How do we define qualitative research in a new MR world?

MRIIThis post looks at the definition of qualitative research and has been produced as part of a project I am doing with the University of Georgia’s Principles of Marketing Research team to update their Mobile Market Research course. Your thoughts and suggestions are invited! Indeed, the thoughts expressed here arise from a lively and informed discussion in the NewMR LinkedIn group.

The growth in new research approaches and the emergence of new tools have raised questions about what the term qualitative means, and by extension what quantitative means. If we can collect hundreds of videos, thousands of pictures, or millions of quotes, does it challenge what we mean by qual and quant.

Traditionally, qualitative research was relatively easy to differentiate from quantitative research. The Table below shows how qualitative and quantitative research were traditionally differentiated.

Table

However, this simple table is less helpful in today’s modern, mobile, social, passive data collecting world. Some tools are used by both qualitative and quantitative researchers, and the types of data collected have expanded and overlapped.

Today, researchers can collect large amounts of text, images, and videos, and these large amounts can be collected from the same sorts of samples that are used for quantitative studies. Using social media and mobile devices allow vast quantities of unstructured data to be collected. These changes have made data that has been traditionally the preserve of qualitative research available to quantitative researchers. At the same time the growth of techniques such as online discussions, MROCs, mobile diaries has resulted in many qualitative projects collecting semi-structured data, with some going as far as using survey software to collect their data.

The changes in the range of data collection options available to researchers mean that it is often not possible to describe qualitative research (or quantitative research) in terms of the tools used to collect the information. For example the same software can be used to collect 10 in-depth mobile dairies as a part of a qualitative project, or 3000 quantitative diary sets.

The difference between qualitative and quantitative research lies in the analysis. Qualitative research depends on applying human understanding and is based on creating a story from the data. Usually the human understanding, i.e. the mind of the researcher, engages with the data that were collected (e.g. text, images, artefacts etc) rather than with the outputs of automated processes. Quantitative research depends on using an algorithm to produce numbers that can be projected to a relevant group.

When qualitative analysis is being conducted, the data has to be restricted to the amount that can be handled. This is why most qualitative research is based on small data sets, rather than the analysis being qualitative because there is less data. However, quantitative tools can be used to augment the qualitative analysis process, for example software that helps organize, tag, and sort text, images, and videos.

Constructionist and Positivist
The table above uses the terms constructionist and positivist, here are my quick and dirty definitions of these two.

Constructionist  A philosophical position that assumes that views of the world are constructed by people, not discovered. It tends to focus on useful descriptions of phenomena, rather than ‘scientific’ truths. Qualitative researchers often take a constructionist position, either because they believe the nature of the things they are researching are too complex to apply a scientific approach, or because their own philosophical position is constructionist.

Positivist A philosophical position that assumes there are real facts and truths, that they can be measured, and that knowledge that should be verified by applying a scientific method. Most scientists and quantitative researchers adopt a post-positivist position, i.e. one that accepts subjective criteria are required in creating assumptions, but endeavors to uses a scientific, ‘fact’-based approaches with data.