What is qualitative research?
Qualitative research is a method of enquiry that explores the meanings that people hold, how they make sense of the world around them and their perceptions of it. It is an umbrella term covering a number of different approaches that have differing methodological approaches. However, the different approaches have several things in common. One thing they have in common is that data collection never relies on experimentation (as it does in quantitative research) and, instead, might involve such methods as structured or unstructured interviews, observation, focus groups and printed media sources. A second common feature is that qualitative methods do not believe the world holds absolute truths that are waiting to be discovered; qualitative approaches are typically more focussed on the way meanings are constructed and negotiated in the social domain. A third common feature is that subjectivity is acknowledged as being an inescapable factor in any type of research and this is embraced rather than ignored.
What are the main features of qualitative interviewing?
The main features of qualitative interviewing depend on the type of interview as there is a lot of variability between approaches:
- The structured interview features the same rigid, pre-defined questions for each person (in the same order). It is on the fringes of qualitative research as it has much in common with quantitative methods in that enter a project with a set research question the researcher is interested in and design the experiment (or in this case the questionnaire) to fit this objective
- The semi-structured interview allows more flexibility as the researcher has a schedule but does not have to stick to it rigidly. Therefore, if anything interesting occurs it can be explored. The participant is acknowledged as being important in the interview shaping process and the researcher makes an effort to understand his or her unique world.
- Researchers conducting unstructured interviews may have a broad range of topics in mind; or they might have none. The interview can go in any direction.
What are the main features of focus groups?
Focus groups feature a small group of participants (typically 4-8). The researcher will usually guide the research in the direction he or she wants (and ensure everyone has a say) but the direction will largely be guided by the members of the focus group who will talk with each other as well as the researcher. Therefore, as well as hearing individual accounts, the social interaction that takes place can provide a wider ranging, richer form of data. The focus group will be recorded (via video or audio only) to be later transcribed. Data analysis depends on the epistemology of the researcher.
What is a theme?
A theme is a recurring pattern across a data set. It is a fusion of individual snippets of data; vocabulary, ideas, meanings, experiences that appear to share some degree of similarity when brought together. Themes vary depending on the approach of the researcher. For example, if several men talked about not wanting to visit the doctors because they didn’t want to appear weak, a semantic theme would confine the theme to exactly what was said and would not make any inferences about it. In this example, a recurring semantic theme might be something along the lines of “avoiding the doctor to prevent appearing as weak”. However, a latent theme might explore the social construction of masculinity and how this is influences the men’s perceptions and experiences. For example, the theme might be “conformity to traditional masculine stereotypes”.
The main steps involved in performing a thematic analysis include:
- This involves transcribing, word for word, the audio recording of the interview, focus group etc. Often it includes non-verbal artefacts, such as coughs and pauses, as these can influence the meaning of what was said. It is an extremely important part of the familiarisation process where the researcher starts to get a feel for any recurring patterns in the data.
- This involves reading and re-reading the data so that the researcher becomes immersed in it. Some initial codes are generated that represent interesting data snippets.
- The data is then coded paying equal attention and importance to every sentence and not ignoring any contradictions or inconsistencies that arise.
- Theme generation. The next step is to collate the codes and look for recurring patterns in the data set. An initial thematic map might be produced at this stage.
- Theme refining. Themes can be combined, excluded, reworked etc. at this stage until they represent an accurate reflection of the story the data is telling. They can be named at this stage and refined until a final thematic map is produced.
- The researcher comments and explores the themes and the meaning within them to paint a picture of the data. The degree of interpretation depending on the researcher’s theoretical position.
How can the quality of validity of a qualitative/thematic analysis be improved?
Critical self-reflection along the entire process of thematic analysis can vastly improve the quality and validity of thematic analysis. It is important the research recognises the limitations of the research and the pivotal role he or she plays in the process and the influences this has on the data collection, the content and the interpretation of findings.
Member checks offer a way to improve validity and involve the researcher being open and building report with the interviewee and summarising and articulating what they believe the interviewee is saying and checking this with him or her. After the data collection, member checking may also allow the interviewees to critically analyse the findings and conclusion that were made and allowing them to affirm the accuracy of them.
Triangulation is another way of improving validity and involves analysing a research question from multiple perspectives. There are various ways of triangulation; a common one is investigator triangulation in which a number of researchers approach the data collection process and triangulate their findings.
Finally, negative case analysis can be used to search and explore the cases within the data that contradict (or at least do not support) the themes/conclusions the researchers have drawn. This brings transparency into research process by identifying (and discussing the impact of) the limitations.