Psychiatric Research

In 1990, Dr David Streiner authored a paper called “Sample-Size and Power in Psychiatric Research” in The Canadian Journal of Psychiatry (The CJP).¹ This article was the first in a long series of popular tutorials on statistical methods published in The CJP during the next 20 years. Many articles in this series have had an enduring popularity, being both widely read and frequently referenced. For example, an article on factor analysis published in 1994² has been cited over 124 times since its publication. In this book, the entire series of articles has been collated and updated, making all of them easily and conveniently accessible.

As the title suggests, this is a book primarily intended for clinical researchers. It will also be of interest to nonresearchers seeking to hone their understanding of statistical procedures they encounter in their reading of the clinical research literature. It is not, however, an introductory textbook of statistical methods. The topics covered range from basic indices of central tendency (Chapter 1) to advanced topics, such as item response theory (Chapter 25). Later chapters provide practical advice on writing papers and grant proposals (Chapters 27 and 28).

What places this book apart from most in this genre is its readability and its practical, rather than technical, emphasis. The book has a clear writing style and places its emphasis on narrative, rather than symbolic, description (words rather than formulas). It is a surprisingly pleasant read. Far from being dumbed down, the discussion is interesting, lively, often humorous, and at times opinionated.

The title of the book refers to statistical perplexity. This may create a perception that the book is targeted toward readers who are puzzled by statistics. In fact, Dr Streiner views perplexity as a positive trait—one that stands in opposition to “blithe ignorance”^{Introduction, p ix} or a willingness to passively accept statistical claims or to use statistical techniques without understanding. Perplexity is taken as a sign of thoughtfulness and curiosity, a precursor to competence rather than a threat to it.

The updating of the older chapters means that none of them are dated. However, there is not a major emphasis on trying to cover all of the latest trends. The book is not intended to be an update, even though it has been updated. The focus is on the bread-and-butter statistical methods in psychiatric research.

The book has an informal and attractive appearance, and is free of production errors. It can be read cover-to-cover as a sweeping and thoughtful overview of important statistical issues. It will also be a useful reference source when one encounters unfamiliar methods or puzzling statistical results while reading the clinical research literature. This book is a useful antidote to the sense of perplexity that should arise in these situations.