What's the issue?
A vital element in successful qualitative data analysis is to respect the difference between qualitative and quantitative research. The difference between qualitative and quantitative research is, as Strauss (Strauss, 1987) puts it, not the least in how data are treated analytically.
A common way to approach qualitative data analysis is the construction of themes. Sometimes these themes have already been decided when designing the study or if the data collection is structured around these predefined themes. In other cases the themes are constructed afterwards.
When themes are not constructed beforehand it is however usual that the data analysis actually starts before the data collection is over and often data collection and data analysis are conducted in parallel, the preliminary analysis being used to decide which areas should be examined in more detail.
Coding is an important part of the qualitative data analysis and is the process of grouping interviewees' responses into categories that bring together the similar ideas, concepts, or themes that have been discovered.
The analysis of qualitative data usually involves the selection of quotes to support the presentation of the findings. Frequently such quotes are anonymous but, if the interviewee is identified, it is common practice to let him or her see the quote and the context (the surrounding text).
Pitfalls to avoid
The issue of confidentiality: It is important to respect the privacy of the interviewees and make sure that whatever information they give to you as a researcher does not backfire on them in any way. This is extremely important when working with data from children. Therefore you should have the data under good control.
- Do not leave transcripts, pictures, videotapes or whatever you are working with lying about in public.
- Do not make unnecessary copies and keep good track of the location of all copies (in both electronic and other formats)
- Do not hand your material to anyone without going over the handling procedures.
The status issue: Despite the fact that qualitative research has a long history within the social sciences, it is still quite common to see a tendency to impose the ideas of quantitative analysis on qualitative data. An example of this is when increasing the number of interviews or focus groups is thought to improve the generalisability of the findings. If generalisability is what you want, use quantitative methods.
The issue of qualitative data analysis as common sense: Everyone engages in some form of qualitative analysis in daily life. This leads some people to the erroneous conclusion that no special training is needed to analyse qualitative data except good common sense. Hopefully, though, the vastly increased use of qualitative techniques in marketing research in recent years has done much to correct these misunderstandings.
The issue of condensation: Invariably, qualitative data analysis is a process of condensation in which a vast amount of data has to be condensed in a meaningful way both theoretically and generally. This relates to at least three different problems:
- Drifting, which means that the results are poorly rooted in the original data.
- Dumping, which means that the results are simply not based on the data and at best present an oversimplified picture.
- Data drowning, which means that too much data has been collected and the researcher fails to get any meaningful grip on the data.
Questions to consider
When designing a qualitative study it is worthwhile to think thoroughly about how the data is to be analysed. Good planning can save a lot of time and energy and, as a rule of thumb, the looser the structure is at the data collection stage, the more time one can expect to spend on the data analysis.
The use of software for qualitative data analysis has increased rapidly over the past years. Researchers are however not quite agreed on whether it improves the quality of the analysis.
Grbich, C. (2007). Qualitative Data Analysis: In Introduction. London: Sage.
Silverman, D. (2006). Interpreting Qualitative Data: Methods for Analysing Talk, Text and Interaction (3rd ed.). London: Sage.
Tesch, R. (1990). Qualitative research: analysis types and software tools. London: Falmer.
In a study on Chilean adolescents the method of parallel data collection and data analysis was used. During the interviewing phase and after completing each interview (and then again after finishing a larger group of interviews) the data was scrutinised to decide which areas should be examined in more detail. This preliminary analysis was useful to redesign the subsequent interviews and to focus on central themes such as the importance and popularity of instant messaging.
(Veronica Donoso, Belgium)