9
was criticized by later authors who argued that it still set the terms of the discussion
according to positivist criteria. Lincoln herself later accepted this critique of parallelism
and offered reconceptualized criteria (1995) in lieu of the earlier ones.
<<Table 1 here>>
Table 1 also shows the terms used by M&H in their discussion of “standards for
quality.” They use the parallel construction of L&G, listing both the positivist terms and
L&G’s interpretive terms. But they also add a third term for three of the four original
criteria. For example, they add the term “authenticity” as equivalent to “internal validity”
and “credibility” in response to interpretive literature coming after the L&G text, for
example, Maxwell (1992). Perhaps sensitive to the emerging criticism of L&G’s
approach, they add an additional set of terms as well—“Utilization/Application/Action”
—arguing that this criterion set is “an essential addition to more traditional views of
‘goodness’” (1994, 280).
Another part of the L&G approach in their 1985 text has proved influential,
perhaps more so because it has provoked less critical discussion. After developing their
parallel criteria, L&G went on to suggest specific “techniques” that would enable a
researcher to achieve each new criterion. So, for example, to demonstrate an interpretive
study’s “credibility” (the term they propose for “internal validity”), they suggest
“prolonged engagement” in the field and “member checks,” among others. Similarly,
recommended techniques for demonstrating an interpretive study’s “confirmability” (i.e.,
objectivity) include a “confirmability audit” and a “reflexive journal.” Here, too, L&G
adopted an analytic practice oriented toward positivist research. For example, in
quantitative research the criterion of “reliability” can be assessed using “test-retest” or
Convention