Module 6: References and Further Reading

This final module steps back from specific methods to focus on the broader principles of conducting high-quality, reliable, and ethical geospatial research. It also provides a curated list of resources to support your continued learning journey in spatial epidemiology.

By the end of this notebook, you will have access to: 1. Links to the excellent online courses that inspired this one. 2. A list of key textbooks for deepening your theoretical and practical knowledge. 3. Direct links to the documentation for all the R packages used in this course. 4. An overview of best practices for ensuring your work is reproducible and ethical.

6.1 Course Recommendations and Key Resources

This course was developed with inspiration from several outstanding, publicly available resources.

  • Spatial epidemiology in R by Hugh Sturrock A comprehensive online course covering a wide range of topics in spatial epidemiology, from which some concepts and datasets in this course were adapted.

  • The Epidemiologist R Handbook An excellent, practical guide for using R in epidemiology. The “GIS Basics” chapter provides a great overview of handling spatial data in R

6.2 Essential Textbooks for Further Learning

To dive deeper into the theory and application of spatial analysis, the following books are considered foundational texts in the field.

  • Geocomputation with R by Robin Lovelace, Jakub Nowosad, and Jannes Muenchow. This is the modern, definitive guide to doing spatial analysis in R, with a focus on the sf, terra, and tmap packages. The entire book is available online for free.

  • Applied Spatial Data Analysis with R, 2nd edition by Roger S. Bivand, Edzer Pebesma, and Virgilio Gómez-Rubio. This book provides a deep, rigorous dive into the statistical theory behind many spatial methods, including the use of the spdep package.

  • Spatial Data Science: With Applications in R by Edzer Pebesma and Roger Bivand. An updated text from the leading figures in the R-spatial community, reflecting the most current approaches and packages.

6.3 Core R Packages Used in This Course

This course would not be possible without the incredible ecosystem of open-source R packages developed and maintained by the community. Below are links to the official documentation for the primary packages we used.

Data Handling and Manipulation

  • sf: The modern standard for working with vector data (points, lines, polygons) in R.
  • terra: The modern, high-performance package for working with raster data.
  • dplyr: A core part of the tidyverse, providing a powerful grammar for data manipulation that works seamlessly with sf objects.

Data Access

  • spData: Contains the diverse sample datasets used throughout this course, such as us_states and world.
  • spDataLarge: A companion package containing larger datasets, including the landsat.tif satellite image.
  • geodata: A package for downloading common global datasets like country boundaries from GADM and climate data from WorldClim.

Visualization

  • tmap: A dedicated package for creating beautiful and flexible static and interactive thematic maps.
  • rayshader: A package for creating beautiful 2D and 3D data visualizations, especially for mapping and elevation data.

Spatial Statistics & Validation

  • spdep: A foundational package for analyzing spatial dependence, including tools for creating neighbor weights and running tests like Moran’s I.
  • spatstat: A comprehensive suite of tools for point pattern analysis, including Ripley’s K-function.
  • caret: A general-purpose package for machine learning and model training that provides excellent tools for creating confusion matrices and calculating accuracy statistics.
  • pROC: A specialized package for generating, analyzing, and plotting ROC curves.
  • smacpod: A package that implements spatial scan statistics.