Visualization for Social Data Science

"This is an important book on an important topic. I particularly like the examples showing different visualizations of the same data and the parallel presentation of graphics and code. And I absolutely love the chapter on visual storytelling. I can't wait to use this book in my classes." - Andrew Gelman, Department of Statistics and Department of Political Science, Columbia University, New York

"A book that gives learners the inspiration, knowledge and worked examples to create cutting edge visualisations of their own." - James Chesire, Professor of Geographic Information and Cartography, University College London

Visualization for Social Data Science provides end-to-end skills in visual data analysis. The book demonstrates how data graphics and modern statistics can be used in tandem to process, explore, model and communicate data-driven social science. It is packed with detailed data analysis examples, pushing you to do visual data analysis. As well as introducing, and demonstrating with code, a wide range of data visualizations for exploring patterns in data, Visualization for Social Data Science shows how models can be integrated with graphics to emphasise important structure and de-emphasise spurious structure and the role of data graphics in scientific communication -- in building trust and integrity. Many of the book’s influences are from data journalism, as well as information visualization and cartography.

Each chapter introduces statistical and graphical ideas for analysis, underpinned by real social science datasets. Those ideas are then implemented via principled, step-by-step, workflows in the programming environment R. Key features include:

• Extensive real-world data sets and data analysis scenarios in Geography, Public Health, Transportation, Political Science;

• Code examples fully-integrated into main text, with code that builds in complexity and sophistication;

• Quarto template files for each chapter to support literate programming practices;

• Functional programming examples, using tidyverse, for generating empirical statistics (bootstrap resamples, permutation tests) and working programmatically over model outputs;

• Unusual but important programming tricks for generating sophisticated data graphics such as network visualizations, dot-density maps, OD maps, glyphmaps, icon arrays, hypothetical outcome plots and graphical line-ups plots. Every data graphic in the book is implemented via ggplot2.

• Chapters on uncertainty visualization and data storytelling that are uniquely accompanied with detailed, worked examples.