Abstract
This is the second edition of a book that sold well, but had become somewhat dated with rapid developments in its field. Although primarily aimed at library and information staff, and with the strong implication that this is a field where LIS staff can make themselves indispensable to their employers, this book will also appeal to anyone in other professions who have an interest in web metrics. The book’s emphasis on LIS staff reflects the fact that they already engage heavily with online content, and here is an opportunity for LIS staff to demonstrate the impact that individuals and organisations have. The book provides an introduction to the various types of web metrics that can be analysed, and includes examples of software applications for those confident in such things. Each type of web metrics that can be analysed is given its own chapter, and the pros and cons of the use of such metrics – for none of them are perfect – are discussed. The author, who has published many books and research articles in the field, is without doubt an authority on the subject, and his commentary on each type of metric is helpful. Intriguingly, he describes himself as a ‘would-be theologian’!
The introduction chapter gives useful background to the whole concept of web metrics, why it is important and why LIS staff should engage with it. Chapter 2 explains the differences between bibliometrics, altmetrics, web metrics and webometrics. It is worth noting that not everyone follows the same definitions as the author does here. The author finishes the chapter by summarising some potential uses of metrics for LIS staff: measuring the impact of the library’s content, measuring the impact of library patrons’ content, identifying important research and resources, helping other staff with their webometric studies and providing patrons with overviews of a subject area.
The speed of developments in the field is reflected in the fact that the author repeatedly gives ‘Twitter’ amongst his examples, although that service changed its name to ‘X’ in July 2023 – the book itself was written in November 2022; fortunately, people are so used to the T name that this in no way invalidates the content of the book.
Chapter 3 discusses the tools available for data collection, and includes discussion of the components of a URL and the operation of search engines, The author emphasises that search engines, though powerful, do not cover everything that is available on the web.
Chapter 4 looks at how to measure web impact and looks at different approaches for web sites, social media, blogs and wikis, as well as the differences between internal metrics (about a particular item) and external metrics (the context of the results compared to other comparable sites). Figure 4.1 gives an example using Google Trends of how interest in British Library (in terms of keyword searches on Google) has declined over time. I wonder if the recent news that the BL has been hacked by blackmailers will reverse that trend. . . . . The author is critical of the idea that search engine ranking provides any useful data at all.
Chapter 5 looks at evaluating social media impact. Services discussed include Twitter (as was), Instagram, Facebook, Wikipedia and YouTube. No doubt future editions will also look at Mastodon and Threads. In this chapter, the author explores the fascinating topic of sentiment analysis, an area where Artificial Intelligence software will no doubt have an important role to play in the future.
Chapter 6 examines relational web metrics and social network analysis, whilst Chapter 7 looks at Web Bibliometrics. Chapter 8 then considers metrics for data and code, a topic that will surely grow in importance over time. Chapter 9 considers the future of web metrics and what it might mean for the LIS profession. This excellent final chapter includes possible future web metrics techniques that LIS professionals may wish to explore.
There are places where the author assumes knowledge that a reader might not have; examples include name-dropping white hat and black hat search engine optimisation in a couple of chapters, and the statistical softwares R and Python. Figure 3.3 on page 56 fails to provide a scale and Figure 6.2 requires a source reference. The name dropping of ‘inorganic financial campaigns’ without explanation of the term on page 96 is unhelpful. Similarly, on page 110–112, the author name drops a number of network analysis softwares as well as a methodology (Louvain Modularity) without giving their URLs or references to articles about them. Pages 117–120 and 147–149 include examples of the use of R codes to undertake analyses, but it is unclear how useful such examples would be for someone who has never used R before. On page 125, again the author name-drops – this time mentioning the i10-index and Matthew Effect without proper explanation.
The author writes in a friendly, approachable style. The book is supported by a lengthy list of references, and by (it must be said) a pretty basic index; the latter could certainly be expanded and made more in-depth in any future edition. I was surprised by just how expensive this book is, as I suspect the price will deter an individual from buying the book out of private interest.
