CAQDAS (Computer-Assisted Qualitative Data Analysis Software)

Program Functions

Most programs work with textual data. Although it is not necessary to transcribe ethnographic notes, narratives, and so forth, it is usually advantageous because most software supports the rapid recontextualization of extracted and/or coded passages by displaying the original text. Most programs support the coding of text, and such codes can be arranged hierarchically, modified, and rearranged as the analysis proceeds. Some software (e.g., MaxQDA, Qualrus, Atlas.ti, and NVivo) can work with rich text that includes different fonts, styles, and colors. Several programs support the analysis of digitized audio, images, and video, in some cases allowing the direct coding of digitized material (e.g., HyperRESEARCH, Atlas.ti, CI-SAID, AnnoTape, and the Qualitative Media Analyser). So far, there is no software that can automatically transcribe from audio recordings. Even the best speech recognition software still relies on high-quality recordings and familiarization with one speaker and cannot cope with multiple respondents' voices.

Most software allows the writing of memos, code definitions, fieldnotes, and other analytic texts as part of the process of analysis, and some (e.g., NVivo, NUD*IST/N6, Atlas.ti, and Qualrus) allow this text to be linked with data and codes in the [Page 88]project database. A key feature of many programs is the ability to search online text for words and phrases. Typically, such lexical searching allows the marking of all the matching text wherever it is found in the documents. Some programs support the searching of the database using the coded text. The most sophisticated of these allow complex searches involving various Boolean and proximity combinations of words, phrases, coded text, and variables (e.g., NVivo, MaxQDA, WinMAX, Atlas.ti, NUD*IST/N6, HyperRESEARCH, and Qualrus).

A third generation of software has extended these functions in two ways. Some assist analytic procedures with a variety of facilities to examine features and relationships in the texts. Such programs (Qualrus, HyperRESEARCH, Ethno, and AQUAD Five) are often referred to as theory builders or model builders because they contain various tools that assist researchers to develop theoretical ideas and test a hypothesis (Kelle, 1995). Another example in some programs is a model-building facility that uses diagrams linked to the main database of codes, text, or images (e.g. Atlas.ti, Qualrus, and NVivo). Second, some programs enable the exchange of data and analyses between collaborating researchers. Some do this by the export and import of data and analyses in a common format such as XML (e.g., Tatoe and Atlas.ti). Others are able to merge project databases from different researchers (e.g., NVivo, Atlas.ti, and NUD*IST/N6).

...