The Narratives Which Connect…

Ethical issues

Several ethical issues have been enlightened and reflected upon earlier in this thesis. It is however necessary to say a few more words about the ethical issues. This research project has included the participants’ personal and private histories. To some extent, this will also include the narratives of some third persons close to the participants. The need for anonymity in this context is sometimes a sensitive issue. Norway is a small country and the group of family therapists is limited. To be absolutely sure that no third persons could be hurt by this research project I have decided to leave one of the stories from one of the participants out of my material.

Initially the plan was to invite the participants to participate in a focus group at the end of the analytic process. The plan was to first conduct a member validation by letting the participants read some of the analysed findings and then having them discuss the relevance of the categories for clinical practice and for family therapy education. However, to include the participants in a focus group would be another way of revealing their identity, and the idea was excluded for ethical reasons. Afterwards I see that it could have been possible to convene another group of family therapists to deliberate the findings.

Validity and Trustworthiness

“The issue of what is valid, true knowledge in the social sciences involves the philosophical question of what truth is” (Kvale, 1989, p. 75)
Concepts like validity, replicability and reliability have numerous definitions. They are concepts drawn from quantitative and positivistic oriented research methods. In the positivistic traditions different areas of validity (construct, internal and external) are quite well defined (Robson, 2002). When the same concepts are transferred to and used in a qualitative research project it is of the greatest importance to the particular understanding of validity and reliability as clear as possible. It is “…validity and self-reflexivity (that) support qualitative research” (Dallos and Vetere, 2005, p. 201).

According to Jonathan Smith (1999) qualitative research should not be evaluated using the same criteria as quantitative research. On the other hand, there is no common agreement on criteria for validating qualitative research (Richardson, 1996). Thus we need to ground and validate our findings so we can use them to hypothesize and generate theory for further research.

The question of validity may be described as both internal and external. This study is internally valid when it describes the meaningful state of affairs within this setting. When that is done the research project has reached trustworthiness. External validity cannot be established unless more research is performed that generalises the findings to population. As mentioned earlier, the idea of researching the narratives that connect a family therapist’s personal and private life with the therapist’s clinical practice was supported by most of the people who heard about the project plan. Therefore, it was important to let the idea of validity accompany all steps in the research project. This leaves the concerns about generalisability and generativity. What is valid in this research project is not necessarily valid in other contexts.

Respondent validation

Respondent validation is suitable for qualitative research in general and for a grounded theory project in particular. “Respondent validation is often used when the researcher is attempting to represent the views, opinions and experiences of the research participants” (Dallos and Vetere, 2005, p. 205).

Respondent validation has been a key validation strategy from the very beginning of this research project. The basic idea in this research project was that the participants’ validations of the initial findings should form the trustworthiness of the project. After analysing the first interview and the first video, the relation between the two of them was analysed. Meaningful links between the interview of a participant and her or his video of a first family therapy session were then brought back to them in a second interview. This is an ethno-methodological approach to interviewing.

The possible, meaningful links between the first interview and the video was an outcome of my preliminary analysis of the relations between the two. However, from the very beginning I was aware of the limitations of links as meaningful for the participants. However, it was my idea that their evaluation of the meaningful link in the second interview should form the main outcome of these analyses. As shown in the findings, we were able to establish meaningful links between all four videos and the first interviews for the first four participants.

Further on, all seven participants were invited to give additional comments to me through e-mail responses. All of them came back to me with at least one such response during the research period.

As mentioned in the paragraph about “Ethical issues” (p. 67) I decided not to perform the focus group as an attempt to protect the participants’ anonymity. Instead, I decided to send each one of them the findings chapter before the thesis was completed. At the end, when the findings chapters were finished I therefore offered all participants a final opportunity to comment on the analysis and categories that had emerged through the research process. This was done to ensure their interviews and videos were given a serious and professional treatment and that their anonymity was kept within a responsible frame.

All together these elements form an important part of the validation strategy are attempts to reach trustworthiness.

Triangulation

“Triangulation as a strategy is based on the idea that different perspectives on the same phenomenon can enhance our understanding whilst at the same time can provide a basis for cross-checking and cross-referencing our findings” (Dallos and Vetere, 2005, p. 205). In the process of building the validity strategy into the research project, triangulation was used to increase this aspect. The triangulation model in this research consists of the literature review, interviews analysis and videos analysis. Being able to create meaningful and powerful links between these three key elements in this research project should be part of establishing both validity and trustworthiness in the interpretation of the research findings.

Audit trail.

The audit trail seeks to participate in the creation of validity by showing how the research project as a whole has been built. This is done by casting a backwards glance on the project to show how it fits together. This backward glance will add transparency to the project.

“If we use different methods of data collection, we will expect them to enrich our understanding of the issue at hand” (Dallos and Vetere, 2005, p. 207). To show how I have reached this enriched understanding I will recapitulate the research project backwards as steps. When a (research) process is converted into steps some qualities that are characteristic for a process might be lost. For example, the categories did not only emerge at one certain step in the project. One category emerged very early (parallel connections) and some categories were given finishing touches up until the thesis was finished.

My supervisor(s) have participated closely every single step on this journey. Some main steps on this road to completing the research project were:

1. 
   In the findings, to be able to present the paradigm cases and the seven GT categories and show that all are grounded in direct quotations from the interviews and from the videos, I had to:
   
2. 
   Construct the categories emerging from my material and form the paradigm cases. To be able to do this, I had to (see Appendix 7):
   
3. 
   Analyse the possible links between the interviews and the videos to form categories and extract the paradigm cases. To be able to do this, I had to:
   
4. 
   Analyse all research codes by sorting them into piles of codes that fit together and going through a process of constant comparison. To be able to do this, I had to (see Appendix 5 and 6):
   
5. 
   Conduct the theoretical sampling procedure where seeking variation was the leading goal when I picked new participants. To be able to do this, I had to:
   
6. 
   Complete the second interview with the first participant and present my findings from analysis of the first interview connected to the video of the first session. To be able to do this, I had to:
   
7. 
   Analyse the first interview and connect it to the video of a first session. To be able to do this, I had to:
   
8. 
   Invite a first participant to join the research project for a first interview and to make a video. To be able to do this, I had to:
   
9. 
   Prepare the information sheet and agreement to participate in the research project. To be able to do this, I had to (see Appendix 1 and 2):
   
10. 
    Get ethical approval for my project plan from the examination board for the Doctorate in Systemic Psychotherapy at The University of East London (Tavistock Clinic site). To be able to do this, I had to (see Appendix 3):
    
11. 
    Prepare my project plan with the help of my supervisor and put in for recognition by the examination board. To be able to do this, I had to:
    
12. 
    Do the pilot preparations by interviewing experienced family therapy teachers in Spain, Ireland, Wales and Norway about the relations between a systemic family therapist’s personal and private life and her or his clinical practice and reading what was written about it. To do this, I had to:
    
13. 
    Apply for and be recognised as a doctoral student at the Tavistock Clinic. To be able to do this, I had to:
    
14. 
    Ask Diakonhjemmet University College for time and money to participate in the Doctoral program at the Tavistock Clinic. To be able to do this, I had to:
    
15. 
    For many years, be puzzled with the questions about the relations between a systemic family therapist’s personal and private life and her or his clinical practice and what the understanding of this area may mean for family therapy education and practice.
    

To give further documentation of this audit trail I will in the appendix put in translated and coded transcripts to show how categories emerged, the information sheet and the consent form for participants (see Appendices 1, 2, 3, 6 and 7).

Reflexivity and self-reflexivity as context

The reflexive process in this research project was initiated from the very beginning. Reflexivity and self-reflexivity have been a part of the research diary and the memos. The participants sent reflections to me after the first and the second interview and at the end of the whole process.

Supervision has been an important element in making clear how I as a researcher have influenced the development of the interviews and the GT analysis. I also had an independent audit. The supervisor audited a sample of the analysis.

Research participants had the opportunity to comment on and develop ideas from the first interview in the second interview. In addition, I have invited the participants to give e-mail reflections after each interview and at two other points. A summary of these e-mail reflections is presented earlier in this chapter (see pp 161-164).

I attempt to show how my personal background and engagement in the field of family therapy education in Norway may have influenced this research project and the research process. My aim is to make a reflective overview of all stages in the research process. This includes the planning of the research process, the interview and analytic process and the phase of writing up the thesis.

As mentioned, I also invited all participants to send me reflective e-mails after all interviews. Most of them made use of this possibility. After about one or one and a half years I invited them again to reflect upon the project and their involvement in it. All of them answered this last invitation and presented their reflections.

I also wrote a research diary from the very beginning of the research project. Since this research project has stretched out over several years this diary has been of great importance to help me recall the sequences as they occurred during the project period.

To further address self-reflectivity, a colleague who holds a research doctorate has interviewed me about my motives, interests and own experience with the topics covered by this research. This interview forms an important background for this chapter. However, I will not use direct quotations from this interview because many of the ideas and reflections that come forward in the interview are ideas and reflections that I have had and also used earlier in this work. For example, I had to be confident that the project had a design that was trustworthy and that the documentation of the research accounts for what has been done and explained throughout the complete research project.

The use of triangulation, with the literature review as one leg, seeks to increase validity in the project by linking the categories and sub-categories to psychotherapy research, supervision theory and other experiential and relevant theoretical inquiries about topics related to my research question.

The videos represent an attempt to increase validity in the project. The videos represent an element of observation that includes behaviour and analogue communication as a part of the material. The videos are from “an ecologically valid context” -namely the therapist’s own therapy room (Dallos and Vetere, 2005, p. 204). They represent a second leg of triangulation.

In addition I have used respondent validation (Dallos and Vetere, 2005). The most important element in this process is bringing my analysis of the possible connections between the first interview and their video of a therapy session back to the “full case” participants. In this second interview they were invited to comment on and validate my constructions of meaningful connections and patterns. Also through inviting all participants to send e-mail reflections and through inviting them to comment on the first draft of the thesis, respondent validation took place.

All together, these elements form the basis for the trustworthiness of this research project. Without trustworthiness there can be no validity.

Generalisability

Generalisability is connected to validity in a way that broadens the understanding of validity. My research findings may be valid in one setting and not in another setting. Generalisability is about how to understand in which contexts these research findings add meaning. Systemic family therapists have been participants in this research project and all were informed beforehand about the topics and the research aims. This may have influenced some of the participants to look for some certain kind of answers that fit these topics. Only one of the participants could not tell any histories that linked his own personal and private life to his clinical practice. The lack of participants with this experience and understanding of their own practice may reduce the research project’s generalisability both inside the field of systemic family therapy and the field of psychotherapy in general.

On the other hand, the many findings that this research project presents from all participants (Erik (2) included) may also trigger curiosity for further research, both in the field of systemic family therapy and in the field of psychotherapy in general. This research project is performed in a country where PPD work in general is weak or unknown and where self therapy has been removed from family therapy education many years ago. In this context, these research findings could promote further research.

Generativity

Generativity is defined as “having the ability to originate, produce, or procreate” (The free dictionary: http://www.thefreedictionary.com/Generativity). In this research project generativity occurs in three areas. First, a middle range theory has been developed by looking for relationships between categories helped by the method of constant comparision. Second, the need for more research seems to be obvious and may be seen as a part of generativity. Third, the findings should have some consequences for our family therapy education programmes in Norway. A PPD module or the like to cover these questions should probably be a compulsory part of such a program.