Table of Contents

Introduction
Why this book?
Defining big data and its value
Social science, inference, and big data
Social science, data quality, and big data
New tools for new data
The book’s "use case"
The structure of the book
Resources

Capture and Curation
Working with Web Data and APIs
Introduction
Scraping information from the web
New data in the research enterprise
A functional view
Programming against an API
Using the ORCID API via a wrapper
Quality, scope, and management
Integrating data from multiple sources
Working with the graph of relationships
Bringing it together: Tracking pathways to impact
Summary
Resources
Acknowledgements and copyright

Record Linkage
Motivation
Introduction to record linkage
Preprocessing data
Classification
Record linkage and data protection
Summary
Resources

Databases
Introduction
DBMS: When and why
Relational DBMSs
Linking DBMSs and other tools
NoSQL databases
Spatial databases
Which database to use?
Summary
Resources

Programming with Big Data
Introduction
The MapReduce programming model
Apache Hadoop MapReduce
Apache Spark
Summary
Resources

Modeling and Analysis
Machine Learning
Introduction
What is machine learning?
The machine learning process
Problem formulation: Mapping a problem to machine learning methods
Methods
Evaluation
Practical tips
How can social scientists benefit from machine learning?
Advanced topics
Summary
Resources

Text Analysis
Understanding what people write
How to analyze text
Approaches and applications
Evaluation
Text analysis tools
Summary
Resources

Networks: The Basics
Introduction
Network data
Network measures
Comparing collaboration networks
Summary
Resources

Inference and Ethics
Information Visualization
Introduction
Developing effective visualizations
A data-by-tasks taxonomy
Challenges
Summary
Resources

Errors and Inference
Introduction
The total error paradigm
Illustrations of errors in big data
Errors in big data analytics
Some methods for mitigating, detecting, and compensating for errors
Summary
Resources

Privacy and Confidentiality
Introduction
Why is access at all important?
Providing access
The new challenges
Legal and ethical framework
Summary
Resources

Workbooks
Introduction
Environment
Workbook details
Resources

Bibliography

About the Author

Ian Foster is a professor of computer science at the University of Chicago as well as a senior scientist and distinguished fellow at Argonne National Laboratory. His research addresses innovative applications of distributed, parallel, and data-intensive computing technologies to scientific problems in such domains as climate change and biomedicine. Methods and software developed under his leadership underpin many large national and international cyberinfrastructures. He is a fellow of the American Association for the Advancement of Science, the Association for Computing Machinery, and the British Computer Society. He received a PhD in computer science from Imperial College London.





Rayid Ghani is the director of the Center for Data Science and Public Policy, research director at the Computation Institute, and senior fellow at the Harris School of Public Policy at the University of Chicago. His research focuses on using machine learning and data science for high-impact social good and public policy problems in areas such as education, healthcare, energy, transportation, economic development, and public safety.





Ron S. Jarmin is the assistant director for research and methodology at the U.S. Census Bureau, where he oversees a broad research program in statistics, survey methodology, and economics to improve economic and social measurement within the U.S. federal statistical system. He is the author of many papers in the areas of industrial organization, business dynamics, entrepreneurship, technology and firm performance, urban economics, data access, and statistical disclosure avoidance. He earned a PhD in economics from the University of Oregon.





Frauke Kreuter is a professor at both the University of Maryland and the University of Mannheim. She is also head of the Statistical Methods Group at the Institute for Employment Research in Germany. Among her over 100 publications are several textbooks in survey statistics and data analysis. She established the International Program in Survey and Data Science and is a fellow of the American Statistical Association. She received a PhD from the University of Konstanz.





Julia Lane is a professor at the NYU Wagner Graduate School of Public Service and the NYU Center for Urban Science and Progress. She is also an NYU Provostial Fellow for Innovation Analytics. She co-founded the UMETRICS and STAR METRICS programs at the National Science Foundation, established a data enclave at NORC/University of Chicago, and co-founded the Longitudinal Employer-Household Dynamics Program at the U.S. Census Bureau and the Linked Employer Employee Database at Statistics New Zealand. She is the author/editor of 10 books and the author of over 70 articles in leading journals, including Nature and Science. She is an elected fellow of the American Association for the Advancement of Science and a fellow of the American Statistical Association. She received a PhD in economics from the University of Missouri.

Reviews

"This book builds a nice bridge connecting social science and big data methodology. Big data such as social media and electronic health records, empowered by the advances in information technology, are an emerging phenomenon in recent years and present unprecedented opportunities for social science research. This book was written by pioneering scientists in applying big data methods to address social science problems. As shown by numerous examples in the book, social science could benefit significantly by embracing the new mode of big data and taking advantage of the technical progress in analysing such data. If you work in social science and would like to explore the power of big data, this book is clearly for you. Indeed, if you do not have previous experience in dealing with big data, you should read this book first, before implementing a big-data project.
As indicated by the title, this book acts as a practical guide and targets readers with minimum big data experience, hence it is very hands-on. … It covers all necessary steps to finish a big data project: collecting raw data, cleaning and preprocessing data, applying various modelling tools to analyze the data, evaluating results, protecting privacy, and addressing ethical problems. … All the important topics concerning big data are covered, making this book a good reference that you should always keep on your desk."
— Guoqiang Yu, Virginia Tech, in Journal of the American Statistical Association, July 2017 "…In summary, although there is a growing number of books related to social science and big data, this volume contains several non-trivial aspects which make it worth to have in the library, possibly along with other similar textbooks as a good complement to them."
—Stefano M. Iacus, University of Milan, in Journal of Statistical Software, June 2017 "This is a well-written book and showcases a good number of examples and applications to demonstrate how the methods are actually used in real life situation using real datasets. Further, topics at hand are motivated by social science data. … The chapters are nicely structured, well presented and motivated by data examples. The main strength of the book is that it still offers a good number of applications that are based on real datasets emerging from social science perspectives. The book will be useful to students, practitioners, and data analyst in the respective fields. The editors did a very good job introducing the book, it aims and goals, intendent audience, clarifying underneath concepts and phrases, a must read before moving to other chapters."
—S. Ejaz Ahmed, in Technometrics, April 2017 "Economists and Social Scientist have a lot to learn from Machine Learning, and Engineers have a lot to learn from Econometricians and Statisticians. This two way sharing is long overdue and it is time to start the conversation. This book is a tour-de-force for anyone interested in participating in such a discussion."
—Roberto Rigobon, Society of Sloan Fellows Professor of Applied Economics, MIT "This ambitious sweep through data science techniques provides an invaluable introduction to the toolbox of big data methodologies, as applied to social science data. 　It provides tremendous value not only to beginners in the field, but also to experienced data scientists wishing round out their knowledge of this broad and dynamic field."
—Kenneth Benoit, Department of Methodology, London School of Economics and Political Science "Most social scientists would agree that ‘big data’ – the term we use to encapsulate the huge amount of electronic information we generate in our everyday lives – provide the potential for path-breaking research not just into our economic, social, and political lives but also the physical environment we create and inhabit. However, few have the knowledge, or critically, the tools that equip them to realize this potential. This book provides a bridge between computer science, statistics, and the social sciences, demonstrating this new field of ‘data science’ via practical applications. The book is remarkable in many ways. It originates from classes taught by leading practitioners in this area to federal agency research staff, drawing in particular upon the example of a hugely successful project that linked federal research spending to outcomes in terms of patents, job creation, and the subsequent career development of researchers. By making these workbooks accessible, the book takes the novice on a step-by-step journey through complex areas such as database dynamics, data linkage, text analysis, networks and data visualization. The book is a treasure trove of information. It leads the field in the important task of bringing together computer science, statistics, and social science. I strongly recommend that all social scientists with an interest in ‘big data’ immerse themselves in this book."
—Peter Elias CBE, University of Warwick "The explosive growth in big data and in new technologies to analyze these data is transforming the practice of research in a variety of fields. Foster, et al. provides a well-timed, valuable guide to the new methods and tools associated with big data that can be used to address critical research questions in the social science field. The breadth of the material is impressive, providing a comprehensive summary of the methods and tools as well as practical guidance for their use. A key feature of the guide is the use of a case study to illustrate how big data techniques can be used to address a research question from beginning to end of the project, including providing examples of computer code targeted to specific steps in the project. Any researcher will find this unique guide to be useful, and it is essential reading for any social science practitioner that wants to use the best available data to conduct influential research in the near future."
—Paul Decker, President and CEO, Mathematica Policy Research "The typical statistics pedagogy has changed. In the past, textbooks assumed that data was hard to obtain, but neatly organized in a single file. Today, data is very easy to obtain from a number of data sources, often very messy, and analysts are now responsible for organizing it in addition to deriving useful insights. Foster, Ghani, Jarmin, Kreuter, and Lane have assembled a book that gives a pointed overview of tools to facilitate the entire digital lifespan of data in this era of analytics. Big Data and Social Science gives an evenhanded look at the myriad of ways to obtain data--whether scraping the web, web APIs, or databases--to conducting statistical analysis to doing analysis when your data cannot fit on a single computer. Meanwhile, they provide sound, diligent advice on pitfalls that still, and will always, exist. A book like this is useful for social scientists, experienced statisticians, econometricians, and computer programmers who want to see the tools available to them. It will also be a helpful text for a budding data scientist who wants a fairly technical preview of the landscape."
—Tom Schenk Jr., Chief Data Officer, City of Chicago "In Big Data and Social Science, the authors have deftly crafted one of the very best "how-to" books on big data that researchers, enterprise analysts, and government practitioners will find equally valuable. From Nodes, to Edges, to Arcs, the book takes the reader along a near-perfect path to understanding the fundamental elements of constructing a practical and realistic model for Big Data Analysis that any organization can execute by simply following the path outlined in this book. Elegant in its simplicity, Big Data and Social Science is one of those books that every research group and data-analysis team will want to have on their reference shelf."
—Tom Herzog, Former Deputy Commissioner, NY State Department of Corrections and Community Supervision   "This book offers a radically different programme of statistical training for those social scientists looking to engage with "big data". Individual chapters cover techniques and analytical approaches, each one introducing software you may have heard about but not used (e.g. Python, SQL, Hadoop, Tableau). The sheer scale of the task is ideally designed for teams of people with skills across computing, data and statistics, as well as hardware support of an institutional nature. Thus, the text covers a full course of big data skills with substantive examples drawn from social data available online. Exercises are supported by Jupyter workbooks which, using Anaconda, follow through description in the text, each one offering additional and complementary programming and analysis skills: a prodigious challenge. Examples of analysis both comprehensible and tractable for the exposition remain unsatisfyingly simple, but the references are extensive and impressive. ...The initial gushing enthusiasm of the author is soon tempered by practical and critical consideration of tools, but overall the book promises more than is possible to deliver. The result is a spare text, heavy on technical skills, which moves efficiently through the subject without quite giving confidence for application to real substantive questions."
-Thomas King, Biometrics June 2018 "...It is clear that the editors have put a lot of thought into the structure and organization of this book …the book demonstrates how to collect information about grants and about people funded on grants from web sources of the funding agencies and universities, how to link information coming from different sources, how to store and organize such data in ways that allow for quick summarization and data exploration, as well for convenient extraction of data for further research. The parts of the book with cautionary tales and advice regarding limitations of data science as an approach to carry out social science research should be required reading for all those involved in data science...Overall, the book could be useful to researchers and data analysts who would like to understand overarching ideas of data science and big data analysis. Social scientists interested in the topic will gain knowledge of basic steps required for working with big data and will deepen their understanding of the tools and the associated language used by the data science community. The book could also be used by instructors of graduate and undergraduate courses that touch on big data."
-Elena A. Erosheva, University of Washington