作者aifor3 (ai)
看板Statistics
标题[问题] 统计描述看不懂 500p徵好心人
时间Wed Dec 9 13:36:00 2015
Power analysis determined that a sample size of at least 20 subjects
per group would achieve 80% power to detect a 2-point change in NRS
scores between treatment groups, with estimated group SDs of 2 for a
2-sided test with a significance level α of .05 (PASS-NCCS, 2011,
Kaysville, UT
这段是说有检定力的分析需要每个group需要20个受试者吗
NRS SCORE是临床评断疼痛的指标
但剩下看不太懂
achieve 80% power to detect a 2-point change
with estimated group SDs of 2 for a
2-sided test with a significance level α of .05
谢谢版上的大家了= =
--
※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 106.1.25.182
※ 文章网址: https://webptt.com/cn.aspx?n=bbs/Statistics/M.1449639363.A.4A3.html
1F：→ allen1985: 你知道power的定义吗 12/09 14:21
2F：推 sean50301: P(reject Ho | Ha is true) 12/09 14:48
3F：推 imaltar: 在power为0.8,alpha为0.05,2 side,假设标准差为2,NRS di 12/09 16:22
4F：→ imaltar: fference假设为2的情况下,估得每个group约需20人,sas 中 12/09 16:22
5F：→ imaltar: proc power可执行次过程 12/09 16:22
6F：推 imaltar: 这个样本估计有考虑到drop out rate吗？ 12/09 17:03