Statistics you can get excited about

Since 1995, graduate students have hailed "Reading and Understanding Multivariate Statistics" as the best source for grasping the expanding variety of statistical techniques.

This year, there's more help to come for students and psychologists in general who read scientific research articles. Building on the success of their first book, Laurence G. Grimm, PhD, and Paul R. Yarnold, PhD, have just released "Reading and Understanding MORE Multivariate Statistics" (APA Books).

With this 430-page volume, the editors have responded to reader requests for the same accessible approach to today's newer multivariate statistics--the umbrella term used to describe a collection of analytic techniques that allow researchers to conduct simultaneous analyses of multiple measures.

"MORE" untangles the use of popular analyses, such as:

• Cluster analysis.

• Q-technique factor analysis.

• Canonical correlation analysis.

• Repeated measures analyses.

• Structural equation models.

• Survival analysis.

In addition, the book discusses topics that do not fall under the rubric of multivariate statistics, such as reliability, validity and measurement.

The book's goal is not to teach the reader how to perform multivariate statistics: Rather, it seeks to help readers understand the purpose and presentation of multivariate techniques, and to teach foundational aspects of reliability and validity. As with its predecessor, each chapter of "MORE" describes the research questions for which a statistical method is most appropriate, the underlying assumptions and rationale of the analysis and the logic behind interpreting the results--all written in jargon-free language and with easy-to-follow examples.

Grimm hatched the idea for the first book when he was writing an undergraduate statistics text.

"It occurred to me that the mainstream statistics books were written for classroom use at the undergraduate and graduate levels, but there were no books for the consumer of research that is published in scientific journals," Grimm says.

He and Yarnold pounced on the idea of writing such a book. Initially, they thought their main audience would be middle-aged professors who felt left behind since their graduate training in statistics but were too busy or embarrassed to take a graduate course in their own departments.

"In other words," says Grimm, "I wanted to read the book myself."

What they found, however, was that graduate students were also using the book to help them understand the required readings in their main statistics text.

After hearing, over and over again, "The book got me through graduate school, but I wish you would have included this or that topic," says Grimm, he and Yarnold developed "MORE," with all new topics.

"MORE" was in preparation for three years, "which is about the time it takes obsessive-compulsive authors to meet their own standards of excellence," Grimm says.