Cody's Data Cleaning Techniques Using SAS, Second Edition

Good book on data management!

Those of us who work professionally in biostatistics with SAS keep Cody's book on the shelf above our computer. The methods and code are invaluable for finding and cleaning the dirty data that accompanies research and large medical datasets.

Get this book when you want to change from being somebody who knows some basic data manipulation into being a SAS data guru. This brilliantly well written text includes lots of well annotated code examples and output which demonstrate how to find and clean bad data. Topics are covered with multiple methods starting with basic tricks that "everybody knows" but with slight tweaks to make them work better and then fancier methods are shown with clear explanations on why you would want to bother with the more advanced techniques. For example, he covers getting basic descriptive numeric statistics with the typical SAS procedures (proc means and proc univariate) but he shows useful tweaks like how to get custom percentages and custom reports of extreme values listed. Then he shows how to make the results look good (with proc tabulate) and he also provides custom "macros" so you can easily copy and paste his methods and apply them to your data. The topics are consistently useful and well presented. I have been using and teaching SAS for 15 years and while I was reading this book I repeatedly found myself thinking "I wish somebody told me that trick years ago." Because of the time you will ultimately save with Cody's techniques, this book is money well spent.

Like other books by Ron Cody, this book is easy to read and understand. The explanation is clear with different examples and ways to write the SAS codes to achieve similar objectives but to suit different situations. You can download the programs and data sets from SAS website to try the codes without typing it yourselves. I normally had problems focusing on a book for a long time and skipped exercises to finish the book the earliest. It did not happen to me when reading this book. Moreover the book covers all the key problems with data especially when you touch it the first time. Even if you are a SAS expert, you will be surprised to learn a better way to code SAS. I highly recommend this book.