Theoretical Saturation

Illustration of Theoretical Saturation

What follows is a necessarily simplified illustration of theoretical saturation intended to clarify its defining features. A researcher conducting a study to understand how HIV-positive women handle the stigmatization associated with HIV infection begins to theorize, from the interview data collected from HIV-positive women, two types of disclosure: managed and mismanaged disclosure. In managed disclosure, women stay in control of whether their HIV status will be revealed and what about it will be revealed and to whom. The researcher discerns what she or he initially takes to be four analytically distinct strategies women use in managed disclosure: (a) full disclosure, whereby women withhold nothing about their disease from anyone; (b) partial disclosure, whereby they reveal only some things about their disease; (c) selective disclosure, whereby they reveal information about their disease to some but not to other people; and (d) full concealment, whereby they reveal nothing about their disease to anyone. In mismanaged disclosure, women lose control of whether their HIV status is revealed, what is revealed, and to whom. These women are “outed” accidentally, as when they are seen entering a clinic serving only HIV-positive patients, or deliberately, as when a person to whom a woman has revealed her disease subsequently tells other people.

Theoretical saturation will have been achieved when this researcher—using theoretical sampling and constant comparison analysis—is able to answer a number of questions, only a few of which are featured here. For example, do the concepts managed and mismanaged disclosure, and full disclosure and concealment, and partial and selective disclosure, exhaust the variation in types and strategies of disclosure? Should the categorization of disclosure be refined to encompass additional types and strategies or to eliminate one or more of them? No matter the number of categories, are they both exhaustive and mutually exclusive? If the researcher decides to use a conditional matrix framework for analysis—one of a number of grounded theory coding families—does she or he have the data to describe the causes and conditions for, and consequences of, using these disclosure strategies? For example, do certain HIV-positive women (e. g., White versus African American, women in general versus just mothers, women diagnosed with HIV infection for more versus less than a specified period of time) prefer one strategy over another? Under what circumstances would a woman decide to disclose or conceal fully, partially, or selectively? Alternatively, if the researcher decided a process coding framework was a better fit to the data, can the [Page 876]researcher show how women moved from one strategy to another, or cycled between strategies?

...