Geospatial Data Science and Spatial Analysis

From mapping neighborhood inequality to modeling how geography shapes health outcomes, the ability to work with spatial data has become an essential skill for researchers, analysts, and practitioners across the social sciences. This course provides a rigorous applied introduction to spatial data analysis, focusing on acquiring, wrangling, visualizing, and analyzing geographically referenced data.

The first part builds core spatial skills: data structures, coordinate reference systems, visualization, and geoprocessing. The second part moves into applied spatial analysis — Census data, spatial autocorrelation, predictive modeling, and spatial regression. The course closes with reproducibility workflows and the ethical dimensions of working with geographic data.

Prereq Foundational R and data wrangling

Tools R · sf · tidycensus · tmap · spatialreg

What you will be able to do

Work with spatial data

Read, write, and transform vector and raster data across coordinate systems

Analyze spatial patterns

Run autocorrelation, clustering, and spatial regression models

Make compelling maps

Design thematic and interactive maps that communicate findings clearly

Work responsibly

Understand geoprivacy, disclosure risk, and spatial justice

Course overview

Part 1 Foundations of spatial data

Getting fluent with how spatial data is structured, stored, and referenced — the building blocks for everything that follows.

Vector vs. raster data; points, lines, and polygons; simple features
Coordinate reference systems, projections, and CRS transformations
Reading and writing shapefiles, GeoJSON, and GeoPackage
Importing Census and administrative data with tidycensus
Attribute joins, filtering, and dplyr in spatial workflows
Exploratory spatial data analysis and summary statistics with geography

sf CRS tidycensus dplyr ESDA

Part 2 Exploration, visualization and geoprocessing

Transforming raw spatial data into maps and insights — from cartographic design to spatial operations and working critically with data sources.

Thematic mapping with tmap and ggplot2; choropleth maps and color theory
Map design principles; interactive mapping with OpenStreetMap
Buffers, spatial joins, intersections, unions, and clipping
MAUP and areal interpolation
Communicating spatial findings to non-specialist audiences

tmap ggplot2 OSM MAUP ACS tigris

Part 3 Spatial analysis and modeling

Detecting patterns, building models, and drawing valid inferences from geographically structured data.

Spatial weights as neighborhood definitions; Moran's I statistic and scatterplot
LISA maps: reading cluster types substantively
Adding spatial features to OLS
Spatial cross-validation
SLM vs. SEM; running models with spatialreg
Likelihood-ratio tests, AIC, and Lagrange multiplier specification tests
Sensitivity to the spatial weights matrix; reporting robustness

Moran's I LISA spatialreg SLM SEM spatial CV

Part 4 Advanced topics, reproducibility and ethics

Extending spatial analysis to time, producing transparent research, and grappling with the ethical stakes of working with geographic data.

Spatial panel data structures; visualizing change over space and time
Fixed vs. random effects in spatial panel models
Geographic re-identification and why location is quasi-identifying
Spatial aggregation, k-anonymity, and geomasking as disclosure control
Place-based algorithmic bias, and policy implications

panel data space-time Quarto geoprivacy spatial justice