Comp. by: VPugazhenthi Stage : Revises1 ChapterID: 9781405132879_4_M Date:1/4/
09 Time:15:19:38 Filepath:H:/00_Blackwell/00_3B2/Gregory-9781405132879/appln/3B2/
revises/9781405132879_4_M.3d
inspired by thesciencesand is often a hall-
mark of social science perspectives. This
recounting of procedures offers a methodo-
logical transparency that is the hallmark of
scientific studies in that it allows readers the
chance to ‘disprove’ a study’s conclusions.
Even beyond formally scientific studies, trans-
parency is often still advocated as a way of
ensuring rigour and validity across a variety
of approaches. Baxter and Eyles (1997) sug-
gest adapting ideas from Lincoln and Guba so
that for all methods, includingqualitative
methods, the following have to be shown: credibility of the account (i.e. authenti-
cated representation of what actually
occurred);
transferability of the material (i.e. making
what occurred intelligible to the audience);
dependability of the interpretation (i.e.
that it is not illogical, or how partisan it
is); and
confirmability of the study (i.e. the ability
to audit the process that made it through
personal reflection, audit processes or
opportunities for informants to reply).The emphasis here is on the clear statement
of procedures undertaken, the techniques and
steps of analysis as a means to enable the rea-
der to examine the process leading to the
results and conclusions. Traditions of transpa-
rency are variable across even sub-disciplinary
fields where, for instance, in a comparison of
the Journal of International Business Studies
andEconomic Geography, around 80 per cent
of articles in the former had formal sections
on methodology and the collection of data,
while in the latter only 30 per cent or less
had such sections (Poon, 2007). A degree
of transparency might also be intended to
enable a reflexive account that positions
the research process, and allows the reader to
see the contingent and situated production
of knowledge. This may in fact be aimed
at undercutting notions of authoritative
social science by demonstrating fallibility and
the limits to knowledge, and may be inspired
by approaches to researchethicsthat fore-
ground the contributions of research partici-
pants. Thusfeministmethodology stretches
from design to dissemination of research,
but must also consider the ‘relationships
among people involved in the research pro-
cess, the actual conduct of the research, and
process through which the research comes
to be undertaken and completed’ (Moss,
2002, p. 12).A stronger definition of methodology points
out the different ways in which these criteria
of reliability reflect theoretical approaches.
For example, sampling within extensive
research would have to meet criteria of
representativeness to support statistical ana-
lysis of the data, while within qualitative
methodsit might aim to capture a particular
group’s perspective, where its validity depends
on the quality of material derived from their
positionality. In this case, methodology is
about joining the stages in the research from
underlyingphilosophyto research questions
to techniques generating data to forms of an-
alysis and presentation of the results.
A methodology thus involves considering
how the specific techniques can be assembled
and used to generate the sort of data that will
enable an answer to the questions posed
through a specific conceptual framework.
There have been extensive arguments about
whether specific conceptual frameworks dem-
and specific methodological linkages and pre-
clude some methods. For instance,feminist
geographyhad a long debate over whether
it required feminist methods, that sought to
empower, give voice to women and treat
them as ‘subjects’ who made knowledge, or
whether it could use the quantitative
methods that feminist theorists had often
critiqued for treating people as objects of
knowledge and using a detachedmasculinism
in its logics. This debate has seen special issues
of journals, such as theProfessional Geographer
in 1995, assessing whether a feminist method-
ology can include various techniques such as
gis.
If methodology is about assessing how to
creatematerialthatwill answer the conceptual
questions in a study, the answer may well be to
use multiple methods. Thus some parts or
issues might be addressed via one method
while others could be addressed by another.
This is often called triangulation, named
metaphorically after thesurveyingpractice
of taking bearings from different landmarks.
Here, methodology is about combining
methods to help validate each others’ findings
and, optimistically, integrating different forms
of data in the analysis (Knigge and Cope,
2006). On the other hand, it might be argued
that the different techniques, with different
ontologies, construct radically different ver-
sions of the world that cannot be brought into
the same epistemologicalapproach. How
can, say, asocial constructivistaccount be
allied to modes of statistical inference that rely
onrealismor assumptions ofobjectivity?InGregory / The Dictionary of Human Geography 9781405132879_4_M Final Proof page 458 1.4.2009 3:19pmMETHODOLOGY