Curiculum Vitae
Other Writings
Book reviews
from the
Dutch Mathematical
Book reviews
from the journal
Acta Applicandae

Book review

Author(s) Barnett, V.
Title Sample survey: principles & methods
Publisher Edward Arnold
Year of publication 1991
Reviewed by Anatoly Zhigljavsky

The book is a revised edition of a well-known work of the author, having the slightly different title "Elements of sampling theory". During the last fifteen years the first edition has been widely used as a text book for students and applied statisticians involved into sample surveying of population. Evidently the main aim of the present edition is to replace that book. There is a good chance that the aim can be achieved due to the following reasons: (i) the author is a well-known statistician and an expert in the field, (ii) the first edition has proved to be of a good quality and has won recognition, and (iii) the price is relatively low.

It may not happen mainly because there are many similar books available, here are only three of them: W. Cochran, Sampling techniques, Wiley, 1977; G. Kalton, Introduction to survey sampling, Sage, 1983; A. Stuart, The ideas of sampling, Griffin, 1984.

The main difference between the two editions is that the present one pays more attention to non-statistical, organizational problems and a discussion of some modern sampling technique is been added. In particular, the survey methods based on the stratification sampling are widely discussed in the book.

According to the author, the subject is presented at an intermediate mathematical level. From the point of view of a mathematician, the level of presentation is even lower. Thus, one of the most complicated mathematical concepts met in the book is just ordinary correlation.

The book is well produced and is relatively cheap. It can certainly be suitable for students following the course of survey sampling. The libraries of the universities providing such a course are advised to buy at least one copy of the book. Also, purchase of the book can be a good investment for individuals connected with the sample surveying of population.