Home page > Data analysis > Text analysis > Qualitative analysis

Qualitative analysis

Brief description

In cases where there are a limited number of complex textual sources (such as one or more literary or philosophical texts, transcripts of interviews, reports, etc.), the research topic focuses on certain relationships between phenomena, the approach is interpretative, and data collection and analysis typically follow each other in a cyclically interactive manner, then it is usual to apply qualitative data analysis. Qualitative analysis often has two approaches: (i) describing the data and (ii) identifying themes in the data. Describing data involves adding comments, analyses and interpretations. Identifying themes involves tagging the data. This includes locating words and/or passages in the text (or relevant parts of audio or video files) and providing them with a label to identify a particular concept or theme.

Tasks / activities

Collecting data	Data for qualitative analysis can be collected in different ways. The interview instrument is a popular way to obtain information, but you can also use existing textual data (such as journalistic texts, reviews and autobiographies), or observations (in the form of notes or recorded audiovisual material, for example).
Encoding data	Encoding data is an important tool for the researcher to get a handle on the material. These codes describe the content-based elements that are important to the research; they may be a more or less objective description of the material, but can also be interpretations on the part of the researcher. This involves segmenting the text into various pieces that can be revelant to the research. This segmentation can happen at various levels: paragraphs, sentences, phrases, words, etc.  The marked segments can be named and classified using  labels and codes chosen by the researcher. This process is also called annotation. The great advantage of using specialized software for this is that it allows you to check the consistency of the encoding at any time and make adjustments if necessary. Although the encoding is usually added manually, some programs offer tools that automatically encode the text.
Exploring data	Before analysing the data, you can make multiple different selections of parts of the qualitative data, such as all segments that treat a certain subject. This can be done by searching for a specific label, whch then yields a neat overview of all results. Segments can be selected by searching for multiple labels simultaneously, by combining  them  with the logical operators ' or', 'and' and ' not'. The various selections that can be made are usually not limited to the encoded text, as you can also make concordances of words that occur in the text(s) or search for document characteristics.
Analysing data	The encoding added to a text can be analysed in multiple ways, depending on which software you use. For example, the program Kwalitan can be used to request a list of all labels, giving you insight into which labels have been used and how often they occur. In addition, labels that refer to the same underlying concept can be combined into categories and, if applicable, you can create hierarchical relationships between the labels and display them in a tree structure, ordering them into a multi-level parent-child structure. Some programs also allow users to subject labels and combinations of labels to simple statistical analysis.

Tools

Various computer programs have been developed to support qualitative data analysis. Analysis performed with these programs is typically called Computer Assisted Qualitative Data AnalysiS(CAQDAS). In our faculty the program Atlas.ti is used. This program can also work with digital audio and video files, which, in the case of interview analysis, for example, can be linked to transcripts.
Other commonly used programs (that are not available in our faculty, but which have demo versions that are useful for short-term research which have student discounts) are Kwalitan and Maxqda
Although these programs all have their differences, they have in common that they can be used to encode (classify) research material, as well as retrieving and analysing this material.

Further information

Online QDA
Website with extensive information about qualitative data analysis. This website also contains a comprehensive introduction to Computer Assisted Qualitative Data Analysis

Lewins, Ann Christina Silver (2007). Using software in qualitative research: A step-by-step guide. Los Angeles, CA (etc.): Sage. Info. [Available at the VU University Library]
This book provides an overview of the tasks and processes that are part of qualitative data analysis as well as comparing a number of computer programs (including Atlas. ti) on how they can support analysis.

Schreier, Margrit. "Computer-Aided Qualitative Data Analysis and its Uses in the Empirical Study of Literature." The Systemic and Empirical Approach to Literature and Culture as Theory and Application. Eds. Steven Tötösy de Zepetnek and Irene Sywenky. Edmonton: University of Alberta, Research Institute for Comparative Literature and Cross-Cultural Studies [etc.], 1997. 155-62.

Other topics in this section: Introduction   Basic text analysis   Content analysis   Corpus analysis   Sentiment analysis   Text mining   Stylometry