Analysis of the data – Content analysis

In this chapter, the analysis process is explained and described how it was implemented in this study. Content analysis is a suitable method in many types of qualitative research; thus, it was also applied in this research. The main purpose of content analysis is to organize the data so that the essential information is not lost. It is a methodological framework, which can be used in various ways, enabling the analyzing of data in versatile manners. The analysis process is based on interpretation and reasoning, where it is possible to proceed from the empirical data to conceptual perception about the phenomenon. The analysis must be based on truthful and credible interpretation as well as justifying the decisions made in the process (Puusa 2011b.) When it comes to data analysis in qualitative research, the data can be rich. Therefore, many interesting viewpoints can be found, even such viewpoints that the researcher does not expect (Puusa 2011b; Tuomi & Sarajärvi 2018). The data of this study also ended up being rich. In the interviews conducted in this study, the order of questions varied as the interviewees started talking about certain topics before the interviewer had proceeded to that question. Therefore, data-analysis may be more complicated in semi-structured interviews, as it may be more

difficult to locate the answers of every respondent regarding the same question and to organize questions and answers that are quite similar (Patton 2000).

According to Tuomi & Sarajärvi (2018), the analysis process can be either data driven, theory guided, or theory driven, depending on how strongly the research and its analysis is based on theory or data. In this study, the analysis is theory guided, also called abductive reasoning, for the research itself is based on theory and previous theory was studied before the interviews were conducted. However, it does not strictly follow a certain theory nor does not try to test a theory. In abductive research, focus is on the meanings and interpretations, the motives, and intentions that people use in their daily lives (Eriksson & Kovalainen 2008; Ong 2012).

The aim of the content analysis is to make the data a meaningful, clear, and compact entity, with which the researcher can make conclusions about the phenomenon under research (Puusa 2011b). When starting the analysis process, the data will be gone through several times. The first task in the analysis process is to organize the data by checking, completing, and organizing the data for saving and analysis (Hirsjärvi 2000; ref. Puusa 2011b.) This was done in this study at the end of the transcription process. The interviews were re-listened and corrected wherever needed. When the researcher has gained a holistic view about the data, it will be divided and examined in more detail. Phrases can be simplified into single expressions. This is also called coding (Puusa 2011b). Coding was done for the data in question by collecting single expressions into an Excel sheet. Subsequently, findings will be categorized and thus, thematized into the most important topics aroused from the data. (Puusa 2011b). In the data of this study, the similar expressions were later unified under one. As Eriksson & Kovalainen (2008) state, the analysis process took place in several phases and the phases mentioned above took place simultaneously.

Themes can be made according to a preplanned structure, or alternatively, new themes can be generated from the data (Puusa 2011b.) In this study, the preplanned themes were followed when analyzing the data, however, the information and even themes were documented when something unpredictable came out. However, the themes must be carefully considered to ensure that they are the most relevant concerning the study. During this study, the themes were determined based on theory when the research question and interview questions were planned.

Also, it is worth noticing that one theme can be generated from several questions in the data and some of them may even relate to several themes. (Puusa 2011b). This notion also became

apparent when interpreting the data under different themes, as multiple meanings under several categories could be found in one sentence.

Ultimately, the chosen themes are taken out from the data with regards to the research questions and compared with each other (Eskola & Suoranta; ref. Puusa 2011b). In practice, similar expressions were categorized under the same theme. Furthermore, distinct phrases were written down in the analysis phase to describe what the interviewee had said about a certain, central phenomenon concerning the study. Especially the sentences that were quoted were translated with the help of Google Translator. This is a crucial phase, as the decisions of the researcher determine the themes (Tuomi & Sarajärvi 2002; ref. Puusa 2011b; Tuomi & Sarajärvi 2018).

Finally, similar types of categories are combined into distinct categories, which should eventually answer the research questions. These categories, also called typologies are used in finding not the most typical phenomena, but rather phenomena that differs from the typical viewpoints known in the field. This was also the case in this study, as in addition to comparing the findings in the existing theory, an attempt was made to find new information. However, the research questions were kept in mind, as the ultimate purpose of the analysis is to find the most essential information considering the goals of the study (Liane 2001; ref. Puusa 2011b). All in all, the analysis of the data must be made carefully, so that new theory or new information to theoretical discussions can be produced. (Puusa 2011b).

After having analyzed the data, the researcher interprets and explains the results generated through research. Also, the results should be synthesized. With the help of synthesis, it is possible to compile the main points of the research results. Conclusions, in turn, are based on the synthesis (Puusa 2011b.) In this study, the conclusions were determined as explained above, first by synthesizing the findings and consequently, turning them into conclusions. However, making conclusions may happen all the time during the analysis process (Puusa 2011b). This also happened during the analysis of the findings in this study. The theoretical conclusions should be made based on the theoretical framework, namely, what kind of new information is acquired from the results. Central research results also form the theoretical contributions, and the researcher should be able to answer the research questions with the help of theoretical contributions (Puusa 2011b.) The conclusions of the study in question will be explored more closely in chapter 5.