Sunday, December 2, 2012

Dissertation on shame. Chapter 28.3 Method



28.3 Method


Interviews in this exploration can in my opinion be seen as social productions. Working together in active interviewing, the researcher and the informants construct stories and give these narratives an interpretation together. Holstein and Gubrium (1995) state that researchers using active interviewing techniques should:

Acknowledge interviewers’ and respondents’ constitutive contributions and conscientiously incorporate them into the production and analysis of interview data    (1995: 4).

Their perspective advocates a form of research that involves collaboration between the researcher and the respondents in order to create a process which involves the co-construction of meaning.

In this exploration it was important that such a sensitive matter as shame in connection with sexual abuse could be talked about in groups where the respondents were familiar with each other and in an environment where they felt secure. It was constructive to have so few in each group so that all respondents could participate freely. A drawback with such small groups was the vulnerability of the participants when one or two were absents from a meeting, or when one member of the group was more verbal than the others and therefore dominated the group. I experienced both of these drawbacks, and this gave me valuable research experience. Using active interviewing seemed to make the participants more relaxed and the group members could speak freely, in spite of the drawbacks mentioned. One of the employees at the Incest Centre in Vestfold, Linda, has spoken with many of the other participants both during and after the study and has the impression that talking together; sharing narratives in the interviews has brought the users and workers at the Incest Centre closer together. The way the interviews were conducted gave a feeling of security. She describes her experience with the interviews this way:

Linda:                         It’s been, in a way, umm, well all the groups say that the way the interviews have been conducted has resulted in us being more welded together than before. We have become better acquainted with each other as individuals. That’s the way I felt. It was really all-right to be in the group I was in. It gave us a whole lot in return. It’s difficult to explain, but it gave us a, umm ((Bites her lip)) umm it felt safe and you gave us umm a whole lot in addition or what can I say. It was a very positive experience. 

Linda’s comment seems to verify to a certain extent that ethical considerations had taken the welfare of the informants seriously, not only did I need information from them, but I wanted to carry out the study in such a way that the participants felt they also could learn something from it. I was hopeful that the interviews were carried out in a helpful and respectful fashion. It is of course possible that Linda could feel my expectations as a researcher and gave this response so as to satisfy my expectations. This is in my opinion a common problem with interviews. The informant will try to be good respondents by trying to give “the right answers” and eagerly pleasing the researcher. On the other hand it is possible in my opinion that participants can come closer to each other when openness and honesty are present in group meetings. This is what Buber calls “I-Thou” meetings and has been used in both social work practice and research (Sim 1994; Itzhaky and Hertzanu 1999; Yassour-Borochowitz 2004).

One of the primary ethical challenges in using active interviewing was making sure that the respondents knew what that they had agreed to participate in. It was necessary to repeat the intention of the study and how the information was going to be used several times, both before the interviews started and during the interviews. It was important to maintain a relaxed atmosphere in the interviews so that respondents could freely ask questions about the study.  Still, I was always in doubt about whether they fully understood the purpose of the study. When is the amount of information given sufficient? How could I be sure that the descriptions I gave of my study were comprehensible for my respondents? Another ethical consideration I faced using active interviewing was evaluating the different tactics I had used as an interviewer and which the participants used as respondents. Looking and listening to the many hours of video recordings was a revealing experience; it showed the many different strategies people use in an open dialog. It was thus not only necessary to be sensitive to the macro-ethics of the knowledge produced in the study, but also to the micro-ethics of the interview situation itself. Although I had these ethical problems in mind throughout the study, I still believe that active interviewing techniques encouraged frank conversations and opened up for arguments about what was being said, thereby contributing to the co-construction of meaning.

Before conducting a qualitative investigation into the concept and phenomenon of shame, I carried out a survey with two case groups. These surveys are carried out because I was curious to find out; in what degree shame-proneness is a phenomenon which at all can be measured; if people who have been sexually abused have a greater degree of shame-proneness than university college students; which possible relation shame-proneness might have to other self-conscious emotions such as guilt and pride; and to learn how it is possible to investigate if TOSCA-3 really measures what it intends to measure (construct validity). The test used was the Test of Self-Conscious Affects (TOSCA-3), which intends to measure shame-proneness, guilt-proneness, detachment, externalization, and two forms of pride. The findings from these two surveys, one comprising 201 university college students and the other 180 adults who had been sexually abused as children, show that subscale means and standard deviations between the two groups were very small, suggesting little difference between the two groups. I had expected the differences between the two groups to be much greater, considering the impact that sexual abuse has on a victim. But the survey suggests that people who have been sexually abused have approximately the same proneness for shame and guilt as those who have not been abused.

The survey also showed that Pearson’s Correlation between shame-proneness and guilt-proneness was high (r=.68) in the case group of sexually abused men and women, and moderate (r=.42) in the case group with university college students. An interesting question which arises here is in my opinion if the high correlation shown in Incest 2005 can be explained because of the experiences of sexual abuse which this group has. This question is examined further in the focus group interviews carried out in the Incest Centre in Vestfold. The high correlation between shame-proneness and guilt-proneness might imply that it is difficult to measure shame and guilt as two independent emotions. I therefore carried out an exploratory factor analysis. It was not possible, in my opinion, to confirm construct validity from this factor analysis. There might therefore be some degree of uncertainty to whether TOSCA-3 really measures what it intends to measure. This finding needs in my opinion further investigation.

I conclude from these two surveys that it seems possible to measure proneness to self-conscious emotions such as shame and guilt, but it is important to keep construct validity in mind when constructing such tests. It also seems that shame-proneness is correlated to other emotions in the test. The greatest correlation seems to be between shame-proneness and guilt-proneness in the survey group of those who have experienced sexual abuse. The correlation her (r=.68) might suggest that victims of sexual abuse have difficulty in treating shame and guilt as two different emotions. This assumption is taken further in this exploration through the interviews of 19 employees and users of the Incest Centre in Vestfold. 
Kaare T. Pettersen
Reference:

No comments:

Post a Comment