Geoscience Reference
In-Depth Information
Mixing methods
Any serious decentering of 'the fi eld' has the effect, of course, of further softening the
division between ethnographic knowledge and other forms of representation fl owing
out of archival research, the analysis of public discourse, interviewing, journalism,
fi ction, or statistical representations of collectivities. (Gupta and Ferguson, 1997,
p. 38)
The emergence of quantitative methods in the 1960s undermined ethnographic
research and related qualitative methods and associated them with purely descrip-
tive and, hence, unscientifi c work. Today, the scientifi c authority of qualitative
methods has been re-established and they are being widely adopted. They are,
however, increasingly seen as not single methods (e.g., participant observation) but
as part of a suite of methods (qualitative and quantitative), any of which might be
used on a given project. The combination of methods is made possible by not only
the broadening of ethnographic and other qualitative approaches across disciplines
but by the re-thinking of quantitative methods and, even, GIS as tools for post-
positivist research (Lawson, 1995; Sheppard, 2005; Pavlovskaya and St. Martin,
2007).
Methods once seen as epistemologically incompatible are being successfully com-
bined within 'mixed method' research paradigms that often include ethnography
(Creswell, 2003). The success of such approaches is due to their ability to produce
knowledge that otherwise would not be possible to create. This is particularly
important in light of the simultaneous expansion of secondary data, mainly in digital
form (e.g., remotely sensed data, census information, consumer databases, etc.), and
its growing prominence in various types of analysis, including environmental
policy.
Environmental geography is well suited for mixed-methods approaches. This
is clearly demonstrated by political ecologists who are combining, for example,
geomatics techniques with ethnographic methods (cf. Turner and Taylor, 2003).
Hong Jiang (2003) argues for the integration of satellite imagery analysis with eth-
nographic accounts of landscape change. Combining these methods produced
insights into environmental and cultural change in Inner Mongolia that would not
have been revealed by either method alone. Paul Robbins' research (2003; see also
Robbins and Maddock, 2000) interrogates professional foresters' and villagers'
concepts of 'forest' in India. Using remotely sensed images as well as in-depth
interviews, his research not only reveals but explains the dissonance between both
groups' categorisation of forests.
Conclusion
The current popularity of ethnography in human geography is a result of the
renewed attention to human subjectivity characteristic of many realms of human
geography including, recently, environmental geography. Where in the past, the
power of ethnography existed in its ability to comprehensively describe and thereby
appropriate other peoples and resources, its strengths today suggest a number of
ways that it can inform an environmental geography that is itself moving beyond
the instrumental analysis of environmental impacts. These include ethnography's
abilities to theoretically (rather than statistically) explain social and environmental
phenomena, to explicate 'on the ground' just how environments and environmental
