviewing themes

Checking the themes work in relation to the coded extracts and the entire data set, generating a thematic plan of the analysis.

5. Defining and analysing themes

Ongoing analysis to refine the specifics of each theme, generating clear definitions and names for each theme, analysing each theme in detail.

6. Writing up

Selecting vivid and compelling extract examples, refining the analysis, relating the analysis to the research questions and literature, producing a scholarly report.

Source: Braun and Clarke (2006) (modified)

The application of these analytical phases in this study is described in more detail below. However, it is important to note that the process was not linear but iterative with constant moves back and forth between the entire set of information, the coded extracts of the transcripts, the emerging categories/themes and the relevant notes, as is usually the case with qualitative analysis (Vaismoradi, et al, 2013). This process reflects the method of constant comparison. Following this principle, the researcher decided on the basis of prior interviews and preliminary analysis, which issues would be given emphasis in subsequent interviews and how analysis of new data would develop (Boeije, 2002). As soon as two interviews had been recorded and transcribed, analysis began and the content of these and subsequent interviews was compared. In this way, some questions were given particular emphasis in subsequent interviews in order to educe information on issues that were raised by the comparison process (Boeije, 2002). For example, issues about language barriers, cultural diversity, the effects of economic

crisis and the provision of care to undocumented immigrants were given special importance by the participants from the first interviews so the relevant questions were elaborated accordingly in the following interviews. Writing was an integral and constant part of the analysis and not something that took place at the end.

Phase 1 – Familiarisation with the data: In qualitative research, it is vital for the researchers to immerse themselves in the data to the extent that they are familiar with the depth and breadth of the content. This procedure was the bedrock for the rest of the analysis. Since the data of the present study consisted of verbal material, the interviews needed to be transcribed into written form prior to the onset of the thematic analysis. The process of transcription was an excellent way for the researcher to obtain an initial awareness of the data, rather than a mechanical procedure, as some researchers suggest (Bird, 2005 and Lapadat and Lindsay, 1999). For this reason, a verbatim transcription of all the verbal information was produced in a way which was practically suited to the purpose of analysis (Edwards, 1993). Afterwards, the first phase of the process involved repeated reading of the transcripts. This reading was oriented in an active way, as it included searching for meanings and patterns. The entire information was read through at least twice before the onset of coding. The familiarisation with all aspects of the data, whether they were semantic or latent, was important. Moreover, during this phase, notes were taken regarding the appearance of concepts. These notes could be retrieved and reviewed in subsequent phases of the analysis. Gathering and analysing data are often conducted concurrently in qualitative approaches, thus adding to the depth and quality of data analysis (Vaismoradi, 2013). However, it is also common to collect all the data before examining them to determine what they reveal (Chamberlain, et al, 2004). In the present research, only preliminary analysis was conducted prior to the concentration of the entire information. Furthermore, all the transcribed interviews were translated into English so the analysis could be discussed with the research supervisors and the findings could be presented in the last phase. The researcher made the translations himself so he could secure not only an accurate transfer of the literal content, but also elaborate the essence of the meanings contained in the use of metaphors by the informants.

Phase 2 – Generation of the initial codes: The second phase involved the production