Limits of retrospective power analysis

被引：108

作者：

Gerard, PD

Smith, DR

Weerakkody, G

机构：

[1] Mississippi State Univ, Expt Stat Unit, Mississippi State, MS 39762 USA

[2] US Geol Survey, Biol Resources Div, Leetown Sci Ctr, Kearneysville, WV 25430 USA

[3] Mississippi State Univ, Dept Math & Stat, Mississippi State, MS 39762 USA

来源：

JOURNAL OF WILDLIFE MANAGEMENT | 1998年 / 62卷 / 02期

关键词：

interpretation of results; noncentrality parameter; statistical power; study design;

D O I：

10.2307/3802357

中图分类号：

Q14 [生态学（生物生态学）];

学科分类号：

071012 ; 0713 ;

摘要：

Power analysis after study completion has been suggested to interpret study results. We present 3 methods of estimating power and discuss their limitations. We use simulation studies to show that estimated power can be biased, extremely variable, and severely bounded. We endorse the practice of computing power to detect a biologically meaningful difference as a tool for study planning but suggest that calculation of confidence intervals on the parameters of interest is the appropriate way to gauge the strength and biological meaning of study results.

引用

页码：801 / 807

页数：7

共 24 条

[1] Effects of research activities on nest predation in Arctic-nesting geese [J].