Материал: Seale - Quality in Qualitative Research

466 QUALITATIVE INQUIRY / December 1999

is that practicing researchers now have to find their way through a mass of conflicting positions, and methodology is in danger of getting a bad name. I want to suggest a way through, following in the spirit of an earlier era, when the call to produce grounded theory empowered researchers concerned to reclaim their craft from the “theoretical capitalist(s)” of the day (Glaser & Strauss, 1967, p. 10). The appeal to scientific foundations made by Glaser and Strauss may no longer be as easily sustainable, but the need for a new conceptualization of the relationship between qualitative social research, theory, and indeed philosophy is as pressing now as it was in the 1960s.

Philosophy is often presented as underpinning the craft of social research, being an arena where various attempts at providing foundations for judging truth claims have come and gone, yet present day opinion seems nowadays, paradoxically, to conclude on antifoundationalism as itself being a philosophical foundation for social research. I think it is time for social researchers to exploit this paradox, by breaking free from the obligation to fulfill philosophical schemes through research practice, while remaining aware of the value of philosophical and political reflexivity for their craft. A confident view of social research as a craft skill could then emerge, relatively autonomous from social theory or philosophy, yet drawing on these arenas of discourse as a resource. The search for overarching criteria for judging quality under this vision is thereby held at a distance, and the elusive nature of quality (we somehow recognize it when we see it, but we cannot prespecify it with methodological rules) is preserved. This also means that we do not have to abandon skills developed under one paradigm because another paradigm has come along.

It is worth summarizing some key shifts in criteriology, which is a kind of offshoot of broader debates in the philosophy of science, to show how I have reached this conclusion. Then I shall make some statements about philosophical positions and foundationalism, after which I shall use the example of triangulation to demonstrate how social researchers might operate within a conception of research as a craft skill in a way that preserves a commitment to producing good quality research.

CRITERIOLOGY

Qualitative creation mythology, in the modernist phase of qualitative inquiry, emphasized difference by making overdrawn contrasts with the supposed “positivism” of quantitative work. In the methodological debate about quality criteria, this, initially, involved substituting new terms for words such as validity and reliability to reflect interpretivist conceptions while retaining a sense that social researchers in both traditions shared similar scientific orientations. A typical example here is LeCompte and Goetz (1982) who argued that techniques for establishing validity and reliability should be

Seale / QUALITATIVE RESEARCH 467

somewhat different in, say, an ethnography compared with an experiment. They drew up a scheme intended as a qualitative parallel to Campbell and Stanley’s (1966) influential account of validity and reliability in quasiexperimental designs, inventing new concepts such as internal and external reliability to legitimate a degree of difference that nevertheless allowed them to advocate pragmatic combinations of quantitative and qualitative work.

Qualitative criteriologists since LeCompte and Goetz progressively moved away from modernist commitments, leading to conceptions of validity and reliability very far removed from positivist or realist perspectives, as constructivism shifted into postmodernism. The urge to generate criteria for judging good quality studies seems irrepressible in these methodological writings, perhaps in part due to the requirement that social researchers impress the worth of their efforts on skeptical audiences, such as researchfunding bodies.

A sometimes bewildering variety of new concepts arose. For example, Altheide and Johnson’s (1994) review of interpretivist positions on validity identifies “successor validity, catalytic validity, interrogated validity, transgressive validity, imperial validity, simulacra/ironic validity, situated validity, and voluptuous validity” (p. 488). A glance at Kirk and Miller (1986), though, shows Altheide and Johnson to have omitted from this list “apparent,” “instrumental,” and “theoretical” validity. Additionally, Kirk and Miller (1986) demonstrate the ease with which new forms of reliability can be conceptualized, dividing this into the “quixotic,” the “diachronic,” and the “synchronic.” This proliferation of concepts reflects the difficulties that qualitative methodologists, committed to creating some overarching system for specifying quality, have had in making their ideas stick. This is in marked contrast to parallel authors in the quantitative tradition where a consensus around certain ideas (for example, the distinction between validity and reliability, or between internal and external validity) has been more easy to sustain. In qualitative research, the project of criteriology experiences particular contradictions because of the difficulty in regulating and constraining an endeavor whose guiding philosophy often stresses creativity, exploration, conceptual flexibility, and a freedom of spirit. Additionally, though, conceptual proliferation is a marker of the paradigm shifts and crises of legitimation and representation, which have characterized “moments” in the recent history of qualitative methods (Denzin & Lincoln, 1994).

The work of Lincoln and Guba reflects these more recent shifts. Lincoln and Guba (1985) argue that establishing the trustworthiness of a research report lies at the heart of issues conventionally discussed as validity and reliability so that four questions have, from within the modernist paradigm, been asked of research reports, namely their truth value, applicability, consistency, and neutrality. Truth value, though, assumes a “single tangible reality that an investigation is intended to unearth and display” (Lincoln & Guba, 1985, p. 294), whereas the naturalistic researcher makes “the assumption of multiple

468 QUALITATIVE INQUIRY / December 1999

constructed realities” (p. 295). Applicability depends on generalizing from a sample to a population, on the untested assumption that the “receiving” population is similar to that of the “sending” sample; the naturalistic inquirer, on the other hand, would claim the potential uniqueness of every local context, requiring empirical study of both sending and receiving contexts for applicability to be established. They are similarly critical of the other two conventional criteria. Consistency, they say, depends on naïve realist assumptions; neutrality depends on an artificial separation of values from inquiry. They criticize LeCompte and Goetz, whom they identify with such questions, for their dependence on axioms such as “naive realism and linear causality” (Lincoln & Guba, 1985, p. 293).

Instead, Lincoln and Guba (1985) propose their own four-point criterion list for naturalistic inquirers. A concern with credibility should replace truth value and “the most crucial technique for establishing credibility” is through “member checks” (Lincoln & Guba, 1985, p. 314). Transferability should replace applicability, or external validity as conventionally conceived. Dependability is proposed as a replacement for consistency, or reliability as conventionally conceived, to be fulfilled by peer auditing procedures. Auditing is also useful in establishing confirmability, a criterion designed to replace the conventional criterion of neutrality or objectivity. Auditing is an exercise in reflexivity, which involves the provision of a methodologically self-critical account of how the research was done. Trustworthiness is always negotiable and open-ended, not being a matter of final proof whereby readers are compelled to accept an account. This, Lincoln and Guba (1985) claim, “stands in marked contrast to that of conventional inquiry” (p. 329), which claims to be “utterly unassailable” once relevant procedures have been carried out. This is, in fact, a rather overdrawn contrast if we consider the fallibilistic spirit of Campbell and Stanley’s (1966) account of threats to validity and reliability.

The criteria offered by Lincoln and Guba in 1985, however, depend on a contradictory philosophical position, because the belief in “multiple constructed realities,” rather than a “single tangible reality” (Lincoln & Guba, 1985, p. 295), which lies at the heart of the constructivist paradigm, is not consistent with the idea that criteria for judging the trustworthiness of an account are possible. Relativism does not sit well with attempts to establish “truth,” even if the term is placed in inverted commas.

Acknowledging this problem, then, in later work (Guba & Lincoln, 1989, 1994), a fifth criterion, “authenticity,” is proposed as being consistent with the relativist view that research accounts do no more than represent a sophisticated but temporary consensus of views about what is to be considered true. In detailing the components of authenticity, Guba and Lincoln (1989, 1994) reveal a sympathy for political conceptions of the role of research that was already evident in their earlier commitment to the value of member checking. Authenticity, they say, is demonstrated if researchers can show that they

Seale / QUALITATIVE RESEARCH 469

have represented a range of different realities (“fairness”). Research should also help members develop “more sophisticated” understandings of the phenomenon being studied (“ontological authenticity”), be shown to have helped members appreciate the viewpoints of people other than themselves (“educative authenticity”), to have stimulated some form of action (“catalytic authenticity”), and to have empowered members to act (“tactical authenticity”). Of course, the view that fairness, sophistication, mutual understanding, and empowerment are generally desirable is itself a value-laden, culturebound position that a Foucauldian deconstructionist might very well enjoy taking apart. It represents an attempt to pull back from the relativist abyss, but the substitution of political goals as foundations for research is problematic in a world where there is no fixed consensus on the desirability of particular goals (Hammersley, 1995). A softer version of political commitment might involve researchers simply offering readers a reflexive account of their politics and leaving it to the democratic process in wider society to resolve clashes of interest. Nevertheless, it is hard not to agree with Guba and Lincoln’s (1994) following conclusion: “The issue of quality criteria in constructivism is . . . not well resolved, and further critique is needed” (p. 114).

These authors, then, along with many others in the qualitative social research community, have traveled on a path beginning with a rejection of positivist criteria and the substitution of interpretivist alternatives. Dissatisfied with the limitations of these, constructivism has been embraced, introducing an element of relativism. Political versions of the value of research have then been imported to save facing the logical implications of relativism, which seem to threaten a nihilistic vision and abandonment of the research enterprise. What is a practicing social researcher to make of all this? How can these inconclusive debates become a resource for researchers rather than a source of frustration and negativity? Before turning to this, I want to consider some more purely philosophical concerns.

PHILOSOPHICAL MOMENTS

At a philosophical level, discussions of the problem of foundations for knowledge are also inconclusive. Nevertheless, philosophical positions are sometimes claimed as being in another sense “foundational” for research practice by criteriologists. Thus, empiricism is claimed as foundational for modernist paradigms, antiempiricism is proposed as foundational for constructivist and postmodern views, and so on. I propose here to examine just one attempt to provide foundations for research practice in a philosophical position, the “subtle realist” conclusion of Hammersley (1992), together with some criticisms of this. I do not propose subtle realism as a solution (though it has attractive qualities as a pragmatic compromise between several