The Narratives Which Connect…

Writing memos

Through the process of interviewing participants and watching videos from their clinical practice, I have written memos. The memos are my interpretation and analysis and give direction and question for further data collection. This memo writing process is one of the frameworks for developing concepts and categories.

Writing memos represents the process in between analysing the material and writing the first draft. “Memos catch your thoughts, capture the comparisons and connections you make, and crystallize questions and directions for you to pursue (Charmaz, 2005, p. 72). Writing memos also became a way of storing ideas and possible connections between different parts of the material, for example between transcriptions and videos. Re-reading the memos in the process of writing a first draft was very fruitful and some of the memos represented prewritten reflections, case material and analysis of great value.

Memos are my contemporary interpretation and analysis of the ongoing research process and give direction and questions for further data collection. This process was the basis for developing concepts and categories. Memos are made in between coding and form the framework for the thesis. The thesis is based on the developed categories and “…the categories reflect the interaction between the observer and the observed” (Charmaz, 1995, p. 32).

Open Coding for the Transcribed Interviews

Open coding consisted of line-by-line analysis of transcribed interviews and it constitutes the first stage in the analysis. Initial coding and analysing is a part of the theoretical sampling process. These processes are closely linked to writing memos and in that way developing the project on its way to saturation. In the analyzing process, the seven participants are treated in different ways. The four first (Elisabeth, Erik, Adam and Karen) participants contribute with nine transcribed interviews and four video sessions of a first therapy session. All the transcribed interviews are line-by-line coded and analyzed as a part of the theoretical sampling process. The videos are subject to theme analysis to make it possible for connections between participants’ clinical work and personal and private life to appear. In this way, they are also a part of the theoretical sampling process.

The next three (Evelyn, Anne and Janne) participants participate with three interviews and the transcriptions are line-by-line coded in the same way. After analysis, these interviews are connected to expanding the categories that emerged during analyzing process.

The coding process gave initially many new ideas and was very helpful in developing the research to get variation and depth. The analysing process offered to the project new research categories that were not already included. This made me look for what could widen my project and find new family therapists as participants. In this period, I used my colleagues to get advice for who could be my next participant. I told them what I was looking for and they could give me ideas for whom to ask.

Analysis

Grounded Theory is well designed for developing theory in the topic under the investigation, and it is based on the idea of theoretical sampling. In addition, Grounded Theory appears to be a structured and well-suited approach to analyse huge amounts of information. It also has the advantage of being both descriptive and interpretative (Polkinghorne, 1989), and would therefore work well for my analysis. Analysis should come immediately after data collection. This means that analysis follow each interview or participant (Strauss and Corbin, 1998, p. 207).

According to Pidgeon and Henwood success in generating theory that is well grounded in data depends upon maintaining a balance between the full use of the researcher’s own subjective understanding and the requirement of ´fit´ (Pidgeon and Henwood, 1996, p. 87). This means that the categories have to ´fit´ the data and that trustworthiness and subjectivity could emerge to a level of new meaning that can create intersubjectivity. Trustworthiness is in the eyes of the readers.

The material has been analysed in two steps. In the first step, I have used Theoretical Sampling from Grounded Theory to guide the selection of participants. Then I have used Grounded Theory coding to analyse the interviews and I used Theme Analysis of the videos as the main methods. Grounded Theory has been used to analyse my transcribed interviews. The relation between the interviews and the video has formed the basis for my theoretical sampling. I have been directed by looking for variation so in the analysing process I was always searching for new material and new stories.

In the second step in analysing my material, I have done axial coding. Axial coding was the main process to create categories from the first and second interview.

Patterns between Narratives

Patterns between narratives that showed an important link between a therapist’s private and personal experience and the therapist’s clinical practice came forward already in the first interview with Elisabeth. All but one therapist could spontaneously tell stories from their own personal and private life and link some of them more or less directly to their own clinical practice. These stories are documented in this thesis.

Thematic analysis of videos

The aim of the thematic analysis of the videos was to look for links between what was going on in the therapy session and the knowledge I had about the therapist from the transcribed interviews. The videotapes were looked through several times.

The videos were coded according to themes that emerged in the sessions. The theme that came forward in this coding was compared with the transcribed and coded interviews. When any match between an interview and a video from the same therapist was found, these parts of the videos were transcribed too. Parts of the client’s stories that the therapist did not give any specific attention are examples of what I did not transcribe.

The transcribed parts of the videos were then analysed and compared with the narratives that came forward in the interviews.

Patterns between Narratives and Videos

With the help of videotaped therapy sessions, observation as method was brought in to be a part of my material. After coding and analyzing first the transcript and then the video, the results were compared and contrasted in looking for patterns and narratives that could be connected. When connections between transcript and video were constructed, the construction was brought back to the participant in the next interview. In this interview my constructed relation between a story in the first interview and the video were presented. This new interview also worked as a validation procedure.

Table 6. The Creation of Patterns that Connects.

The Coding Procedure and Process, Bottom up and Top down

Bottom up coding was open coding of the nine first interviews. The categories that emerged through this process were used to “top down” analyse the remaining three last interviews. Several hundred codes emerged during the initial Grounded Theory analyzing processes (Appendix 5). At this stage, the whole material was analysed again, and this time with axial coding with the aim to develop categories.

After some considerations, I chose to do the analytic work manually. I used scissors and an empty floor and started to cut the more than 300 pages of transcriptions and sort them according to codes that were related. After the first rounds, it was possible to incorporate some of the piles and to move the work on to a large table.

When all interviews were coded and analysed as a part of the theoretical sampling process, the first nine were cut in pieces using scissors and spreading them out on the floor. I started to organize the small pieces of papers for each participant in piles after how they related to each other. These piles formed themes and some of the piles could after some time go together under a common theme. In the end, I put the piles for each participant into envelopes and gave each envelope a name reflecting the content inside the envelope. The content of the envelopes for each participant helped me construct a diagram.

Open Codes and Research Categories

These open codes were sorted manually, as told before. At the same time, the videos were analysed with Thematic Analysis. After a long and demanding process the first categories slowly appeared. When they first appeared, the analyzing process came over into a “smoother water”.

In the first rounds of analyzing the interview, the main idea was to use the result of the analysis to find the next participant. However, it became important to start writing memos already at this stage. These memos came to be the first theoretical constructions of my thesis.

As mentioned, I manually cut all the interviews into pieces according to open codes and related codes and put them into piles. The first case was compared and contrasted with the second case. Then these two cases were compared and contrasted with the third. At last, these three cases were compared and contrasted with the fourth. I put all the piles into different envelopes and gave the envelopes names after the topics of what they contained.

Most of these names were long names like: “Therapy, histories and experts that reflect with their blind spots”. Then I started to put these envelopes together according to how they were related to each other. The first categories came out of this sorting process. All together nine interviews were included in this process.

Top Down, Special Topics and the Representatives for these Topics

Already in the first round with coding interviews, the first category emerged. When I was analysing one of my cases I found some possible connections between my first interview and the video that was not clear, but that I wanted to explore further. In the video the therapist met with a couple. The woman in the couple said that she not could live with the man any more but since they had two children they needed help to communicate as parents. When the therapist was exploring what was the problem they told about several serious problems in the family. Two of the problems were that the man was a diabetic and that he was a heavy drinker. In the rest of the session, these two problems were the topics. The therapist asked some very qualified questions about his diabetes and she explored his drinking problem in detail. In the end of the session, the woman said that she had decided beforehand that she would not talk about her husband’s drinking problem. She seemed astonished that that was all they were talking about.

In my analysis from the first interview I came to know that the therapist was a nurse and in one sentence she said: “… and my husband is drinking too much” (2, 23). Her interest for his diabetes could be connected to her background as a nurse but could it be a connection between her small comment on her own husband drinking too much and her focus on this drinking problem of the husband on the video? I decided to ask her both these questions in my next interview.

When I met her again, I presented my findings and when I asked for a possible connection between her focus on diabetes and alcohol abuse she said after some time of interviewing:

“This could have been me and my husband. He has diabetes and he is drinking too much. We have been going for couple’s therapy for one and a half years with this problem” (3, 36).

At this point I realised that sometimes there are “obvious connections” like this. However, these “obvious connections” are not always easy to discover or to act on. However, this made me interested in this aspect, and I immediately knew many stories like this from my own experience and from therapists I have met in the family therapy field. I named these kinds of obvious connections parallel connections.

Parallel Connections

Some connections between the therapist’s own personal life and the client’s experiences are based on special crises or dramatic events in the therapist’s life. This can be connected to for example death or divorce. To understand these kinds of processes we have to consider time. Therefore, it is two kinds of connections that are obvious. First, there are connections that correspond in theme and time, and second, there are connections that are a part of the therapist’s life history and that now are a present problem for the clients. For most therapists it is probably a difference between situations where the therapist and the clients have the same kind of problems at the same time, and situations where the clients’ problems are a part of the therapist’s life history, but not an active ongoing problem at the time being.

When the same kind of problems or difficult life situations occur simultaneously for the therapist and the family it can sometimes be impossible for the therapist to go on working with these families or as a family therapist at all.

When the same kind of problems or difficult life situations has been a part of the therapist life history it can be a very important source to a qualified therapy process. “Parallel connections” has got its own chapter in this thesis (see chapter 7).