Reviewed by
Jennifer Badham
Queen's University Belfast
Professor Cioffi-Revilla is the founder of the George Mason University computational social sciences (CSS) programme and Center for Social Complexity. His achievements in this area are widely recognised, including the recent award of Fellowship of the American Association for the Advancement of Science. This textbook draws on this experience, taking a theoretically oriented approach to presenting CSS over a vast interdisciplinary scope.The second edition differs little from the previous edition (reviewed in JASSS by Kalvas 2015) except for the welcome addition of a large number of problems and exercises for each chapter, some with solutions. These problems are diverse, with everything from multiple choice fact recall (such as names of theorists) through to potential essay questions and group projects.
Following the Introduction, Chapter 2 introduces the concepts and terminology of computer science, such as modularisation and data structures. This chapter is intended as an orientation and much of it is fairly dry and superficial, consisting of little more than definitions. However, there is a more substantial discussion of UML. Chapters 3 and 4 cover automated information extraction (big data or data science methods) and social networks respectively.
The next three chapters cover social complexity: origins and measurement (chapter 5), laws (chapter 6), and theories (chapter 7). While these chapters concern political complexity, the “extent to which a society is governed through nonkin-based relations of authority” (pg 205), they integrate ideas from across the social sciences, such as social development (anthropology/archaeology) and bounded rationality (psychology).
The final chapters concern simulation: methodology (chapter 8), variable-oriented models (chapter 9, system dynamics and queuing models), and object-oriented models (chapter 10, cellular automata and agent-based models). These chapters have a nice balance of theory about the modelling process, examples of research pursued with models, and practical issues such as relevant software.
The intended use of this book is for a graduate course in CSS. That raises the question of what is meant by ‘computational social science’, with at least two well-established definitions. Computational scientists typically focus on the network and data science aspects, to “collect and analyze data at a scale that may reveal patterns of individual and group behaviors” (Lazer et al. 2009). However, as should be evident from the description of the contents, this text covers the broader field of “interdisciplinary investigation of the social universe on many scales, […], through the medium of computation” (p. 2). In this social science orientation, CSS is the use of computational tools to measure, analyse, understand and model social complexity (Conte et al. 2012), and that emphasis on social complexity is the core of this book.
Despite being an introduction, the textbook is very comprehensive. This scope makes it relatively demanding of potential readers. The material is fairly dense, with formal definitions, mathematical notation and a theoretical orientation. While examples are provided for many of the definitions, the examples are often simply stated rather than explaining their relevance. Generally, the balance seems reasonable for the intended use in formal graduate studies particularly if, as is often the case for CSS, the student group is interdisciplinary and can bring diverse expertise. The exception is the treatment of social networks, which does not reach the quality of the remainder of the book. For example, there is no real discussion of features of social networks such as relatively high clustering, some networks are described just by the node entity (ignoring the relationship that makes up the edges) and a key example is unusually that of a decision tree.
Each chapter starts with a historical summary that lists notable events and (if known) the relevant scholar, and ends with recommended readings. Combined with the new problems for this edition, the textbook provides a range of pedagogical tools to assist the instructor.
It would also be a useful text for any CSS researcher to have on their bookshelf. It describes a range of relevant methodologies and is likely to stimulate new research questions and how to approach them in whatever aspect of CSS is of interest.
ReferencesKALVAS, F. (2015). Review of “Introduction to Computational Social Science”. Journal of Artificial Societies and Social Simulation, 18(1): jasss.soc.surrey.ac.uk/18/1/reviews/2.html.
Return to Contents
of this issue