Sas Language: Reference, Version 6

Rating:
Summary: Some balance is in order
Review: Some of the comments of the previous reviewer are appropriate, others are misleading, which I attempt to correct.

The book is titled 'SAS Reference' so what would one expect other than an exhaustive listing of SAS commands. There are dozens of other books out there which will teach you both Stats and SAS (or any other statistical tool you choose such as JMP, SYSTAT, SPSS). To say that "all" SAS Institute and commercial documentation have a SAS insider's bias is plainly inaccurate. If one doesn't know enough stats then these books are not to blame.

Most programmers who code in C, Fortran or Java would not consider SAS (or SPSS, SYSTAT) to be complete programming languages. Though many general programming do concepts apply, if the authors attempted to compare C with SAS, then the latter would appear tame howsoever powerful SAS may touted to be. The only statistical language that would come close to C would be R/S/S-plus family of languages.

SAS is meant for statistical analysis not for baking cookies. Is it any wonder you need to know stats? Similarly one would expect users of Mathlab or Mathematica to know some math - it is not aimed at teaching Math to English Majors, but make the job of engineers/scientists, who use math as a tool, easier.

Rating:
Summary: Exhaustive, layed out in order, but very poorly done.
Review: This book covers the whole of the SAS language up to about 6.06. While the core of the language is presented, and that information is presented in a fairly logical order, you'll have to keep in mind that since SAS seems to have been invented by people who were trying to do something different than anyone else in the industry had ever done, that there is nothing to relate any of the information to anything in any other programming language.

Although it says reference on the cover, it is about half reference. The other half covers a lot about the Rules of the SAS Language (which you will need to know to get anything done, or even have an idea of how to do it), and other SAS-unique concepts (and there are so many).

The index is almost useless. You will need to know SAS and everything it can do before it is useful (SAS almost always has its own command-name for any programming operation, and the index contains only those names; hardly a single mention of generic programming tasks or needs and where to find SAS statements that fulfil those needs).

The price of almost $50 is steep, but despite the poor nature of the book, you will still need this book if you are serious about SAS because even though it won't help you figure out what SAS command to use, when you do find out what that command is (any command not in one of the extensive array of sebsequent version updates), this book does provide a complete syntax, which can be the only peace of useful information.

It has been my experience that all SAS Institute documentation and all commercial documenation tends to explain things from the viewpoint of a SAS programmer, that is, someone who programs using SAS concepts and methods. This makes all such documentation difficult to impossible to understand to people who do not think in the "net" of the SAS "way" of programming. Having a heavy lean toward statistical analysis doesn't help, either (large portions of SAS functionality are obscure and useless if you don't have advanced statistical understanding).