As promised in this post, I said that I would review “Exploring Animal Social Networks”, by Croft, James, and Krause. I finished the book a while ago, but I’ve been stuck in the hell of my examen de synthèse and haven’t had time to do a proper review.
Having said that, I’m not going to do a proper review anyways. Why? Well, as full disclosure, I’ve spoken with Dr. Krause about the possibility of joining his lab as a postdoctoral researcher when I’ve finished my Ph.D; while it’s only a preliminary discussion at this point, any review I would make of this book would probably be tinged in a way I don’t feel like trying to control. So, I’ll just mention a few points of interest about the book, and I’d be happy to answer questions in the comments.
The book’s publication, two years ago, came at a perfect time. Social network analysis (SNA), the study of relationships among individuals by encoding them as a mathematical object known as a graph, has been a topic of some interest in fields such as sociology for decades, but has only recently attracted widespread interest in biology. In animal behavior, the use of SNA is even more recent. Its usefulness has been hampered by a lack of statistical tools to accommodate sampling, and it is only in the past fear years that methods from mathematics, statistical physics, and nonparametric statistics have converged to enhance the usefulness of SNA for behavioral biologists.
Exploring Animal Social Networks synthesizes recent advances in the field and provides an extremely accessible introduction to the methods and applications of SNA to animal behaviour. They discuss issues of interest to practicing researchers: sampling, coding, and descriptive and inferential analysis; their statistical test of choice is randomization tests, and the authors discuss the applications of these tests using real data. The tests they discuss are not a cure-all (and recent papers are engaged in a lively debate over the best applications of these methods), but they are careful to point out where researchers might go astray.
In reading it, I found the book quite engaging and well-written. Don’t be fooled by the mentions I made above of the various mathematical fields which inform SNA; this is not a mathematical text, and though there are a few equations sprinkled through the text, readers of all ability should be able to read it with ease. If I have one criticism, it’s that it’s too short by half and I felt that they were too quick to pass off the details to other texts, like Wasserman and Faust. It would have been nice to see them dedicate more time to integrating the older methods of SNA into a context that behavioral biologists can understand.
The other problem – which is not within their control – is that the book is going to be outdated within five years of its publication. The consensus on how to apply these methods to biological question is still emerging, and it seems likely that the pace of publication won’t slow down any time soon. An updated version would be a good idea within the next few years.
My own experience with the text is in light of the fact that I’ve been following SNA methods for years (including the work of the authors), and have done a bit of work applying it to my own research; from that perspective I can say that aside from some of the discussion on statistical methods, there was little in the book that I hadn’t seen in other places. However, I still greatly enjoyed the book’s presentation, and I can imagine that naive readers will find it one of the better introductions to the area that they are likely to find.
My verdict? If you’re a researcher or student looking to explore the application of SNA to biological questions and have no previous experience with the topic, this is a no-brainer: buy the book. And though the material is deeply mathematical at its core the authors present it with a minimum of mathematical formalism, which should be accessible to even the most math-phobic of readers.
If you’re more experienced then you won’t find much new here, but the book manages to combine a great introduction with a good round-up of recent research; the combination will be of value even to researchers well-familiar with the field.
Exploring Animal Social Networks, by Darren P. Croft, Richard James, and Jens Krause (2008). Princeton University Press: Princeton, New Jersey.