Advanced Workshop on Spatial Analysis in Social Research
May 17-20 , 2001
Ann Arbor, MI

An Advanced Workshop on Spatial Analysis in Social Research will be held in Ann Arbor, Michigan, May 17-20, 2001. The workshop is jointly sponsored by the Interuniversity Consortium on Political and Social Research (ICPSR) and the NSF-Funded Center for Spatially Integrated Social Science (CSISS). The workshop is organized by Luc Anselin (CSISS and University of Illinois, Urbana-Champaign) and Hank Heitowit, Director of the ICPSR Summer Program in Quantitative Methods. The workshop will last slightly more than two days, starting with a social event on Thursday evening and ending with a wrap-up session on Sunday morning.

Meeting Goals
The objective of this workshop is to establish a dialogue between leading methodologists in spatial analysis and in the mainstream social sciences, in order to:

• facilitate the dissemination of state of the art spatial analytical techniques to the methodology in political and social research;
  
  
• assess the importance of spatial analysis in general, and spatial data analysis in particular to social science methodological questions;
  
  
• promote the application of state of the art spatial analytical techniques to substantive research questions in political science and sociology and/or to important social science data sets.

In recent years, the attention paid to spatial aspects of analysis in the social sciences has grown dramatically, both from a theoretical as well as from an applied empirical perspective. In the theoretical debates in political science, sociology, demography, criminology and other social science fields, notions such as "social interaction", "social capital", "diffusion and contagion", "spatial mismatch" and similar concepts contain an explicit spatial imprint and require an explicit modeling of two-way or multi-way interaction over space. This suggests that interdependence (or spatial autocorrelation) is the rule, rather than the traditional paradigm of independence. Similarly, the rapid advances in the technology of spatial data handling, such as geographic information systems, global positioning systems, and the like as well as the explosion in available geo-coded data has created a great demand for sophisticated methods to interpret and analyze these data. Such methods, while common in geographical analysis and the natural sciences, have yet to become part of the standard technical toolbox of the social scientist. In part to address this issue, the National Science Foundation established the Center for Spatially Integrated Social Science (CSISS) as one of its infrastructure projects under the 1999 round of funding.

The workshop will bring together a select group of 20-30 leading scholars from three main constituencies: spatial methodologists (scholars dealing with GIS, geographic data visualization, exploratory spatial data analysis, geostatistical analysis, and spatial regression analysis), social scientists who have begun to apply or are considering the application of spatial analysis in their work (criminologists, demographers, political scientists), and social science methodologists (multivariate methods, survey design, latent variable models, time series, etc.).

The workshop is envisaged as a two-day event, consisting of intensive discussion in plenary sessions, with participation by all, followed by a wrap up session on the morning of the third day. Each day will be distinct in purpose and format. During the first day, a series of 45-60 minute presentations by leading spatial analysis scholars would provide a survey of the main concepts, methodological issues and techniques. These lectures would be aimed at jump-starting those social scientists who are less familiar with spatial analysis and provide a common vocabulary for the discussion in the second day. It is envisaged that the following topics would be covered:

• GIS data models, spatial analysis from a GIS perspective;
• spatial data visualization and exploration;
• geostatistical analysis;
• classical spatial data analysis and spatial regression analysis;
• Bayesian perspectives.

The lectures will be complemented with a series of selected readings that would be made available to participants ahead of time. Also, day 1 will conclude with a series of software demonstrations.

The second part of the workshop would consist of a discussion of the current state of the art of "spatial analysis" in specific social science fields based on brief presentations by the participants. The main goal of this discussion is to outline an agenda for high priority methodological issues that need to be resolved. A second aspect of the discussion is to explore the potential of applying spatial analytical techniques to a number of demonstration projects, dealing with substantive research questions or with applications of a spatial perspective to the analysis of large social science data sets (managed by ICPSR).

The final wrap up will consist of a summary and discussion of future activities and will conclude by noon.

To the extent practically possible, all materials from the workshop (lecture overheads, supporting materials, demos, etc.) will be made available at this website.

It is the intention to follow up the workshop this May with a second meeting about a year later at which the results of the substantive analyses would be reported. An edited volume is planned, to be targeted at a leading university press, that would contain both the pedagogic pieces on spatial analytical techniques as well as the description of the implementation and results of the substantive "demonstration" projects.

Travel and Accommodations
ICPSR will cover all participant costs associated with travel, lodging and meals. A set of rooms has been set aside from Thursday through Sunday nights, May 17-19 at the Bell Tower Hotel in Ann Arbor.

Note: Participation in this Workshop is by invitation only.

Organizers

• Luc Anselin, CSISS, University of Illinois at Urbana-Champaign
  
  
  
• Hank Heitowit, ICPSR, University of Michigan