This repository is the companion to Allen Downey's 'Think Stats, 3rd Edition', an open educational textbook on exploratory data analysis in Python. It is primarily a collection of Jupyter notebooks containing book chapters, exercises, and solutions, plus a supporting Jupyter Book build pipeline. The content teaches practical statistics through real datasets: NSFG pregnancy and respondent surveys, BRFSS health data, Palmer Penguins, NHL goal timing, U.S. EIA electricity generation, historical stock data, and more. Topics progress from frequency tables and PMFs through hypothesis testing, bootstrapping, regression, time-series forecasting, and survival analysis. Although it has application-like automation (Makefile targets, CI, notebook build scripts), the dominant character is data_research: executed notebooks, bundled datasets, and statistical analysis workflows rather than a runnable software service.