Bausell R. Barker, Li Yu-Fang. Power Analysis for Experimental Research: A Practical Guide for the Biological, Medical and Social Sciences

Cambridge: Cambridge University Press, 2002. — 363 p.Power analysis is an essential tool for determining whether a statistically significant result can be expected in a scientific experiment prior to the experiment being performed. Many funding agencies and Institutional Review Boards now require power analyses to be carried out before they will approve experiments, particularly where they involve the use of human subjects. This comprehensive, yet accessible, book provides practicing researchers with step-by-step instructions for conducting power/sample size analyses, assuming only basic prior knowledge of summary statistics and the normal distribution. It contains a unified approach to statistical power analysis, with numerous easy-to-use tables to guide the reader without the need for further calculations or statistical expertise. This will be an indispensable text for researchers and graduates in the medical and biological sciences needing to apply power analysis in the design of their experiments.Introduction page
The conceptual underpinnings of statistical power
Strategies for increasing statistical power
General guidelines for conducting a power analysis
The t-test for independent samples
The paired t-test
One-way between subjects analysis of variance
One-way between subjects analysis of covariance
One-way repeated measures analysis of variance
Interaction effects for factorial analysis of variance
Power analysis for more complex designs
Other power analytic issues and resources for addressing them
Technical appendix