GeoDa

GeoDa is a free and open source software tool that serves as an introduction to spatial data science. It is designed to facilitate new insights from data analysis by exploring and modeling spatial patterns.

GeoDa was developed by Dr. Luc Anselin and his team. The program provides a user-friendly and graphical interface to methods of exploratory spatial data analysis (ESDA), such as spatial autocorrelation statistics for aggregate data (several thousand records), and basic spatial regression analysis for point and polygon data (tens of thousands of records). To work with big data in GeoDa it should first be aggregated to areal units.

Since its initial release in February 2003, GeoDa's user numbers have increased exponentially to over 520,000 (June 2022). This includes lab users at universities such as Harvard, MIT, and Cornell. The user community and press embraced the program enthusiastically, calling it a "hugely important analytic tool," a "very fine piece of software," and an "exciting development."

The latest version 1.20 contains multi-layer support, several new local cluster features, including univariate and multivariate local Geary cluster maps, redcap, skater, spectral clustering and max-p, and local join count maps for categorical data. It also implements several classic non-spatial cluster techniques (principal component analysis, k-means, and hierarchical clustering) implemented in Hoon et al.'s (2013) C Clustering Library, as well as HDBScan.

A new workbook is under development. In the meantime, here are interim resources, including an overview of features in 1.20.