3 Research Design and Methodology

3.1  Approach to the study

3.1.1  Research approach

↓36

This study was designed to provide empirical data on the concepts of disability, cultural values concerning disability and its Social Representations in KwaZulu-Natal. The study also attempts to analyse the various stigmata associated with various disabilities and how this influences the way people interact with each other and increases exposure to the risk of HIV infection. To approach this matter I was particularly interested in people’s individual experience of disability. Experience is an overused term in social science and therefore I want to clarify my view at this point. In accordance to Kleinman and Seeman (2000) I understand experience as a representation of “the intersubjective felt flow of events, bodily processes, and life trajectory which always takes place within a social setting.” On the one hand experience is an outcome of cultural categories and social structures that interact with psycho-physiological processes in such a way that a personal social reality is constructed. Social reality on the other hand is a product of meaningful social interaction as perceived from perspectives of those involved, and not from the perspective of the observer (Kleinman and Kleinman, 1991). I aimed to capture these individual experiences of disability and the social reality of the phenomenon. Exploring these two aspects I had to inquire deeply into people’s notions and interpretation of disability. Due to the fact that I had not entered the field with any sort of preformed hypothesis, more inductive and explorative ways of research seemed necessary while approaching my topic. Such an inductive approach allowed me to generate concepts in the course of my research (David and Sutton, 2004). Quantitative ways of collecting data, which are usually more deductive, therefore seemed unsuitable while approaching the topic. On the contrary, qualitative methods such as observation, interviewing, drawings and ranking exercises, were able to access the individual experiences and Subjective Theories and with this the Social Representations of disability and seemed therefore to be the better choice for my study.

↓37

As a researcher I tried to be sensitive to the priorities and notions held by those whom I approached. A useful qualitative approach for this purpose was ethnography. Ethnography literally means ´writing about people´. It sees people as experts in the field of what the researcher wants to find out about, rather than as objects (Burns, 2000). Ethnography initially involved descriptive data collection as the basis of interpretation and has often been used while studying sick-role behaviour. Using this approach I hoped to create a dynamic picture of the subjective life experiences of people with disability and in this way capture the social reality of these people. I am however aware of the fact that meanings and interpretations are not fixed entities and that they are generated through social interaction and can change over time. I also aimed to capture these dynamics within my study.

Using an ethnographic approach meant that I had to study people’s behaviour towards disability in an everyday context. I had to base my research on analysing people’s conversations and reflections about their life. To accommodate people I had to use a less structured research design and keep myself open and flexible with my choice of participants and where and how I would conduct interviews. It also meant approaching only a small number of people and collecting more descriptive data (Hammersley 1998).

As is often done in qualitative research, I would select particular cases or certain bounded realities and study these in depth. In my study these cases or realities were related to disability. Using an interpretive paradigm, it enabled me to investigate the subjective experiences of the participants. Through this I hoped to gather a greater understanding of the broader phenomenon than would have been otherwise possible. The focus of my research was to understand Subjective Theories of and the reaction to disability and how this links to HIV/AIDS. In accordance with the tradition of the Symbolical Interaction theory (Blumer, 1938) I therefore based my research on the symbolic character of social actions. Research with such an approach concentrates on the way in which the participants perceive and reflect others (Flick et al., 1995). It also takes into consideration the experiences people have had in their lives and how they relate meaning to this, therefore the reconstruction of Subjective Theories became the instrument for my research of the world of people with disabilities.

↓38

At the time of my study it was very popular in AIDS related research to use a participatory action approach, where the participants can immediately profit from the research and are included fully in the research process as equal partners. Often these types of studies are done by a research team and have some financial backup (Preston-Whyte, 1996). I usually approached my participants as an individual and therefore did not feel confident to use this method on my own. I did however include participatory elements into my study. I included DPSA and DICAG14 into the process of choosing valuable participants. They also influenced the design of my research questions and were included in feedback discussions through focus groups at the end of my research. It is also possible that the various interviews and visits might have had a positive and empowering effect on my participants, as I was often able to give advice and comfort in the aftermath of the interview. I can however only speculate about this.

3.1.2 Research context

KwaZulu-Natal (KZN) is one of 9 provinces in South Africa. It is situated in the south east of the country. I chose this province as the research site for my study as it is one of the epicentres of the HIV-pandemic. I have also spent many years volunteering and visiting KwaZulu-Natal and know the area well. Within KZN I focused on the Durban metropolitan area which is commonly known as eThekweni. Data from other municipalities has also been collected so as to get a broader picture.

The broader context of KwaZulu-Natal will be described in chapter 4. Here I only want to raise the point, that the process of collecting data has been influenced by the KwaZulu-Natal economic situation as well as its cultural and historical heritage. The issues that are of particular importance in this context are:

↓39

These four points are not only typical for KZN but also for the broader South African context. Important to know is that these factors have influenced the way I was able to collect my data. For instance it was not possible for me, particularly not, as a European, to walk into any township and see ´what it is like´ as there would have been serious safety concerns. Such areas I had to approach through my gatekeepers and their local affiliates.

The above mentioned points have also influenced the sampling. It was, for instance, relatively easy to identify veterans from the struggle, who became disabled through their involvement in the struggle against apartheid. These people are very often well integrated within the local and national disability organisations and therefore easily approachable. As I was not interested in their story alone, I had to make a particular effort to also meet people of ´ordinary descent´. In the course of my research this became easier as I could approach people through local branches of disability organisations.

↓40

In general, people were very open and it was not difficult to find participants for my research. One reason for this might have been the fact that people with disabilities are in general of little interest to the public and therefore have not yet been over researched. They seemed to have experienced the interviews rather as welcome highlights and they added interest to their lives. Generally speaking there was only one group of people who were difficult to approach: People who had a disability and were HIV positive were difficult to identify and seldom agreed to be interviewed. Some participants argued that the double stigmatisation made it impossible for disabled people with HIV to come out. The same problem caused difficulties for my initial interview partners (gatekeepers) to provide me with contact details. As people got to know me, relatives, caregivers and sometimes people with disability themselves did however volunteer to discuss these issues. This took some convincing and patience on my part, but did provide me with some original data of people with disability and HIV. I sometimes had difficulties visiting people in their homes or in their communities as the current levels of violence made me vulnerable to crime. I therefore always had to link myself to other projects so as to safely access areas I was not familiar with. My sample is therefore influenced by these doctrines. By approaching people with disability through their local disability organisation or locally based research projects, I found that quite a number of my participants spoke English well enough to be interviewed in this language. I did however use certain Zulu words, as the meanings would otherwise have got distorted. A few interviews, particularly those ones which were based in rural areas, were conducted in Zulu. For these interviews I used a translator to facilitate the interviews. The presence of a third person might have influenced the outcome of the interviews. The translator that I used was well qualified and had experience in facilitating research on disability as well as on HIV/AIDS. The interviews were therefore conducted to the highest standard possible.

Usually I had very few problems in approaching people for interviews. At the beginning of my research I expected to encounter sceptics as I was just another white European trying to write about African people, their custom, culture and problems. I however encountered very little resistance and was usually met with interest. In some cases the fact that I was a European seemed to be an advantage rather than a disadvantage as people did not feel threatened or under any cultural restraints when talking to me. One participant who disclosed his HIV status to me (I was the second person he had told about it) explained this in the following way: As I was from a foreign country and not associated to his culture or village I would disappear in much the same way that I had appeared that afternoon on his doorstep. He was convinced that I would not talk about his status nor would I have any chance to do so as I was not “black” nor did I belong to his community.

3.2 Study sample

3.2.1 Structure of the research project

In ethnography the researcher needs to recognise the fundamental need to go where the participants spend their time. The quality of my study therefore depended to a large extent on the group’s acceptance of me as the researcher. Gaining entry into a group is usually best accomplished through a mutual contact that might either become or can recommend a person that can operate as a ´gatekeeper´ to the study field (David and Sutton, 2004).

↓41

During my research in KZN I received support from DPSA15 and DICAG16, who I tried to include into my research as much as possible and whose regional managers operated as my gatekeepers. The research unit HEARD17 at the University of KZN also provided me with support. All three stakeholders invited me to team meetings and workshops; they helped me to find possible interview partners and provided me with professional feedback at various stages of my research. Supported by these three organisations I went through the following steps in the research process:

  1. Contact with DPSA, DICAG and HEARD.
  2. Development of the questionnaire and question guide.
  3. Pilot study.
  4. Improving the research instruments.
  5. Data collection with in-depth interviews.
  6. Transcription.
  7. Coding, categorising and summarising of the individual interviews.
  8. First discussion of interim results with DPSA.
  9. Further Interviews.
  10. Further analysing of interviews and theory development.
  11. Focused group discussion with DPSA
  12. Final conclusions.

Another major task was the actual collection of ethnographic data. Following Burns (2000) ethnographic research cycle (see feature 3.1) I started by making broad descriptive interviews (pilot study). This was to help me to gain an overview of the social situation and what is generally of importance for people with disabilities. Then after recording and analysing the initial data I narrowed down my questions and made more focused inquiries.

↓42

Feature 3 1 Ethnographic cycle (Burns, 2000)

The first step of the analytical work was done after having completed the pilot study. I conducted this study to test my research instrument but also to collect the first set of data and information from my ´gatekeepers´ whom I could use as my prospective interview candidates. The ongoing analysis of the interviews18 and field observations included a process of discovering further questions. Even so I came into the field with specific questions; through analysing my data I was able to discover new questions that would help to describe the phenomenon of disability in KZN. In the final interpretation and writing process I tried to draw conclusions as objectively as possible. As Burns describes qualitative researchers are generally concerned with the effects that their “own subjectivity may have had on the data they produce” (Burns, 2000). He therefore suggests to use triangulation and critique by colleagues as an additional check on bias (Burns, 2000). Both methods I have used to a limited extent. I am however aware of the fact that my own subjectivity will have influenced the collection, analysis and interpretation of my data. The reader should therefore see the entire conclusion in this thesis as my personal interpretation.

3.2.2 Profile of the participants

Data was mainly collected from three different sources: There were documents and policies, key interviews and focus groups. The selection of interview partners was based on two main criteria: The interview candidates had to either be disabled themselves, or work/live with disabled people. The participants that I chose for the in-depth interviews then became my key informants. Key informants are defined as those people “who, by virtue of the nature of their position and knowledge, can capture and reveal relevant cultural phenomena, especially on disability” (Room et al., 2001).

↓43

In my study these key informants were predominately people with disability themselves, their parents or their caregivers and professional staff in schools or hospitals. Even though the setting had already been quite specific it was still necessary to select participants. These key informants were chosen through theoretical sampling and snowball sampling. Through snowball sampling, a person, who is identified as a valid member of a specified group is asked to provide names of other people who fit the requirements (David and Sutton, 2004). I used this approach as I did not always have sufficient potential interview partners. The selection of the recommended interview partners was however based on the method of theoretical and convenient sampling. Using this method I tried to find interview partners who could answer emerging or unanswered questions.

Even though I concentrated my study predominantly on the Durban or eThekweni area I also visited other KZN municipalities. In the map below one can see the areas I approached (feature 3.2).

Feature 3 2 Study sites in KwaZulu-Natal

↓44

Therefore people with different disabilities and from different municipalities were approached. As a result I have spoken to quite a variety of people (see table 3.1).

Table 3 1 Profile of participants

Interview number

Details

Area

1

Woman with hemiplegia and in wheelchair, provincial manager of DPSA

eThekweni

(urban)

2

Mother of a child with severe learning disabilities, CBR19 consultant

eThekweni

(urban)

3

Mother of child with severe learning and physical disabilities, manager of DICAG

eThekweni

(urban)

4

Man with spinal injury, in wheelchair, paralysed through gunshot

eThekweni

(urban)

5

Mother of child with severe learning difficulties, manager of day-care centre in Claremont

eThekweni

(urban)

6

Visually impaired man, teacher at High School

eThekweni

(urban/rural)

7

Business man, visually impaired

eThekweni

(urban/rural)

8

Deaf woman

Ugu DC

(urban)

9

Man with spinal injuries, in wheelchair, CBR consultant

Sisonke DC

(rural)

10

Woman, teacher at Claremont day care centre

eThekweni

(urban)

11

Woman, teacher for children with learning disabilities at High School

eThekweni

(urban)

12

Mother of child with severe learning disabilities, Claremont

eThekweni

(urban)

13

Man, occupational therapist of Murchison Hospital

Ugu DC

(rural)

14

HIV positive woman

Ugu DC

(urban/rural)

15

Nurse in Claremont clinic, sexual educator

eThekweni

(urban)

16

Man with albino condition

eThekweni

(urban)

17

Man with cerebral palsy

Umkhanyakude

(rural)

18

Man with loss of hearing and AIDS

Umkhanyakude

(rural)

19

AIDS patient care giver (for no 18), woman

Umkhanyakude

(rural)

20

Man, traditional healer

eThekweni

(urban/rural)

21

Girl with albino condition, who visits a school for children with disabilities

eThekweni

(urban)

22

Priest and caregiver

eThekweni

(urban)

23

Woman, traditional healer

eThekweni/Empangeni

(urban/rural)

24

Woman with slight cerebral palsy, who was victim of sexual abuse from a young age

eThekweni

(urban)

25

Man with leg amputation who was also HIV positive

eThekweni/Bothas Hill

(urban/rural)

↓45

The main part of my research was conducted while participating in people’s lives. I visited people in their homes, during workshops and in projects they were involved in. This gave me a good inside into their living, working and schooling reality and also enabled me to create a personal and comfortable atmosphere for my interview partners. I also collected data through participating in these workshops or in the daily running of a clinic or day care centre. I often had to take down the information from these institutions as memory protocols, as much insight was given to me through informal meetings, break discussions or presentations that could not be taped as these had not been planned.

3.3 Methodology

3.3.1 Method for data collection

General approach

Predominantly using interviews for data collection I had the choice to use unstructured, semi-structured or structured interviews. I decided against structured interviewing, as this type of interviewing allows very little or no flexibility. Approaching so many different people I expected to need a highly flexible instrument. Structured interviews would also have provided me with little scope to find out the notions, feelings or perceptions of the respondents that might not fit into the pre-ordained response categories. This was however necessary when I entered a field with little or no previous research available. In addition to this I was also worried that the detachment and impersonal approach of structured interviews would prevent trust and rapport building between me and the participants. As part of my interviews inquired into highly personal and confidential information (e.g. rape and HIV status) I needed an instrument that would allow me to build a comfortable relationship with the respondent.

↓46

Semi-structured interviews allowed me more flexibility while conducting the interviews. Even though I developed an interview guide I only used it as an instrument to remind me of the relevant information, which I wanted to find out about. The questions of my guide were not necessarily used with its fixed wording or order but rather used to give the interview the required direction, so that the content focuses on the crucial issues of my study. This approach gave me the following methodological and ethical advantages:

A typical semi-structured interview is the In-depth-interview. I used this type of interview to obtain detailed information about life experiences and individual interpretations of disability. I designed these interviews methodically like problem focused interviews described by Witzel (1982) and Flick (1995). With these interview techniques, questions are raised and formulated so as to focus on a specific problem. As a technique to explore specific issues I used “Funnelling”. With this approach I had gradually to guide the direction of the interview through beginning the interview with a set of broad questions followed by progressively asking more specific questions.

↓47

During my research I used the elements as described by Witzel (1982), such as qualitative interview group discussions and analysed the interviews case by case. The fourth element that Witzel suggested, the biographical approach, I only used to a small extent and focused more on certain key points in the life of my participants. I also used the instruments suggested by Witzel (1982), a short questionnaire, a question guide, transcription and a postscript. In this approach the questionnaire was used to collect statistical relevant data like gender, level of education, occupation, type of disability and so on. Witzel used the questionnaire to prompt the participant in case the interview lost its focus or the participant stopped talking. I very seldom used it in this way. The questionnaire was predominantly used to gather crucial “hard data”.

As Witzel suggested, I used one central entry question for the interviews. Crucial communicative strategies that I used while conducting the interviews were prompting, mirroring, encouragement and Ad-hoc-questions (Witzel, 1982). I prompted with questions like: “What happened there in particular?”, “Can you explain that in more detail?” or “Could you give me an example?” As mentioned before, participants were identified through theoretical and convenient sampling (Burns, 2000). Once I decided what kind of person I would like to interview next, I tried to find somebody through my already established contacts (gatekeepers) and took the person most accessible to me.

The Pilot Study

↓48

For the pilot study I developed a questionnaire and an interview guide. The questionnaire helped me to gather statistically relevant data. For the question guide I identified three main areas that would help me to inquire into people’s understanding of disability and assess HIV-risk factors for people with disabilities. These areas were later used to analyse the data. Questions were formulated for each of the following areas.

Although some of the questions in the pilot study were formulated too academically, four of the participants were able to answer them as they had better educational backgrounds. The question guide did not always give the participants the opportunity to explain things in the order that they preferred. The instrument therefore proved that it was not flexible enough. In addition to this, one respondent even needed help with some questions, as they were too complicated or not specific enough. I therefore realised that I had to adapt my instrument so as to better facilitate further research.

↓49

In the initial pilot study I therefore sometimes diverted from the interview questions for a while and allowed the participant to speak about his/her feelings. The fact that I was investigating their personal situation and wanted to know more about their own subjective interpretation of their lives gave me the latitude to make this diversion. The interviews therefore took a long time and I got much additional information that I did not necessarily need for my study.

The final research instrument

As a consequence of the pilot study I was able to improve on some of the questions in the final question guide. I added an additional instrument that could help me to focus the interviews on what I regarded as the more relevant facts. For this purpose I developed a ranking exercise that I would add to my “entry question”. With this I hoped to give my interview partners an open approach towards the interview and at the same time focus on my research question as well as identify the areas that most mattered to the respondents.

↓50

The ranking exercise can be seen as a stimulating item that is given to the participant at the beginning of the interview and is similar to the technique used in focused interviews (Merton & Kendall 1946/79 and Flick 1995, p. 94). Adding this instrument to my research created a “between methods” triangulation (Denzin 1989, p. 237f.). Although the ranking exercise (stimulus) was used more as a didactical instrument to help focus the participant’s attention on the relevant information, it later proved to be useful in reconfirming some of the research results.

Based on this ranking exercise I would start the interview in one of the three main areas of my question guide (mentioned above). Often it was not necessary to prompt my interview partners after the ranking, because they automatically expressed their thoughts. At the same time I was able quickly to gather the relevant data, as participants kept focused on my topic through the ranking items. A similar method was used by Muthukrishna (2006) while conducting research on children with disabilities, who were affected by HIV/AIDS. The items that I used in this ranking exercise were identified through my pilot study. Through a process of reduction I went on to choose 10 main clusters that would equally represent the three main areas of my research.

The 10 main clusters or key themes were developed out of a list of 30 themes that had emerged in the pilot study. In reading and re-reading data, these common clusters of issues emerged. Some of the themes overlapped and could be united under one cluster. At the end of my analysis I came up with 10 clusters that had a culturally relevant meaning and could be used for the ranking exercise in the main study.

↓51

Clusters identified through reduction:

Once these key themes or clusters had been identified I took each theme and identified or reformulated specific questions that allowed me to probe that theme in more detail. These questions were then used as a basis for formulating my question guide. In the course of the interview I preferred to keep this question guide as an aide-mémoire (David and Sutton, 2004). In this way the question guide would not dictate the order of questions but it would rather be an instrument to which I could return during the course of the interview. In this way I had a basic orientation of what I wanted to find out about and at the same time I did not restrict the flow of words from the respondents. Questions had only to be used when a participant did not respond to all key themes as they were presented in the ranking exercise.

↓52

For all of the interviews following the pilot study I used the questionnaire, the ranking exercise, the question guide and the postscript as my research instruments. An interview session was usually conducted in the following way: While I was arranging the technical equipment for the interview the participant was filling in the questionnaire. Most participants were literate. I helped filling in the questionnaire in only a few cases. Afterwards I once again explained to the participant what the purpose of the interview was and that all information was going to be treated as confidential. At this point some participants chose to ask me more details about my research. After the formal agreement to start the interview I presented all 10 clusters/key themes on cards to the participants and asked them to arrange them in descending order from the most to the least problematic. Often participants immediately started to explain why they arranged them in the way they did. This made the interview situation very comfortable. In this way each interview started in a different section of my question guide and it was sometimes not necessary to use the questions from the guide at all.

On the other hand I was able to use these questions to access more specific information where necessary. The flexibility of the semi-structured interview provided, in this way, more opportunity for the respondent to discuss what he/she meant (Burns, 2000), and this may well have strengthened the validity of the data. Even though the ranking exercise guided people’s thoughts the relatively open question at the beginning of my interviews allowed the participants to express their thoughts in a, for them, comfortable manner. In this way they have expressed their experience of disability in an unrestricted way, which then allowed me to explore their Subjective Theories. Consistently referring to my interview guide enabled me to keep focused on my critical questions. The ranking exercise also made it easy for the participant to come back to these issues.

All but two of the interviews were recorded and later transcribed. The other two had to be written down during the interview, as one participant was not comfortable with being taped, while the other participant being a traditional healer had no electricity available in his shop. Rejections of being taped are explained by the Zulu shaman Mutwa with the notion that the recording device could capture parts of a person’s soul (Mutwa, 2003). As the participant was still very entrenched in tradition it is conceivable that this was his reasoning behind not wanting to be taped. He did not however explain himself and I did not press the point. Another reason for his reluctance could be the fear of getting into trouble for giving advice about HIV. This is especially likely as the participant was a physiotherapist and maybe not sure how much information he was allowed to give me.

↓53

In addition to this, one focus group interview was used to reconfirm and discuss research results with stakeholders. In this interview I presented, to a small group of people with disabilities,20 my final results. I also presented various charts and tried to reconfirm if the participants could identify with my interpretation or if some basic misunderstandings had to be rectified. The purpose of this group was also to use the interaction between the participants to generate discussion about the topic. I hoped that the discussions would be more detailed and wider ranging than the one-to-one interviews.

From an etic perspective this group interview also gave the participants the opportunity to criticise my work and thus have more control over the research outcome. The group I used was small and consisted of 5 people. As a stimulus for discussion I used diagrams that I had developed on disability, quotations and results from the ranking exercise. In a twenty minute presentation I outlined the basic concepts of each section of my research and asked the participants for improvements or rectifications. I had a small aide-mémoire prepared to guide me through the process and to keep the participants focused on my topic.

3.3.2 Data analysis

During my analysis I tried to convert the concrete realities that I had found in the field into a conceptual understanding of the phenomenon of disability and bring this into relation to HIV/AIDS. During the construction of my theory I used analysis based on grounded theory as my main tool. This type of analysis is used to identify themes, concepts and meaning. It is a form of classifying content (Burns, 2000). The greatest strength of grounded theory is that it can be conducted after the interview, so that the initial setting is not disturbed. The coding is also determined after the event. In addition to that it was possible with this approach to choose initial categories in the coding process that were generated by the respondents and their perception of their environment. For the interpretation of especially rich or contradicting data I used discourse analysis. Discourse analysis is used for a wide range of topics, from semiotics to narratives (David and Sutton, 2004). All of them have in common the resistance to a mechanistic reduction and application of research and are therefore descriptive case studies. I have used these case studies while trying to understand reaction chains, particularly how the interpretation of disability led to HIV exposure. Even though for the main part of my research I used analysis based on grounded theory, which means that I was counting and interpreting certain patterns that emerged in the data, I used discourse analysis to explore certain cases in more depth. Questions that answered the how and why certain individuals were exposed to certain situations was of interest at this stage (David and Sutton, 2004).

↓54

Starting with a set of data (transcribed interviews) I read through the material and tried to identify units of meaning that would give me information about one of my three study areas (1. Concepts of disability, 2. Living conditions of people with disabilities, 3. Sexual culture and HIV/AIDS). Using open coding (Flick 2002, p.259;;Strauss 1994, p. 57ff) I identified common themes and coded them as I worked through the data. These codes were initially text based, so meaning was not distorted at this point. At this stage codes could therefore refer to a phrase, sentence, or the idea of a paragraph, as used and described by the interview partners.

The main part of my coding was supported by the computer software programme known as MaxQDA. In the next step I tried to find codes with meanings that would describe the same phenomenon and therefore fit into one category (axial coding, Flick 1995, p. 201; Strauss 1994, p. 63). The ensuing categories were once again named, using the data as a guide to decide what a category should be called. My knowledge about theories of disability and disease seeped into the process at this point. Rereading the data I tried to find common themes for all three study areas (Strauss, 1994). At this point I began to particularly examine the extent to which participants share a common theme or issue and where they were contradicting each other. At the end of my analysis I had reduced my material to five main themes each with its own categories.

Theme 1: Living conditions of disabled people in KZN

↓55

Theme 2: Concepts of disability

↓56

Theme 3: Sexual culture and HIV in KZN

Theme 4: Disability as an intervening condition

↓57

Theme 5: Coping strategies

↓58

Thereafter I studied these issues and themes in order to interpret information and make assertions about my findings. The 5 main themes that I identified influenced the structure of this thesis tremendously and drew the outline for the following chapters. During the process of interpretation I adopted the “think display” approach into my analysis. Davis and Sutton suggest (2004) that this approach should be used beyond the representation of final research outcomes and should also be used in the act of analysis. They go on to explain that by “means of visual reduction the complexity of qualitative data can render comprehensible in rather the same way that quantitative data is rendered comprehensible by means of tables, graphs and statistical procedures” (as above). Davis and Sutton recommend matrix displays (grids) and network diagrams (flow and links) as the most useful representational devices, of which I predominantly used flow charts and diagrams. The end results have been recorded in this thesis and are aimed at describing and simplifying single cases as well as enabling comparison between cases.

As a final analytical step I tried to identify the interrelations between the main themes and drew “reaction chains” that would help to answer my research questions. Here I explored, in particular, the connection between Social Representations and Subjective Theories on disability and their influence on interaction with people with disabilities and their exposure to HIV/AIDS.

In order to control my analysis for bias I sometimes let a friend/fellow researcher question my coding system. This provided me with alternative points of view but also with reassurance. As I only used case-studies for my research this thesis did not provide sufficient evidence of statistical populations but rather for theoretical proposition. It is also impossible to establish reliability and validity in the traditional sense. Throughout my research I tried to achieve validity through using multiple sources of evidence. In this sense the final results of the ranking exercise, the in-depth interviews and the focused group interviews were used for triangulation. Reliability was approached with Burns’ understanding of qualitative research. He presses the point that reliability in such studies “is more focused on dependability and that the results make sense and are agreed on by all concerned” (Burns, 2000). The latter I tried to achieve with the focus group interview that I conducted.


Fußnoten und Endnoten

14  DPSA – Disabled People South Africa, DICAG – Disabled Children Action Group

15  DPSA - Disabled People South Africa

16  DICAG - Disabled Children Action Group

17  HEARD - Health Economics and HIV/AIDS Research Division

18  Methodology is described further down.

19  CBR – Community Based Rehabilitation, which provides care and information for people at grassroots levels

20  Here I used stakeholders form DPSA and DICAG that I had already interviewed.



© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML-Version erstellt am:
23.01.2009