This is not a title to attract the overworked library-based information specialist looking for a weekend alternative to TV, and certainly there is no lack of publications on the theory of statistics and application to specific areas such as business and education. There is also a growing number on employing statistics in libraries, and general information work. So what has this book got to offer? One selling point is given by its subtitle: ‘A Primer for Using Open Source R Software for Accessibility and Visualization’. Most of us will already be familiar with the proprietary SPSS software – one of more than seventy available. However, R (originating in New Zealand where the development team is based) is a growing organism in that packages have been developed by many contributors for a variety of applications featuring statistics and graphics. Users can download the software appropriate for Windows, Linux and Mac operating systems. Although based on R, some packages have been developed in Java and C++.
Friedman briefly explains that R uses a command line interpreter instead of machine language. His book is divided into three main parts: introduction to statistics, making sense of statistics and visualisation. The first part introduces research design, data collection and an overview of running R. The second part goes into further details about formulae and methodologies (e.g. mean, variability and deviation), whilst the third part deals with the less familiar skills of visual presentation of statistical data for which R is particularly valuable. It is this section which makes the book one for the experienced statistician as well as for students. Although designed as a textbook for students and practitioners and employing natural language, much hands-on practice would seem to be essential in order to become familiarised with theory and application.
An attempt has been made to include library-based examples of the uses of statistics which are not U.S.-centric (which the ‘References’ are). One illustration (p. 6) is of an English Post Office Savings Bank deposit book from 1939 (incorrectly captioned 1869), and a University of Adelaide survey which included students’ smoking habits is used to illustrate the use of the MASS package to create a ‘data.frame’ (pp. 230, 231). On the other hand one wonders at the identity of the U.S. ‘Librarian [sic] Association’ whose survey is mentioned (p. 143). Three appendices have been provided: ‘A’ consists of a very useful collection of statistical formulae; ‘B’ presents a Z score table; and ‘C’ gives a handy set of R commands.
This book is not as daunting as it first appears, especially if used as a practical guide to contemporaneous use of the software. Its chief value for students and practitioners lay in the detailed coverage of the topic, the visual examples of formulae, and especially the final section illustrating in colour (a bewildering 655 are available) the graphical presentation of data.
© 2016 Edward Reid-Smith
http://dx.doi.org/10.1080/00049670.2016.1242105