CHAPTER 3 RESEARCH METHODOLOGY
This chapter covers the method that was used to carry out the study. The overall view-point of the researcher in terms of the philosophy that informed the study is explained herein. A brief discussion in which the pros and cons of the qualitative paradigm compared to the quantitative paradigm is executed. After that, the research design is elaborated on. In the section following the research design, there is a description wherein the research population as well as the sampling methods are outlined. Thereafter the literature review proceeds to look at the actual procedure itself. The study was anchored on two data collection methods namely the FGDs and in-depth interviews, which are defined in terms of their characteristics, strengths and weaknesses. After the brief discussion on the procedures, there is an explanation of how the research tools of the study were developed and a description of the pre-fieldwork preparations, followed by a section on going into the field. Data analysis, trustworthiness and ethical considerations from the last part of the methodology before concluding.
The overall philosophical approach that was adopted by the researcher and that guides the study is stated after looking at two other philosophies.
Positivism believes that the world is unaffected by and is independent of the researcher (Snape, Spencer, Ritchie, Lewis, Elam, Arthur, Nazroo, Legard, Keegan, Ward, Finch, O’Connor, White, & Woodfield 2003:16). It endorses the methods of the natural sciences as being appropriate for social enquiry (Snape et al 2003:23).
Interpretivist and constructivist stances tend to underpin some qualitative methods while positivist stances tend to be supported by quantitative methods (Snape et al 2003:18). The aim of the researcher in the interpretivist-constructivist viewpoint (which started in the 1980s) is not to predict social phenomena but instead to understand it. Lather (2006) states that though both fall well within the ‘understand paradigm of inquiry’, they seek understanding in diverse ways. The interpretivist researcher has the goal to understand social phenomena by attempting to access the meanings that people assign to social phenomena. Constructivists understand meanings as something that through experience is constructed. The research is thus focused on understanding and identifying the processes of how people acquire or construct different meanings over time (Stinson & Bullock 2013:1257).
Constructivism has its origins in sociology and its aim is to display ‘multiple constructed realities’. This is achieved through the shared investigation of meanings and explanations, by both respondents and researchers (Snape et al 2003:12).
Interpretivism holds the view that the researcher and the social world impact on each other. The researcher focuses on exploring and understanding the social world using both the researcher’s own understanding as well as the respondents’ understanding (Snape et al 2003:17). In considering the input of respondents and researchers, interpretivism and constructivism thus have a commonality.
In interpretivism, facts and figures are not distinct and it is inevitable that the findings are influenced by the researcher. This can be mitigated by the researcher being transparent about their assumptions. It is recognized that the natural sciences’ methods are inappropriate since the social world is not governed by law-like regularities but rather, it is mediated through meaning and human agency (Snape et al 2003:17). Qualitative research is largely associated with interpretivism (Snape et al 2003:23).
The positivist view-point was inappropriate for the qualitative study being undertaken by the researcher. After some consideration, the researcher selected interpretivism as the philosophical stance of her study. As with any school of thought, there are pros and cons. While interpretivism is responsive to respondents, it nevertheless lends itself to some criticism. In this regard, two issues that are normally raised are concerned with identifying the aspects in a researcher’s findings and interpretation that should be accepted, as well as with the possibility of the findings not being generalizable (Eslami, 2013:193). To mitigate these problems, the researcher depended on the trustworthiness of the study, as discussed towards the end of this chapter where the researcher ensures that there is reflexivity while at the same time focusing on the respondent’s views when discussing the findings. The researcher also ensured the transferability of the results by engaging in thick description. Authenticity ensured that a range of realities were described.
Qualitative research was selected for the study.
In choosing qualitative research in terms of its appropriateness for the study, the researcher took several things into consideration including the following: what exactly the researcher was trying to find out; the level of detail of the phenomenon required; how other researchers have dealt with a similar topic and the extent to which alignment with this literature is wanted; practical considerations including the duration of study; if quantitative or qualitative methods assist in learning more about the subject; and commitment to a particular research model and the implication of a particular approach (Silverman 2013:13).
The research was not as much concerned with numbers and statistics (questions that lean to ‘how many’) as it was with exploring and attempting to understand the information that was gathered, collected and analysed in relation to the phenomenon of HIV and AIDS communication for urban Zimbabwean women (questions that are inclined towards ‘what’, ‘how’, ‘why’).
Qualitative research has the goal of refining information from samples to understand and interpret phenomena. Quantitative research differs in that it uses representative sampling to find out a generalizable law (Eslami 2013:192).
Qualitative research is defined as “A research study based on non-numerical data”. Non-numerical data covers “pictures, words, statements, clothing, written records or documents, or a description of situations and behaviour” (Christensen et al 2011:29).
Describing qualitative research
Qualitative research is interpretive research that relies on many types of subjective data and investigates people in specific situations in their natural environment. In this type of research, the researcher examines the data before and after collection the course of the study and continually attempts to understand it from the subjective perspectives of the respondents, which is pivotal. Thereafter, the research becomes an ‘objective outsider’ and relates the interpretive-subjective data to the research purpose and questions. Qualitative research is, by comparison, more flexible than quantitative research in that the research questions are permitted to evolve, and even change during the study since the focus is on exploration of phenomena. Therefore, qualitative research can generate theories. Quantitative research is more rigid as usually the focus is on testing a hypothesis (Christensen et al 2011:52-53).
Qualitative research employs multiple data collection methods and this assists the process of obtaining the best description of a phenomenon and the interpretation of how respondents best understand it. This is evidence for triangulation in the study, and will be elaborated on later. This research type does not rely on numerical data interpretation (Christensen et al 2011:53-54).
Qualitative research is also described as a type of social enquiry whose focus is the way in which people make sense of their experiences and the world they live in. The different approaches in qualitative research have the same goal which is “to understand, describe and interpret social phenomena as perceived by individuals, groups and cultures”. Qualitative approaches find use in the exploration of change or conflict (Holloway & Wheeler 2010:3). Some of the features of qualitative research are:
The data have priority or primacy; the framework is not predetermined but is derived from the data directly.
It is context-bound and therefore context sensitivity is important.
Immersion into the natural setting by the researcher is necessary.
Focus is on the views of the respondents and their perceptions, meanings and interpretations.
Thick description is used: describing, analysing and interpreting as well as going beyond the respondents’ constructions.
There is the basis of equality in human beings applied: between the researcher and the respondent, and the relationship is close.
Reflexivity (explained later in detail) is required whereby the researcher is the main tool and their stance is made explicit (Holloway & Wheeler 2010:3-4).
Table 3.1 compares some of the differences between qualitative and quantitative research.
Table 3.1’s characteristics of qualitative study can be applied to this study as follows: the researcher’s aim was to explore, understand and also describe the respondents’ life world using an approach that is context-based; the sampling was purposive and limited to the cities of Harare and Bulawayo in Zimbabwe; the data collection was in the form of in-depth interviews as well as FGDs; the data analysis used the Colaizzi method; the research team had a close relationship with the respondents; the researcher used various strategies to ensure rigour in the study, as explained later under the section on trustworthiness.
Two research designs were selected for the study for the purposes of triangulation: phenomenology and qualitative case study.
The aim of phenomenology is to understand the ‘constructs’ people use in daily life to make sense of their world. Meanings that are embodied within conversation or text are uncovered. It has its origin in both sociology and philosophy. It has also been used synonymously with the term ‘ethnomethodology’ (Snape et al 2003:12). It is described more concisely as a “Qualitative research method where the researcher attempts to understand and describe how one or more respondents experience a phenomenon”. The aim is for the researcher to gain access to the life world of the respondent or respondents. The life world is described as the inner world of subjective experience or ‘phenomenal space’, where the ‘lived experiences’ are; where the immediate consciousness of a person exists; where the person can feel, sense and have ‘inner’ talk (Christensen et al 2011:368). Lived experiences refers to how respondents are making sense of their personal and social world. Of most importance are the meanings that specific experiences, events and states hold for the respondents (Smith 2015:25). The researcher attempted to understand and describe the experiences of the respondents from the perspective of the respondents, by making use of various variables of the study. How the respondents made sense of or understood their experiences, was important.
There are two types of phenomenology, namely Husserl (descriptive) and Heidegger (interpretive) phenomenology (Reiner 2012: ). The researcher applied descriptive rather than interpretive phenomenology. This was because the researcher decided to bracket any biases (Reiner, 2012: ). This agrees with Colaizzi’s method, which instructs the researcher to validate the findings by returning to the study respondents (Reiner 2012:
). Colaizzi’s method of analysis, which the researcher used in the study, is one of the methods of analyses used in descriptive phenomenology (Reiner 2012: ).
Case study design
In case study, the research centres on a question and not on a method. Case study is a method that subsumes a variety of methods depending on what is desired. It is a means to an end; the end being answering a question; it is not an end. It is a design frame within which a wide range of research methods can be used (Thomas 2016:92). The case study design is good for obtaining a rich picture and gaining analytical insights (Thomas 2016:22). Case study is designed for exploration in a ‘real life’ context (Thomas 2016:10).
Case study research is a major approach to qualitative research. In this method, the researcher gives a detailed description and account of at least one case. A case is a bounded system such as a person, group, organization, activity, process or event (Christensen et al 2011:374). In the current study, the researcher takes the meaning of a case to be that of a group, consisting of young urban women aged 20-29 living in and have lived in either Harare or Bulawayo for at least the past 12 months. To achieve this the women undertook in-depth interviews. In this context, the unit of analysis is the group.
Case study is greatly associated with qualitative research, albeit it is used in a variety of ways. It may even seem as though it is a synonym for qualitative research although this is not the case (Snape et al 2003:51).
The fundamental question in case study research is, “What are the characteristics of this single case or of these comparison cases?” Differing sources and methods of data collection can be used for case study research: in-depth interviews, documents, test results, and archival records (Christensen et al 2011:375). The researcher selected in-depth interviews for the case study, together with FGDs. Life history and context are also important to capture for case studies, and this was done. An instrumental case study is different from an intrinsic case study. The latter is only interested in understanding the individual case. The researcher made use of the instrumental case design as what is important for the study was to understand the phenomenon and not one specific case. A template was designed for this purpose. This design gives insight into an issue or it can be used to develop, refine, or alter some theoretical explanation (Christensen et al 2011:375). The researcher’s study sought insights into HIV and AIDS communication among young urban Zimbabwean women as this will assist in drawing up guidelines for the target population.
Sampling involves selecting a set of people from a population; it is drawing a sample (Christensen et al 2011:163).
A sample is the set of elements that is selected from a population (Christensen et al 2011:150).
A population is the complete set of elements (basic unit selected) from which the sample is selected (Christensen et al 2011:150).
Defining study population
The population that was selected for the study, was that of women aged 20-29, are Zimbabwean, and have been living in either Harare or Bulawayo for at least the last 12 months. The figures from the last census in Zimbabwe indicate that 9.2% of the population were females aged 20-24 for females, while 8.7% made up females aged 25-29 (ZIMSTAT 2013: 18). This translates to 17.9% females for the age group 20-29, or just over 2.3 million females, since the population of Zimbabwe at the time of the census was just over 13 million, as indicated in Chapter 1.
Qualitative research is different from quantitative research as it does not aim to estimate the incidence of phenomena in the wider population: it does not focus on statistical representation or scale. Priority in sample design is given to the ability to represent salient characteristics. Therein lies the precision and qualitative rigour (Snape et al 2003:81-82). Qualitative research focuses on in-depth understanding rather than on the breadth of the study of many cases. Information-rich cases are sought. People who the researcher believes to be information-rich are continually selected (Christensen et al 2011:162). In short, the emphasis is on ‘quality’ of data rather than on ‘quantity’ of data.
In non-probability sampling, deliberate selection with specific features in mind is carried out: the goal is not to have a statistically representative sample but rather the characteristics of the population are used as the basis of selection and the chances of selection for each element remain unknown. The searching out of specific features is what makes for suitability for small-scale in-depth studies. Although probability sampling is generally considered to be the most rigorous type of sampling for statistical research, it is not considered appropriate for qualitative sampling (Snape et al 2003:78).
Non-probability sampling was used during the study as part of a combination of sampling techniques: for the FGDs and the individual interviews, convenience, purposive, cluster and snowballing sampling were used; for the key informant interviews, purposive and snowball sampling were used. The key informants were those selected from organizations that work in HIV and AIDS communication, deemed to be experts in this area, namely government entities and their partners (NGOs).
In purposive sampling, sample members are selected with a ‘purpose’ to represent a location or type in relation to a key criterion. It is also known as criterion-based sampling (Snape et al 2003:79). After specifying the characteristics of the population of interest, the researcher locates individuals matching the needed characteristics (Christensen et al 2011:159). Purposive sampling was carried out for the FGDs as well as for the in-depth interviews for both the sub-population and the key informants, as specific characteristics were required for the selection criteria for the respondents to be met.
In convenience sampling, the researcher chooses the sample according to ease of access (Snape et al 2003:81). It makes use of people who are easily or readily available, are easily recruited or who volunteer for sample inclusion (Christensen et al 2011:158). This was done for sub-population in-depth interviews and FGDs, as ease of access to the respondents and their availability for data collection, had to be at their convenience.
This method makes use of sampled people who are asked to identify other potential respondents with the inclusion characteristics (Christensen et al 2011:159). This proved very effective for the duration of the study. This sampling type was selected as it was easier for the initial respondents (the young urban women and key informants), than it was for the researcher, to identify and refer for possible inclusion, potential respondents who met the study inclusion criteria.
This is when clusters are randomly selected rather than individual type units (such as individual people) in the first stage of sampling. A cluster is a collective type of unit that includes multiple elements. Examples of a cluster are neighbourhoods, families, schools, classrooms and work teams (Christensen et al 2011:157). The researcher adapted this type of sampling and ‘purposively’ selected clusters for the study. The researcher purposively selected clusters in the form of residential areas to ensure that high, medium and low-density suburbs were all represented in the sampling for both interviews and FGDs, in both the cities of Harare and Bulawayo.
Inclusion and exclusion criteria
Key informants were eligible for interviews if their organization worked directly in HIV communication (implementing partners/NGOs), and were not, for example, donors. The eligibility criteria used for the sub-population in the study are shown in Table 3.2.
In qualitative studies, the samples are usually small for three key reasons:
There is a point of diminishing return where even if the sample size is increased, no new evidence is realized.
Incidence and prevalence statements are not of concern as no statistical inference is required.
The findings are rich in detail and so the size must be kept reasonably small for it to be practical to analyse them. It would otherwise become unmanageable to conduct and analyse, groups and observations in a reasonable time frame (Snape et al 2003:83-84).
The sample size for the FGDs was 62 (eight FGDs), 25 for the interviews of the women, and five for the key informant interviews.
Collecting qualitative data can be divided into two broad groups: one approach focuses on naturally occurring data while another focus on generated data through the research interventions. The researcher chose to use generated data for the study (Snape et al 2003:34). According to Bryman (2001), generated data involve ‘reconstruction’ which is reprocessing and retelling of attitudes, beliefs, behaviour or other phenomena. In the final analysis, this type of data offers understanding of the meaning that people attach to their own perspective on and interpretation of their beliefs and behaviours (Snape et al 2003:36).
Of the various methods available for qualitative research, the researcher used a combination of in-depth individual interviews together with FGDs. During pretesting, the researcher established that the two should be implemented and completed for a specific location/suburb to avoid going back for the other. By using FGDs and in-depth interviews, triangulation was achieved.
TABLE OF CONTENTS
LIST OF ACRONYMS AND ABBREVIATIONS
LIST OF TABLES
LIST OF FIGURES
1 ORIENTATION OF STUDY
1.2 Study rationale
1.3 Source and background of research problem
1.4 Statement of Research problem
1.5 Research Aim
1.6 Research objectives
1.7 Research questions
1.8 Theoretical Framework
1.9 Definition of key concepts
1.10 Research methodology
1.11 Ethical considerations
1.12 Significance of study
1.14 Outline of thesis chapters
2 LITERATURE REVIEW
2.2 The HIV and AIDS epidemic
2.4 Theoretical framework
3 RESEARCH METHODOLOGY
3.3 Research paradigm
3.4 Research design
3.7 Sampling method
3.8 Sampling procedure
3.9 Data collection
3.10 Data analysis
3.12 Ethical considerations
4 DATA ANALYSIS AND PRESENTATION OF RESULTS
4.2 Pilot phase
4.3 Data collection phase: demographic data for target population (urban women aged 20-29 years)
4.4 Results from individual interviews
4.5 Results for FGDs
4.6 Key informant organizations
4.7 Results for key informant interviews
5 DISCUSSION, CONCLUSIONS, CHALLENGES & LIMITATIONS
5.2 Demographic data
5.3 Theme 1: Communication
5.4 Theme 2: HIV and AIDS knowledge
5.5 Theme 3: Perceptions and beliefs
5.6 Theme 4: Young urban women’s context
5.7 Challenges and limitations of the study
6 RECOMMENDATIONS, GUIDELINES AND CONCLUDING REMARKS
6.3 The role of World Health Organization (WHO)
6.5 Concluding remarks
LIST OF REFERENCES.
GET THE COMPLETE PROJECT