The recent dissemination of a spatial analytic perspective in the social sciences (outside of the discipline of geography) is often attributed to the rapid spread of GIS technology to the desktop and the availability of a vast array of geographically referenced socio-economic data. This has led to the use of GIS for data organization and visualization as well as increasingly in an inductive approach to exploring data for meaningful patterns and structures (exploratory spatial data analysis). While these have undeniably been important factors, an equally crucial aspect has been the need to operationalize “new” theoretical constructs that explicitly incorporate space in the analysis of human (economic) behavior. Many of these concepts are similar (though not always acknowledged) to the models proposed by economic geographers and regional scientists in the 1960s, and stress the importance of location, neighborhood, region and spatial (social) interaction. Current examples in economics are the emphasis on spatial externalities and regional clusters (e.g., Krugman, Arthur, Porter), theories of interacting agents and interdependent decision making (e.g., Pollak, Ioannides, Durlauf, Brock, Brueckner), the importance of social interaction and group effects (e.g., Akerlof, Aoki) and neighborhood effects (Borjas). Similar examples can be cited in recent work in other social sciences, such as sociology, political science and criminology. Unlike their antecedents of the 1960s, description and discussion of these theories appears in the “core” journals of the mainstream disciplines, such as the American Economic Review, Journal of Political Economy and Econometrica for economics.
Empirical validation of the new “spatial” concepts and models requires an explicit spatial econometric methodology that tackles issues of spatial dependence and spatial heterogeneity, as well as their extensions in the space-time domain. Spatial econometrics is a subset of spatial statistics in that rather than being “statistics for (any) spatial data”, it concerns itself with statistics for spatial (socio-economic) models, where the model specification is dictated by theory. These subtle differences aside, it is important to acknowledge that a growing number of mainstream econometricians (e.g., Kelejian, Prucha, Bera, Baltagi, Pinkse) have started to contribute to the spatial econometric methodology and that spatial econometrics has gained recognition as a useful subset of the econometric toolbox.
These recent developments in the mainstream social sciences in general and in economics in particular raise a number of challenges for the next generation of “spatial analysis.” Central to this is the need to move “beyond mapping” (generally recognized in the GIS community, but not necessarily in the mainstream disciplines) and to tackle the methodological and theoretical issues that address the complexities of the current models. I see the potential for new developments in three important domains:
- Extending concepts of “space”
Spatial analysis needs to go beyond dealing with physical geographical
locations to include location in “social” space (social distance, economic
distance). This will require further consideration and development
of distance metrics for “social” space, for space-time dynamics and notions
of “topology” in space-time (the counterpart of the “weights” matrix in
spatial autocorrelation analysis). Promising avenues are current
work on GIS data models, object-oriented GIS, and the like.
- Broadening the analytical toolbox
The toolbox of spatial econometrics and spatial analysis needs to be
extended to deal with the challenges posed by the analysis of socio-economic
space-time data. While much progress has been made, some unresolved
issues are the estimation of space-time dynamics for limited dependent
data (such as discrete choice data, duration data), modeling changing choice
sets, distinguishing spatial dependence from spatial heterogeneity, effective
visualization of model fit, etc. For many of these research questions
analytical solutions are impossible or prohibitive, such that computational
approaches must be followed (e.g., simulated moments, simulated likelihood,
Markov Chain Monte Carlo). This requires advances in computational
geography in the form of the development of new and/or efficient algorithms
to tackle the complexity of realistically sized data sets
- Technology transfer
Most of the current commercial GIS software comes in the form of (partially)
open environments that allow the user to include customizations and extend
the functionality. In a modern component oriented computing environment,
there is therefore no longer a high priority to have commercial spatial
analytical tools included in the “box”, but rather to have the mechanisms
to mix and match components to accomplish specific tasks. Since the
commercial world will always be behind the curve when it comes to “state
of the art” in terms of the statistical methodology it delivers, such a
toolbox (such as MapObjects) allows analysts to integrate their own selection
of analytical methods with core GIS functionality. In contrast to
the toolbox approach, shrink-wrapped commercial GIS software has tended
to offer the lowest common denominator when it comes to spatial analytical
(let alone statistical) methodology. In my view, this has had two
major drawbacks. One is that uninitiated users identify “spatial
analysis” with the (limited) set of techniques offered by a software vendor.
The other is that the analysis is presented as being “easy” and underlying
assumptions, algorithms and limitations are hidden from the user.
Both issues pose challenges to software developers as well as to current
GIS education, both in the academic and in the private sector (by vendors).
The theoretical questions posed in the mainstream social sciences offer an important challenge to the methodology of spatial analysis. However, this also constitutes a major opportunity for the spatial analytical perspective (as part of a geographic information science) to contribute to the theoretical debate in the core disciplines.
Email: lanselin@utdallas.edu
http://www.spacestat.com