Spatial Regression Models for the Social Sciences. Jun Zhu
in an easy-to-follow approach is much needed.
Therefore, we have attempted to write this as a primer type of textbook for social scientists who would like a quick start to learning spatial regression methods. While the methods are many and counting, we have decided to focus on the methods that are commonly used by social scientists and tend to be useful to them. These methods include exploratory spatial data analysis, methods dealing with spatial dependence and/or spatial heterogeneity, and more advanced spatial regression models.
Why This Book?
There are a number of existing books on spatial regression methods in the field of spatial statistics. They provide comprehensive coverage of some, if not all, of the three main components of spatial statistics: spatial point pattern analysis, lattice (or areal) data analysis (i.e., spatial regression models and methods), and/or geostatistics; a few books also address spatial interactive data analysis. However, these textbooks tend to be written for natural scientists or regional scientists and require that readers have a good understanding of mathematical statistics. In addition, they are not necessarily tailored to areal (or lattice) data analysis in social science research, which is the most useful component, at least presently. This is not a criticism of these books; rather, we simply see a lack of a spatial regression textbook for social scientists interested in learning the models and methods in a comprehensive and easy-to-follow manner and with limited training in mathematical statistics. This is where our book comes in: like these other books, it discusses spatial regression models and methods, and unlike the other books, it is written specifically for social scientists.
The distinguishing features of this book include the following:
1 It is geared toward social scientists who are familiar with standard regression methods and would like to learn spatial regression models and methods.
2 It provides relatively comprehensive coverage of spatial regression models and methods for social scientists.
3 The spatial regression models that it covers are commonly used by social scientists and are of interest to them.
4 It introduces the spatial regression models and methods in a generally easy-to-follow manner.
5 All figures and illustrations have color versions available at study.sagepub.com/researchmethods/quantitative-statistical-research/chi
Who Is This Book for?
This book could be particularly useful for social scientists who are familiar with standard regression methods and desire to learn spatial regression models and methods. It can be used as an introductory book to get to know spatial regression methods and apply them readily. It can also be used for a one-semester quantitative social science course at both undergraduate and graduate levels.
In addition, this book could be useful for social scientists who are interested in using spatial regression methods in their research. This includes instructors, researchers, and students in a wide range of social science disciplines such as sociology, demography, criminology, anthropology, human geography, economics, education, communication, history, law, political science, psychology, urban studies and planning, and others. The book may serve these social scientists as both a textbook and a reference book.
Although it is chiefly intended to be a core textbook for social science courses that focus on areal (or lattice) data analysis using spatial regression models and methods, it could serve as a supplemental textbook for social science courses that provide a more general coverage of regression methods.
A website for the book at study.sagepub.com/researchmethods/quantitative-statistical-research/chi contains color versions of the figures from the book, together with the tables that form Appendix B: Results Using Forty Spatial Weight Matrices.
Acknowledgments
In 2008, we published an article titled “Spatial Regression Models for Demographic Analysis” in the journal Population Research and Policy Review to introduce spatial regression models for demographic research. It was well received, largely because spatial regression models were still new to social scientists at that time. Around 2010, we started to write a book of spatial regression models and methods for social scientists. Between now and then, we have presented the book idea and many rounds of revision in a variety of venues, and the feedback has been positive and enthusiastic: There does appear to be a need for such a spatial regression method book for social scientists.
This book is not a work that is solely ours; rather, it is a work collectively accomplished by us with tremendous support from our mentors, colleagues, students, editors, friends, and families. First, we would like to thank our students, Donghui Wang, Maria Kamenetsky, Supriya Joshi, and Taylor Hackemack, for having provided assistance with figures, tables, and data analysis.
We received valuable advice and suggestions from mentors and colleagues. Paul Voss was instrumental in teaching Guangqing Chi spatial regression methods. Stephanie Bohon, David Levinson, Stephen Matthews, David Plane, Stuart Sweeney, Tse-Chuan Yang, and Stephen Ventura provided advice and suggestions for our book.
We would also like to thank the institutions where we work(ed) during the development of the book: The Pennsylvania State University, University of Wisconsin–Madison, Mississippi State University, and South Dakota State University. This book is supported in part by the National Science Foundation (Awards CMMI-1541136, OPP-1745369, DGE-1806874, and SES-1823633), the National Aeronautics and Space Administration (Award NNX15AP81G), the U.S. Department of Transportation (Award DTRT12GUTC14), and the National Institutes of Health (Awards P2C HD041025 and U24 AA027684).
This book would not be possible without the tremendous support from the SAGE team. Helen Salmon, our Acquisitions Editor, was essential in guiding us to revise the book to be high quality and to tailor it better for our readers. Professor Shenyang Guo, the Editor of the Advanced Quantitative Techniques in the Social Sciences Series, has provided insightful comments for our book. Megan O’Heffernan has been very helpful in assisting us during the final touches of the book. Also, reviewers provided many constructive comments and suggestions, which have helped us revise this book from a reference book into a textbook that is more pedagogical. They are J. S. Onesimo Sandoval, Saint Louis University; Daoqin Tong, Arizona State University; Peter Rogerson, University of Buffalo; Wenwu Tang, University of North Carolina at Charlotte; Changjoo Kim, University of Cincinnati; and Karen Kemp, University of Southern California.
We are grateful to Cindy Sheffield Michaels, now Editor of Special Publications for ASHRAE, for having worked with us from the very beginning. She not only edited the content and format but also provided constructive suggestions to make the book readable from the reader’s perspective. What’s more important, she encouraged us to move this book project forward when we ran out of steam, several times.
Guangqing’s eldest daughter, Claire, has asked multiple times when the book will be published, because Guangqing promised her at the very early stages of this project that he would acknowledge his family so that Claire’s name will show up in a book. So, Guangqing’s last but not the least acknowledgment goes to his family—Yunjuan Jiang, Claire Chi, Gloria Chi, and Lydia Chi—for their patience and support. As for Jun, she would like to take this opportunity to express her deep gratitude toward all those who have mentored her with kindness and compassion, especially her parents, academic advisers, and colleague mentors.
Guangqing Chi and Jun Zhu
October 2018
About the Authors
Guangqing Chi is Associate Professor of Rural Sociology and Demography in the Department of Agricultural Economics, Sociology, and Education with courtesy appointments in the Department of Sociology and Criminology and Department of Public Health Sciences at The Pennsylvania State University. He also serves as Director of the Computational and Spatial Analysis