Post hoc power analysis: is it an informative and meaningful analysis?

General Psychiatry - Tập 32 Số 4 - Trang e100069 - 2019
Yiran Zhang1, Rita Hedo1, Anna C. Rivera2,3, Rudolph P. Rull3, Sabrina M. Richardson2,3, Xin Tu4
1Department of Family Medicine and Public Health, University of California System, Oakland, California, USA
2Leidos, San Diego, California, USA
3Naval Health Research Center, San Diego, California, USA
4Family Medicine and Public Health, University of California San Diego, La Jolla, CA 92093, USA

Tóm tắt

Power analysis is a key component for planning prospective studies such as clinical trials. However, some journals in biomedical and psychosocial sciences ask for power analysis for data already collected and analysed before accepting manuscripts for publication. In this report, post hoc power analysis for retrospective studies is examined and the informativeness of understanding the power for detecting significant effects of the results analysed, using the same data on which the power analysis is based, is scrutinised. Monte Carlo simulation is used to investigate the performance of posthoc power analysis.

Từ khóa


Tài liệu tham khảo

Cohen, 1988, Statistical power analysis for the behavioral sciences, 2

Heonig, 2001, The abuse of power: the pervasive fallacy of power calculations for data analysis, The American Statistician, 55, 19, 10.1198/000313001300339897

Kraemer, 2006, Caution regarding the use of pilot studies to guide power calculations for study proposals, Arch Gen Psychiatry, 63, 484, 10.1001/archpsyc.63.5.484

Levine, 2001, Post hoc power analysis: an idea whose time has passed?, Pharmacotherapy, 21, 405, 10.1592/phco.21.5.405.34503

Korn, 1990, Projection from previous studies, Control Clin Trials, 11, 67, 10.1016/0197-2456(90)90033-X

Korn, 1990, Projecting power from a previous study: maximum likelihood estimation, Am Stat, 44, 290, 10.1080/00031305.1990.10475742

Kowalski, 2007, Modern applied U statistics, 10.1002/9780470186466

Tu, 2004, Power analyses for longitudinal trials and other clustered designs, Stat Med, 23, 2799, 10.1002/sim.1869

Thông tin
Thông tin xuất bản

General Psychiatry

Tập 32 Số 4

e100069

Thông tin tác giả