This repository contains all the materials used for teaching an introductory workshop on Small Area Estimation (SAE) methods, specifically designed for geospatial data scientists. This course introduces participants to Small Area Estimation (SAE) — a set of powerful statistical methods that allow us to produce reliable estimates for geographic or demographic domains with limited or no sample data. SAE is increasingly relevant for spatial analysis, where we often face data gaps at fine spatial resolutions.
This repository includes:
- Workshop slides: Explaining the concepts behind direct, area-level, and unit-level estimators — from intuitive overviews to formal models.
- Hands-on R practicals: Step-by-step guide to demonstrate how to apply SAE methods using toy data sets.
This workshop is aimed at:
- Urban and spatial data analysts
- Researchers and students working with survey or census data
- Public policy analysts seeking to improve local-level decision-making with limited data
No prior experience with SAE is required, but basic familiarity with R and survey data will help you get the most out of the practical.
By the end of the workshop, you will:
- Understand why SAE methods are useful when working with sparse survey data
- Be able to implement direct, area-level, and unit-level SAE methods in R
- Know how to incorporate auxiliary data and spatial structure into your models
- Be equipped to choose the appropriate method depending on data availability
Make sure you have R ≥ 4.5.1 and install the required packages listed in the practicals. Follow the instructions in each RMarkdown or Quarto file to reproduce the results.