A client recently asked whether I would recommend a particular software program to assist in analysing qualitative research data. There are a number of these types of programs around; two or three are well known. She told me her department was considering purchasing it to help them collate and interpret a growing number of consultation findings and evaluation reports. I told her that although I had not used any computer software programs to assist with qualitative analysis, I was well aware of their limitations, and this is what I explained to her.
Software programs such as SPSS (Statistical Package for the Social Sciences) are well used and respected to help analyse quantitative data. They allows for in-depth data access and preparation, analytical reporting, graphics and modeling. There are also online survey software packages that can prepare reasonably sophisticated analysis of statistics, segmentation, trend analysis and some basic text analytics. Analysing statistics and quantitative data is a straightforward process in that it requires a straightforward application of codes and formulae. The same cannot be said about the analysis of qualitative research findings.
Arguably, the most important development in modern social research and evaluation has been the adoption of a mixed methods approach. Data for any one project is likely to be gathered from a range of sources and in many forms, including interviews, focus groups, paper-based surveys, online surveys, social media and telephone interviews. When working with mixed method research and trying to make sense of disparate data forms, it can be enticing to reach for something that promises to bring all of the information together for you and help you make sense of it.
Computer software designed for qualitative analysis can work as a useful data management tool for efficient storage, organising and retrieval of qualitative data. But these programs do not perform analyses. Computer software for qualitative data is predominantly designed to help you find important patterns in your data that speak to your research questions and then look under the surface of your data to explore the rich layers of underlying meaning. They can assist with basic administration tasks and can perform them efficiently. They take swathes of text from qualitative research findings and source commonly-used words, create categories, and find logical connections that can be used to organise and interpret the electronic data. In essence, it scans text for keywords, well-used phrases and other patterns of text. Text software may be useful if you want to analyse and link together verbatim statements, perhaps from a bundle or surveys that contain open-ended questions. It cuts down the amount of time needed to work through the text.
But the soundness of analysis of qualitative information relies on the judgements made by skilled researchers or consultants. This is particularly the case within complex evaluation or social research projects that used mixed research methods. In these types of complex projects, qualitative analysis is more to do with the analytical processes in the mind of the consultant or researcher, their experience, contextual knowledge and their ability to take a range of input and complex data and apply it to a problem and make sense of it.
The most successful evaluation and social or market research projects are those conducted by well trained and experienced researchers or evaluator who have spent years interacting with real people, in real time, listening to what they say, as well as listening for what they don’t say, as this can be revealing. They will have experience or at least an understanding of your sector and appreciate the current issues and trends. Computer software for qualitative research cannot replicate these skills or experience.
When a client buys a service from a consultant, they are not just buying a focus group, a series of interviews, a survey or a literature review. They are buying the expertise of an experienced person who has a good understanding of their sector and can appreciate their problem. There are many aspects to approaching qualitative research, above and beyond text analysis of dead data.
So, my advice to my client was that if she were to invest in the software, to be aware of its limitations. If she is simply looking for a way to better store and organise qualitative information then computer software programs may be the solution. But if she is hoping for the program to be able to make some sense of the qualitative data, investment in a qualitative computer software programs is not likely to be money well spent.