Customer Reviews

Biostatistical Analysis (4th Edition)

5 star:		(12)
4 star:		(4)
3 star:		(1)
2 star:		(2)
1 star:		(1)

 

 

This book and BIOMETRY by Sokal and Rohlf are the two standard statistics books that live on the shelves of most biological researchers.

Zar makes methods and interpretations of statistical analyses accessible and understandable. There are suffient numbers of statistical tests in the text to cover most of my statistical needs. The book also includes its own tables of critical values (something that BIOMETRY does not -- there's a paperback companion book of tables for that book). Zar also includes examples that are easy to follow, as well as enough mathematical background to allow one to understand the hypothetical and much of the mathematics behind the statistical methods.

I typically turn to Zar first when I have a statistical question. This book continues to be a highly prized ally of mine in the data analysis game.

I highly recommend this book to all biology grad students and researchers. It's well worth the price -- an investment that pays off!

5 stars, no doubt about it.

This book is popular because it is well written and authoritative. It is written for biologists, medical students and researchers who do not have any prior knowledge of probability or statistics and may have little mathematical training as well. It serves as an introductory text providing many homework exercises. It can also be used as a reference. It is very thorough and covers most of the important topics required for biological problems. The needed probability is introduced when necessary.
There is the usual emphasis on hypothesis testing and regression. Correlation and analysis of variance are also very well covered. Important issues of sample size determination are covered and many solutions are provided in easy to use box descriptions.

As the author points out in the preface, in order to make this text a good reference it is extensive (663 pages of text followed by appendices and a large number of tables). It also includes a wealth of useful reference articles and books. Consequently, there is too much material for a one semester course. The author provides instructors with guidelines for sections to cover in an introductory course.

Notable topics covered in this text that is rarely found in introductory biostatistics books include multivariate methods especially the multivariate analysis of variance (MANOVA)and inference for circular data.

Recent developments in meta analysis, Bayesian statistics and bootstrap methods are not covered. In fact, these topics are not covered at all. Also, the important topic of missing data is omitted. Outliers are only covered briefly and just a few references are given but the major references, the texts by Hawkins and the treatise of Barnett and Lewis are neglected.

I am currently working on an elementary text that will have the advantage of some real world applications and modern developments. There are a few other elementary statistical texts for biology that are worth considering including Motulsky's "Intuitive Biostatistics" and

This textbook covers a very broad spectrum of biostatistics. It is the best reference for design and analysis of experiments, regardless of the application. This opinion is coming from a person with an MS in statistics. -Brian Mayeux statistician for NIOSH (CDC).

This book is very concise and very mathemically based. It does a good job at presenting the mathematical models of statistics but it does a lousy job at explaining biostatistics in a conceptual manner. Maybe a great book for those who love mathemical models and notations but definitely not recommended for someone who is trying to understand biostatistics conceptually. If you are a scientist with a weak bacdground in statistics, you may find it a waste of time to try to understand this book.

I have found this book more approachable and user friendly than Sokal and Rohlf. Zar is an excellent desk reference, and has solved a number of statistical problems for me. I reccomend it to anybody who regularly uses statistics in a biological context.

The author knows the difficulties found by researchers when they need to analyse their data. He writes about all topics of statistical analysis in a very simple way, and the solved examples makes our understanding smooth. He knows how to drive us in such a way that analyzing biomedical data becomes compreehensible and easy! A book that biologists and people from the biomedical area must consult everyday! This book should be translated to other languages so much more people could learn from it! Lucia M. Singer, M.Sc. Ph.D.

As someone who recently retired from analyzing ecological data after a decade of it, I found this book to be pretty good one-stop shopping. I wouldn't say it's an introduction to stats so much as it is a systematic compilation of all the "traditional" statistical topics (t-tests, regressions, etc). As such, it contained some useful formulas that do not occur in regular "Stats 101" texts, such as sample size estimators for various analyses.

However, there are two things it is missing. As mentioned by other reviewers, there's no coverage (in the edition I have, anyway) of iterative techniques like bootstrapping, Monte Carlo approaches, etc. Those are coming up a lot in everyday statistical work these days.

More important is something missing from nearly EVERY beginning statistics text (and, often, from college education), which is the place of statistical testing in scientific logic. Too many beginners with statistics get stuck on fishing for significant differences in a stale old dataset rather than really thinking about their subject matter. In the absence of context, statistical "significance" can be deceptive and meaningless. One place to start on this subject is with Murphy & Myor's really good book called Statistical Power Analysis. I learned a ton from that book, which is a good companion for nearly any regular stats text. Happy crunching..

You will not go wrong with this book. It is the best general
biostatistical book in print. You will need to study it for it is not a cook book, but it is well written. An absolute must book.

This book is thought of as the "Statistics Bible" by the grad students at my school. This book is all a biologist will ever need for statistics reference.

As a working biologist with little formal mathematical training (and NO statistics training) I was looking for a text that had everything I needed to know about a statistical test--when to use it, when not to use it and how to use it. Zar fits the bill perfectly. I particularly appreciate that for each test he provides real world examples and then works the test out completely. Even when I've had difficulty understanding a test at first (with the ANOVA, for example), I was able to follow the example and convince myself that I had really grasped it.

Another reviewer mentioned the "pencil and paper" aspect of Zar as a negative, but I disagree. I love SPSS as much as the next person and I use it all the time, but my understanding of what test to use when and what that test is really doing come from Zar. A fantastic book!