In the past, the emphasis of statistics in forestry, and other applied fields, has been on an assessment of statistical significance, or the probability that the null hypothesis will be rejected when it is true (i.e.., the probability of committing a "Type I error"). However, there is growing awareness (e.g., see Peterman 1990a, 1990b and Toft and Shea 1983) that researchers should also be concerned with the possibility that statistical methods may fail to reject a false null hypothesis (i.e., a "Type II error" might be committed). The statistical theory and methods by which this important issue can be examined are referred to as power analysis.
This handbook is intended as an introduction to power analysis. It contains basic definitions and a review of power theory for t-tests and ANOVA F-tests, both of which are widely used in the analysis of forestry data. Examples, with step-by-step instructions and SAS programs for performing the necessary calculations, are provided. The handbook also contains a discussion of the two primary applications of power analysis: experimental design (e.g., sample size calculations); and the interpretation of the results of statistical analyses using post hoc power analysis. Although this handbook does not cover tests for categorical data (e.g., tests for proportions or log-linear methods), the same general principles apply to those methods as well. A detailed exposition of power analysis for a broad collection of tests can be found in Cohen (1977).
Download Biometrics Information Handbook No. 2 (205 KB)
To view this document you need the current version of
Adobe Acrobat Reader, available free from the
Adobe Web Site.
Updated October 16, 2008