Spatial Data Analysis
Introduction
This course materials are designed for a 1-day workshop on spatial data analysis. Rüttenauer (2024) provides a handbook chapter accompanying these workshop materials.
In recent years, more and more spatial data has become available, providing the possibility to combine otherwise unrelated data, such as social, economic, and environmental data. This also opens up the possibility of analysing spatial patterns and processes (e.g., spillover effects or diffusion).
Many social science research questions are spatially dependent such as voting outcomes, housing prices, labour markets, protest behaviour, or migration decisions. Observing an event in one region or neighbourhood increases the likelihood that we observe similar processes in proximate areas. As Tobler’s first law of geography puts it: “Everything is related to everything else, but near things are more related than distant things”. This dependence can stem from spatial contagion, spatial spillovers, or common confounders. Therefore, basic assumptions of standard regression models are violated when analysing spatial data. However, more importantly, spatial processes are interesting for their own sake. Spatial regression models can detect spatial dependence and explicitly model spatial relations, identifying spatial clustering, spillovers or diffusion processes.
The main objective of the course is the theoretical understanding and practical application of spatial regression models. This course will first give an overview on how to perform common spatial operations using spatial information, such as aggregating spatial units, calculating distances, merging spatial data as well as visualizing them. The course will further focus on the analysis of geographic data and the application of spatial regression techniques to model and analyze spatial processes, and furthermore, the course addresses several methods for defining spatial relationships, detecting and diagnosing spatial dependence and autocorrelation. Finally, we will discuss various spatial regression techniques to model processes, clarify the assumptions of these models, and show how they differ in their applications and interpretations.
The field has developed very quickly over the past few years, and R now provides a rich set of packages for various spatial data operations. For a more in-depth introduction into spatial data analysis in R, have a look into the materials references below.
The material introduces the use of geographical information to connect and analyze different spatial data sources very briefly. This introduction is limited to the fundamentals of using geographical information in R. Stefan Jünger & Anne-Kathrin Stroppe have provided a comprehensive GESIS workshop on geospatial techniques in R. The focus of this workshop will be on techniques for spatial data analysis, such as spatial regression models.
Schedule
| Time | Session |
|---|---|
| 09:00 – 09:30 | Refresher on R for spatial data |
| 09:30 – 11:00 | Spatial Relationships (W) and Spatial Dependence |
| 11:15 – 12:00 | Practical exercise |
| Lunch break | |
| 13:00 – 14:30 | Spatial Regression Models (SLX, Error, lagged DV) |
| 14:45 – 16:00 | Interpreting Results: Spatial Impacts |
| 16:15 – 18:00 | Practical exercise |
Some useful packages
By now, R provides a lot of functionalities for GIS applications and spatial econometrics, and further extensions. There are lots of packages providing a huge variety of spatial functionalities and methods (see e.g. R. Bivand, Millo, and Piras 2021). Important packages for fundamental spatial operations are:
Spatial data workhorses: sf (Pebesma 2018) and terra
Visualization: mapview (Appelhans et al. 2021) and tmap (Tennekes 2018)
Spatial weights and other relations: spdep (R. S. Bivand and Rudel 2018)
Spatial interpolation and kriging: gstat (Gräler, Pebesma, and Heuvelink 2016)
Spatial regression models: spatialreg and sphet (R. Bivand and Piras 2015)
The packages have constantly developed over the past years, and older packages such as rgdal, rgeos, and sp are currently retiring (Blog post)
Further Readings
Great up-to-date introduction to spatial R: Lovelace, Nowosad, and Muenchow (2019), updated version available online
Great open-science book on Spatial Data Science Pebesma and Bivand (2023)
Comprehensive introduction to spatial econometrics: LeSage and Pace (2009)
Relative intuitive introduction to spatial econometrics: Ward and Gleditsch (2008)
Article-length introductions to spatial econometrics: Elhorst (2012), Halleck Vega and Elhorst (2015), LeSage (2014), Rüttenauer (2024), and Rüttenauer (2022)
Course materials
I highly recommend the great Introduction to Geospatial Techniques for Social Scientists in R including, see Stefan Jünger & Anne-Kathrin Stroppe’s GESIS workshop materials. Nice materials on GIS, spatial operations and spatial data visualisation!
For those looking for a more in-depth introduction, I highly recommend Roger Bivand’s course on Spatial Data Analysis: Youtube recordings, Course Materials
I’ve learned most of what I know about spatial econometrics from Scott J. Cook and his workshop on Spatial Econometrics at the Essex Summer school.