Spatial Analysis

Development operations and research are inherently spatial. Development outcomes vary greatly in different parts of the world and even within countries due to regional differences in natural endowments or public expenditures. Investments in school construction, infrastructure or irrigation, for example, tend to benefit specific regions. An important development objective is to reduce spatial inequality between regions by providing adequate public services and by creating economic opportunities. Projects therefore often use geographic targeting to prioritize placement of interventions.

To investigate these and related issues, researchers and policy analysts use tools and analytical techniques that can be summarized as spatial analysis. These approaches rely on evolving technologies such as geographic information systems (GIS), remote sensing and global positioning systems (GPS) to collect, integrate and visualize geographically referenced information. Analytic techniques to interpret such information range from simple mapping to spatial econometrics and spatial process modeling.

The pages on this site present examples of analytical applications in which the use of spatially referenced information helped provide new insights. The examples are drawn from Bank research and operational projects across sectors and regions.

Spatial analysis can benefit project design and implementation as well as analytical work in many different ways.

• Simple mapping helps explore patterns of information collected in the field or socioeconomic data compiled by statistical offices.
• Remotely sensed data such as satellite images or air photos provide a geographic context to the evaluation of project sites.
• Data integration by overlaying different types of spatially referenced information can create new indicators and variables for visualization and analysis. Geographic analysis enables one to use ‘space as an indexing system’ to link heterogeneous information sources.
• Spatial process models aid in the analysis of transport networks, migration or disease distribution.
• Incorporating the spatial pattern of data points in statistical or econometric analysis corrects for possible estimation bias and can provide insights beyond typical cross-sectional or time series results.