NB. I was sent this book as a review copy.

From Princeton University Press

If you want insights into what makes a good collaboration dream-like and a bad collaboration nightmarish, this is the book for you.

In short, The Strength in Numbers details an extremely important piece of research, with reference to many other studies, which aims to analyse collaborations within STEM, and figure out not only measures of collaboration effectiveness, but also ways to make your own collaborations more likely to succeed.

Academia is a funny old game, where there is extensive training in certain aspects of the job (the fundamental tools of science, for instance), and others are left to the researcher to try and piece together as they go along. Some obvious and frequent examples of these are:

• How to write and give talks effectively
• How to mentor young researchers
• How to best disseminate your own knowledge

and perhaps most importantly, how to create an effective collaboration.

Before reading this book, this wasn’t something that I had ever considered. I figured that to collaborate you find someone who has similar interests to you, you talk about an idea, you try and do the idea, you write about the idea, you publish the idea. Indeed I hadn’t thought about this despite the number of collaborations which have fallen by the way side, or had real difficult frictions throughout, or ended in acrimony.

We may think of ourselves as logical beings, but so often we do the same thing over and over again, with bad results, expecting something to magically change. Of course, that is generally not the way the world works, so we can ask ourselves the question, ‘how do we make collaborations work?’

The first question that the book looks into, via a questionnaire of some 600 academics from 100 institutions (in the US) is: What categories of collaboration effectiveness can we find? They define four main categories:

• Dream collaboration
• Routinely good
• Routinely bad
• Nightmare collaboration

One of the most interesting findings is that, paraphrasing from Anna Karenina, “All happy collaborations are alike, each unhappy collaboration is unhappy in its own way”. The examples are indeed fascinating, where some issues may be very familiar, whereas others may seem anything from implausible to unbelievable (though complete credibility is granted!).

They go on to discuss the variance of each of the above categories. From this, they look into what makes a good collaboration effective. This may seem relatively simple, but as soon as you start taking into account long term impact of research, the training of young researchers, the building of inter-university and university-industry links, the building of long term collaboration, etc. you see that this is really a tricky topic.

The next question that is looked at is: What are the determinants of research collaboration effectiveness? ie. given that we have found a way to define a good collaboration, what are the factors that went into making it work?

And finally, and perhaps most importantly in the last chapter of the book they ask the question: What can be done to enhance research collaboration effectiveness? In this chapter they define different models of collaboration, from tyrannical management to consultative management, and four separate categories in between. This is perhaps, for a researcher, the most important part of the book, as it becomes clear that there are ways to set of a consultative collaboration strategy which is most likely to lead to a dream collaboration, and circumvent the possible pitfalls of a nightmare collaboration.

While essentially the book is an extended paper (in terms of reference to the study, and the detail that it contains), it reads very well, and while this may not be a research area or methodology which is of specific interest to you, I would recommend that anybody who collaborates either within academia, and/or with industry takes the insights in this book to heart.

Overall, this book covers a fascinating piece of research and I truly believe that the ideas within it are ones which every researcher should be thinking about every time they embark on a new team effort. Humans are messy, science is messy, so when we put the two together, we should keep in mind the best ways to organise these chaotic worlds in a fashion which leads to positive outcomes.