PsychStatsWiki

本页面只读。您可以查看源文件，但不能更改它。如果您觉得这是系统错误，请联系管理员。
====== 相关样本t检验的效应量、置信区间和效力 effect size，confidence interval and power for relevant samples ======
===== 一、相关样本t检验的效应量 =====
1.**效应量**（effect size）：表示两个分布的重叠程度，它反映研究中处理效应的大小。效应量不受样本量影响。

Indicates the degree of overlap between the two distributions, which reflects the size of the treatment effect in the study. The effect size is not affected by the sample size.

2.相关样本t检验的效应量＝差异样本均值/差异样本方差

Effect size for correlated samples t-test = difference sample mean/difference sample variance

即Cohen's d = ES = {{::相关样本t检验效应量公式.png?30|}}

===== 二、相关样本t检验的置信区间=====
两个相关样本差异的95%置信区间为：

The 95% confidence interval for the difference between the two relevant samples is：\\
{{ ::相关样本t检验置信区间.png?200 }}

===== 三、相关样本t检验的效力 =====
研究结果的统计效力、效应大小、样本容量以及显著性水平相互联系。当显著性水平为0.05时，统计效力与效力大小、样本容量的关系如表所示。

The statistical power of the findings, effect size, sample size, and significance level are interrelated. When the level of significance is 0.05, the relationship between statistical validity and effect size and sample size is shown in the table.（Upper: two-tailed test; lower: one-tailed test）

{{ ::相关样本统计效力与效力大小_样本容量的关系表.png?700 }}