Computer-assisted qualitative data analysis software (CAQDAS) Contemporary qualitative data analyses can be supported by computer programs (termed
computer-assisted qualitative data analysis software). These programs have been employed with
or without detailed hand coding or labeling. Such programs do not supplant the interpretive nature of coding. The programs are aimed at enhancing analysts' efficiency at applying, retrieving, and storing the codes generated from reading the data. Many programs enhance efficiency in editing and revising codes, which allow for more effective work sharing, peer review, data examination, and analysis of large datasets. Sometimes researchers rely on computers and their software to scan and reduce large amounts of qualitative data. At their most basic level, numerical coding schemes rely on counting words and phrases within a dataset; other techniques involve the analysis of phrases and exchanges in analyses of conversations. A computerized approach to data analysis can be used to aid content analysis, especially when there is a large corpus to unpack.
Trustworthiness A central issue in qualitative research is trustworthiness (also known as credibility or, in quantitative studies, validity). There are many ways of establishing trustworthiness, including
member check, interviewer corroboration, peer debriefing, prolonged engagement, negative case analysis, auditability, confirmability, bracketing, and balance. Transferability of results has also been considered as an indicator of validity.
Limitations of qualitative research Qualitative research is not without limitations. These limitations include participant reactivity, the potential for a qualitative investigator to over-identify with one or more study participants, "the impracticality of the
Glaser-Strauss idea that hypotheses arise from data unsullied by prior expectations," the inadequacy of qualitative research for testing cause-effect hypotheses, and the
Baconian character of qualitative research.
Participant reactivity refers to the fact that people often behave differently when they know they are being observed. Over-identifying with participants refers to a sympathetic investigator studying a group of people and ascribing, more than is warranted, a virtue or some other characteristic to one or more participants. Compared to qualitative research,
experimental research and certain types of nonexperimental research (e.g.,
prospective studies), although not perfect, are better means for drawing cause-effect conclusions. Glaser and Strauss, influential members of the qualitative research community, pioneered the idea that theoretically important categories and hypotheses can emerge "naturally" from the observations a qualitative researcher collects, provided that the researcher is not guided by preconceptions. The ethologist
David Katz wrote "a hungry animal divides the environment into edible and inedible things....Generally speaking, objects change...according to the needs of the animal."
Karl Popper carrying forward Katz's point wrote that "objects can be classified and can become similar or dissimilar, only in this way—by being related to needs and interests. This rule applied not only to animals but also to scientists." Popper made clear that observation is always selective, based on past research and the investigators' goals and motives and that preconceptionless research is impossible. The Baconian character of qualitative research refers to the idea that a qualitative researcher can collect enough observations such that categories and hypotheses will emerge from the data. Glaser and Strauss developed the idea of theoretical sampling by way of collecting observations until
theoretical saturation is obtained and no additional observations are required to understand the character of the individuals under study. ==In psychology==