Computer-Assisted Qualitative Data Analysis Software

Data Management

Data management involves organizing and storing data sources (e.g., transcripts, audio clips), tracking key characteristics of data sources (e.g., demographic data of interviewees, document identification for archival images), collecting process notes (e.g., coding, theoretical memos), etc. Without CAQDAS, researchers might create physical files, organize computer folders and cloud-based storage, and record logs of key information. With CAQDAS, the process is eased, as most software programs act as a repository that stores and organizes all the materials for each project in one centralized location. Most CAQDAS programs will name and sort data sources, allow tagging of files with ancillary information, track process information (e.g., theoretical memos, query logs), and create automatic backups.

Data Coding

Data coding refers to marking a chunk of data (whether text, image, audio, or video) with a tag that signals a particular meaning or category. Without CAQDAS, researchers might print a hardcopy of a set of interview transcripts and then code the text with highlighters, colored ink, or annotations written in the margins. Alternatively, with word processing software, researchers might code with font colors, font faces, underlining, or comments.

With CAQDAS, the process is similar in that researchers apply codes to excerpts. But the specific tactics for coding vary by program. Some CAQDAS programs feature automatic coding. These programs search datasets and automatically apply a designated code to every occurrence of a particular word or phrase. Typically, researchers will designate how far to expand the excerpt to code—from the word or phrase only, to the line or sentence in which it occurs, or to its full paragraph. Other CAQDAS programs rely more heavily on researchers’ coding decisions. Instead of being automatic, researchers read through the data sources and select excerpts to be coded. With greater researcher involvement, passages can vary in length—from short phrases to several paragraphs or pages long. Also, researchers can better identify passages that share a meaning even if they do not contain particular words. Researcher-driven coding can still be assisted by CAQDAS, such as when researchers perform complex keyword searches (e.g., searching for any word among a list of words) across entire datasets to locate passages that might be a potential match for a code. Many CAQDAS programs also allow for hierarchical clustering of codes, such that codes can be designated to be automatically grouped together under a higher-order code.

...