作者bubble0410 (疯狂的阳光)
看板Statistics
标题[问题] effect
时间Thu May 31 02:57:55 2018
1.If a treatment has a very small effect, what is a likely outcome for a hypothesis test evaluating the treatment?
A. A Type I error
B. A Type II error
C. Correctly reject the null hypothesis
D. Correctly fail to reject the null hypothesis
E. Answer cannot be determined from the information provided
2.Under what circumstances can a very small treatment effect be statistically significant?
A. If the sample size big and the sample variance is small.
B. If the sample size and the sample variance are both big.
C. If the sample size is small and the sample variance is big.
D. If the sample size and the sample variance are both small.
E. None of A-D is true.
-
最近在自学社统跟练习题目，有2题我看不懂它题目本身的意思
不知道我是漏念了什麽，还是只是看不懂英文……
想请问这两题是在问什麽？
我想不通a very small effect是什麽意思，要从哪查起
1.「a treatment has a very small effect」
2.「a very small treatment effect」
--
※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 39.8.232.92
※ 文章网址: https://webptt.com/cn.aspx?n=bbs/Statistics/M.1527706678.A.6C1.html
1F：推 lalial: a very small effect "size" 05/31 03:52
2F：推 lalial: use a t test formula may help you understand the quest 05/31 03:56
3F：→ lalial: ions 05/31 03:56
4F：→ andrew43: 效果量很小表示虚无假设不对，但除非power够大（如样本 05/31 07:50
5F：→ andrew43: 数很大、组内变异很小）否则很难拒绝它。 05/31 07:50
6F：推 andrew43: 你可以查 statistical effect size 05/31 08:06
7F：→ yhliu: 当 treatment 确实有 a very small effect 时,做 hypothese 06/01 06:06
8F：→ yhliu: test 将很不容易 reject null hypothesis, 除非样本数非常 06/01 06:10
9F：→ yhliu: 大, 或允许很大的 type I error 机率. 也就是说,犯 type II 06/01 06:12
10F：→ yhliu: error 的机会是可能比较大的. 06/01 06:14
11F：→ yhliu: 承上, 若 true treatment 是 very small 而检定结果显示 be 06/01 06:16
12F：→ yhliu: statistical significant, 除了运气便然, 就是 sample size 06/01 06:18
13F：→ yhliu: 相对於 sample variance 来说很大, 以至於 standard error 06/01 06:20
14F：→ yhliu: 很小, 也就是说, sample size is big 以及 sample variance 06/01 06:22
15F：→ yhliu: is small. 06/01 06:22