ACCESSING AND GENERATING DATA 125actual occasions of use” (1989, 7). The sorts of analyses Jordan calls for are interpretive in their
reflexivity: their willingness to subject numerical sources of evidence to critique, rather than
accepting their production at face value. This is the normative heart of McHenry’s argument, and
it harks back to Kirstie McClure’s reading (1999) of the circular legitimation of the science of
statistics: the Royal Society of Statistical Science announced itself as the source of its own legiti-
macy. As T. Mitchell notes in the context of their twentieth-century usage:
Colonial power had long made use of statistics, whether for administrative needs or to
produce a larger “illusion of bureaucratic control.”... But the circulation of statistics among
a “public”... enabled them to take on the form of an “objective culture”... , [leading to]
a divide between two worlds, a sphere of figures, numbers, facts, and trends on the one side,
and the world to which these refer on the other. The latter must stand as its opposite, the
realm of the material, the real. (2002, 103)Dean McHenry calls on us to heed such misplaced concreteness^8 and to interrogate the largely
unspoken and silent assumptions embedded in its products, along with the processes through
which they are produced.
SEPARATING ACCESS FROM ANALYSIS
We wish to underscore what we noted at the beginning of this introduction, that the distinction we
are drawing between methods of accessing or generating data and methods of analyzing them is
useful for analytic and conceptual purposes, but it does not hold for all methods and can be
profoundly misleading in some cases. To take the latter point first: Analysis commences when one
begins to conceive of a research project, to frame one’s research question, read others’ writings on
the subject, and design one’s study. One may be more or less conscious of this—we urge re-
searchers to be more so—which is one of the reasons why “data analysis” as a formal undertaking
typically is listed as a penultimate step, rather than as an initial one. This positioning, too, is a
methodological issue, as it posits that analytic activities are performed on data alone, not on one’s
readings or thought processes. In our view, separating analysis from other research activities is a
narrow and limiting construal of what transpires in crafting research—although we borrow it for
heuristic purposes.
The other point that warrants articulation is that some methods lend themselves more to the
distinction between access and analysis than others. In some it is clear that analysis is a mental
activity performed on data in hand (though one need not necessarily wait until the “end” of a data
accessing phase to do so). Such is the case with formal semiotic analysis, for example, or category
or metaphor analysis, or any other form in which the researcher works primarily (if not exclu-
sively) with word data, whether accessed through documents or through interviewing. In other
modes of research, however, it is far more difficult to separate out analytic processes from those
entailed in accessing and generating data: Temporally, they are much more intertwined. This is
particularly the case for ethnographic, participant-observation, case study, and (participatory)
action research. Practitioners of these methods typically do not see themselves as analyzing their
data in a fashion that is separate and distinct from what they do in accessing them.
We suggest, with all due respect, that this is one of the practices or habits of mind that has led
to a misconstrual that what such researchers do is lacking in rigor. As noted in the introduction to
the book, given that these methods are, at this point in academic time, crossing boundaries into
disciplinary locales in which their presuppositional grounding is not widely known and/or is not