Qualitative coding bubbles in Quirkos

Qualitative analysis software

What does software for qualitative data analysis actually do?

Overview

Using Quirkos for qualitative analysis

Most qualitative analysis software focuses on providing a useful tool to help researchers understand their data. Most time in qualitative research is usually taken up by deciding what a piece of data is about, and software can help keep track of excepts of text or other sources, and how they have been described as belonging to potentially hundreds of topics or themes of interest. They are essentially tools for managing, sorting and exploring qualitative data. However, it is worth describing what qualitative software won't do.

Currently, there is no software available that can automatically understand text, and do all the analysis for you. While some packages can identify a simple level of understanding (for example, positive or negative comments), a software package can't know what you are interested in, and since they usually work by assessing the probability of a section being about a particular topic, give better results with large sets of data (for example thousands of sources).

Code-and-retrieve

This is the primary function of most qualitative data analysis software.

The first role of a Computer Assisted Qualitative Data Analysis Software package (CAQDAS), sometimes called QDAS (Qualitative Data Analysis Software) is to manage the qualitative data of a project. First a user will import their data into the programme, be it text, audio, pictures or video. The software stores and categorises the data sources so they can be easily accessed and cross-referenced.

Secondly, the user will create a set of themes, topics or nodes, which represent the categories of interest to the researcher. For example these may be negative comments or discussion about a particular product or service. There are many different approaches to creating this ‘framework’ of codes which are used to categorise and analyse the text, this blog post article describes some of the methodology. Again, the role of the software is to allow the user to manage and sort these themes, and is able to ‘remember’ many more topics than a human user could at one time. The software will also allow the user to create a hierarchy of codes, which might have sub-categories, and allow grouping or searching for easier use.

Now the user brings these two elements together: they look through the data, and identify particular sections as being about something of interest, by associating it with one of the codes/themes in the framework. The software records the association, and thus a piece of data is ‘coded’ or 'tagged' as being about a particular theme. Researchers then go through the data, line by line, identifying and coding data as being about something.

The ‘retrieve’ section of the code-and-retrieve function is the ability of the software to show all of the data coded at a particular node or topic. Thus, the user may want to see everything that different sources say about a topic – for example all negative comments. The software will create a view or a report by retrieving coded data according to the request of the user. Essentially, these functions are the front end of a computer-based database which records and displays data stored in it. However the data in this case is generally qualitative: rather than numbers, the data is text, visual or audio segments of generally unstructured data.

It's worth briefly noting some of the limitations with qualitative software, for more see the classic text by Fielding and Lee (1991). They note that this process has a tendency to quantify the qualitative data, and make it temptingly easy to equate the numbers of quotes for a particular theme as being significant. It is also difficult in this approach to capture more nuanced 'untypable' aspects of the data, and the wider temporal and societal story. That it is called qualitative analysis software at all is a misnomer, the software cannot do the analysis process. It is only a tool to assist the brain and judgments of the researcher, and does not negate the need to know the data in great detail.

user of qualitative software

Exploring qualitative data

distilling qualitative data into an idea

While retrieving data stored under a particular theme is a valuable function, most qualitative analysis software packages also provide a number of tools to further explore the coded data. For example, the user might be interested in data only from certain sources (such as respondents with a particular age range) or might want to see connections between coded themes. Users may wish to search through the text of a project for keywords, and see how much text has been coded onto a particular code.

Generally qualitative software can also help to facilitate team work, for example a user might want to see coding done by an individual working as part of a team, and compare the difference between the understandings of two different coders working with the same sources. Often, researchers might chose a section of data to code independently, to check the validity of their approach, and comparing the data using a function such as inter-rater reliability. This would provide a statistical measure illustrating the similarity of the coding from two different people.

Some packages also go further, by providing features to help researchers with their conceptual understanding of the data. At a basic level, this might include the ability to record notes or memos about the sources or their coding, or even provide a space to draw conceptual diagrams linking themes.

Exporting and sharing qualitative data

Finally, qualitative research software will have various methods to export and share the coded data. It will often be necessary to include illustrative quotes to include in a report, to generate a complete list of all the coded data in the project, or to create graphical summaries of the project. Additionally, there may be the requirement to export the data into a format that can be used in another software package for further analysis or visualisations, especially for statistical work.

All packages have slightly different approaches to these tasks, and provide other additional features. There are dozens of options for software to assist with qualitative data analysis, with varying levels of complexity and cost. These include open-source and free packages, as well as fully-featured commercial options. A good, independent overview can be found at the CAQDAS networking project based at the University of Surrey.

Quirkos is one such package for qualitative researchers, with an emphasis on ease of use, low cost, and providing cross-platform operation with a graphical interface. It is currently the newest entry into the market, taking advantage of modern database technologies, and a very different user interface which provides live views of data as it is coded.

A detailed overview of the functions available in Quirkos can be found on the features page. However, we believe that experimenting with software is the best way to assess its suitability, so we provide a one month free trial of the full version of our software.

If you are looking for more information about using qualitative analysis software, there are some good textbooks available. Highlights include Silver and Lewins (2014), and a chapter by Clive Seale in Silverman (2013). However, a software approach does not suit all researchers and methodologies, this article by Pope et al. (2000) gives a wider overview of thinking about the analysis process. You can also look at local networking organisations dealing with qualitative research tools, such as the CAQDAS network in the UK, Digital Tools for Qualitative Research in the USA and Computación Cualitativa Consultores in Latin America.

an example of a qualitative report