Data Analysis Using SAS ENTERPRISE GUIDE, by Lawrence S. Meyers, Glenn Gamst and A.J. Guarino, Cambridge, Cambridge University Press, 2009, xix + 378 pp., \pounds35.00 or US$51.99 (paperback), ISBN 978-0-521-13007-3
One of the key features of this book is its dual-purpose design which makes it useful either as a stand-alone resource or as a supplement to a statistics course. In a total of 11 building blocks, it addresses a variety of topics ranging from an introduction to SAS Enterprise Guide (EG) through Exploratory Data Analysis to specific statistical modelling techniques. It provides a well-balanced ratio between statistical topics and the software in which the examples are implemented, making it easy for a reader with a moderate working knowledge of any other statistical software to adapt the examples. The book forms a handy tool to undergraduate and early graduate students. As a regular SAS user, I found it quite easy to read and follow the examples. Indeed, its simplified statistical language and the presentation of examples in a user-friendly graphical interface make it equally easy to follow by any regular data analyst with or without a strong statistical background.
It is important to highlight a number of issues – some of them quite subtle. Firstly, the limitation of the log transforms mentioned on p. 137 does not seem right for numbers less than unit. Secondly, as a regular SAS user over the years, I have noted a general feeling of confusion between the conventional SAS application and SAS EG especially among some members of the non-SAS community. Thirdly, despite the aforementioned flexibility, the book may still be viewed as software-specific in some respects – hence carrying the potential risk of sliding into obsolescence as new SAS EG versions come. Finally, the professional background of the authors (psychology) appears to have confined the general approach of the book to statistical analysis. To the authors, these issues may present both a challenge and an opportunity for enhancing future versions of the book. For instance, providing an elaborative link between the two SAS streams through illustrated examples would potentially widen its audience. This can be achieved by simply drawing examples from a wide range of disciplines, illustrating easy data exchange between the two SAS applications and attaching a CD containing well-illustrated examples from both streams.