ENVIRONMENTAL IMPACT ASSESSMENT AND MONITORING: THE ROLE OF STATISTICAL POWER ANALYSIS

摘要

Statistical power is an important part of traditional hypothesis testing, yet, until recently, it has been given relatively little attention in this field .It is recommended that practitioners use power analysis to determine the sample size required by quantitative impact assessments, explore the effect of different estimates of sample variance and important effect size on the sample size required, and not take action as if there is `no effect' when a test fails to reject the zero hypotheses ,unless the power of the test is high. Managers should encourage research into what constitutes an important effect, take steps to ensure study designs for impact assessments are evaluated a periori, and include power analysis in standard protocols for designing quantitative impact assessments.