Trustworthiness Criteria

Trustworthiness criteria provide yardsticks by which the rigor, VALIDITY, systematic nature of methods employed, and findings of qualitative studies may be judged. It is the name given to criteria appropriate to emergent, interpretive, nonpositivist, nonexperimental models of inquiry that frequently are labeled constructivist, critical theorist, action inquiry, phenomenological CASE STUDIES, or naturalistic paradigms. These criteria were the first to be proposed (Guba, 1981; Lincoln & Guba, 1985) in response to serious queries regarding the soundness, RELIABILITY, and rigor of qualitative research done under new-paradigm models.

Conventional (positivist) criteria of rigor are well established and understood: INTERNAL VALIDITY, EXTERNAL VALIDITY, reliability (replicability), and OBJECTIVITY. These four rigor domains respond, respectively, to an isomorphism to reality, to generalizability, to the stability of findings (and therefore, the ability of any given study to be replicated), and to freedom from investigator bias. Phenomenological, or new-paradigm, inquiry, however, specifically rejects each of those four domains as unachievable (in either a practical or a philosophical sense) or inappropriate (in a paradigmatic sense). It has proposed in their stead four criteria that are more responsive to both the axiomatic requirements of phenomenological inquiry and the requirements of methodologies and design strategies often more appropriate to such research, evaluation, or policy analyses, which are frequently, although not always, qualitative in nature.

Early efforts to create standards for judging nonquantitative and emergent-paradigm inquiries elicited the trustworthiness criteria: credibility, transferability, dependability, and confirmability. These criteria, deemed “trustworthiness criteria” to signal that attention to them increased confidence in the rigorousness of findings, were devised as parallel to the four criteria of conventional research. These criteria, designed as parallel to criteria for experimental models, have been criticized as “foundational,” that is, tied to a foundation emanating from conventional inquiry rather [Page 1145]than from characteristics inherent in phenomenological (constructivist) inquiry. Like conventional criteria for rigor, these criteria reference primarily methodological concerns; because of that, they are still highly useful when examining the methodological adequacy of case studies.

Credibility refers to the plausibility of an account (a case study): Is it believable? Is the report deemed an accurate statement by those whose LIVED EXPERIENCE is reported? Transferability refers to the report's ability to be utilized in (transferred to) another setting similar to that in which the original case was conducted. Judgments about transferability are unlike those about GENERALIZABILITY, however, in one important aspect. In conventional research, generalizability estimates are asserted by the researchers. In assessing transferability, however, judgments about the usefulness of results are made by “receiving contexts” (Guba & Lincoln, 1989; Lincoln & Guba, 1985), such as users and consumers of the research findings, and individuals who wish to utilize the findings in their own situations. Dependability references the stability, trackability, and logic of the research process employed. Dependability is assessed through a process audit, which analyzes the methodological decisions made and certifies their soundness. Confirmability certifies that the data reported in the case can be pursued all the way back to original data sources (e.g., FIELDNOTES, documents, records, observation logs, etc.).

Trustworthiness criteria are primarily methodological criteria; they are responsive to method (although some of the suggested methods—Guba & Lincoln, 1989—are directly tied to phenomenological premises, e.g., member checking to ascertain whether the researcher has understood the respondent's social constructions), rather than directly tied to paradigmatic assumptions. However, because new-paradigm inquiry often demands nonconventional, qualitative methods, and because trustworthiness criteria are tied heavily to the use of such methods, they continue to have applicability in judging the rigor (for instance, the level of engagement and persistence in the content) of an essential element—method—of emergent-paradigm inquiry.

...