学术探索
学术期刊
新闻热点
数据分析
智能评审

Confidence intervals are a more useful complement to nonsignificant tests than are power calculations

被引：204
作者：
Colegrave, N
Ruxton, GD
机构：
[1] Univ Edinburgh, Inst Cell Anim & Populat Biol, Ashworth Lab, Edinburgh EH9 3JT, Midlothian, Scotland
[2] Univ Glasgow, Div Environm & Evolutionary Biol IBLS, Glasgow G12 8QQ, Lanark, Scotland
来源：
BEHAVIORAL ECOLOGY | 2003年 / 14卷 / 03期
关键词：
D O I：
10.1093/beheco/14.3.446
中图分类号：
B84 [心理学]; C [社会科学总论]; Q98 [人类学];
学科分类号：
03 ; 0303 ; 030303 ; 04 ; 0402 ;
摘要：
[No abstract available]
引用
收藏
页码：446 / U4
页数：2
相关论文
共 5 条
[1]  
Cohen J., 1998, Statistical Power Analysis for the Behavioral Sciences, V2nd
[2]   The abuse of power: The pervasive fallacy of power calculations for data analysis [J].
Hoenig, JM ;
Heisey, DM .
AMERICAN STATISTICIAN, 2001, 55 (01) :19-24
[3]   The insignificance of statistical significance testing [J].
Johnson, DH .
JOURNAL OF WILDLIFE MANAGEMENT, 1999, 63 (03) :763-772
[4]  
Lipsey M.W., 1990, Design sensitivity: Statistical power for experimental research, V19
[5]   A COMPARISON OF THE 2 ONE-SIDED TESTS PROCEDURE AND THE POWER APPROACH FOR ASSESSING THE EQUIVALENCE OF AVERAGE BIOAVAILABILITY [J].
SCHUIRMANN, DJ .
JOURNAL OF PHARMACOKINETICS AND BIOPHARMACEUTICS, 1987, 15 (06) :657-680
1