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。
==== 1. 独立样本t检验的效应量 (Effect size of the independent sample t test)====
效应量(ES)的大小代表两个总体分布的重叠程度。在独立样本t检验中，效应大小的计算公式如下：

The size of the effect size represents the extent to which the two population distributions overlap. In the independent sample t test, the formula for calculating the effect size is as follows:

{{ ::独立样本t检验的效应.jpg?nolink&150 |}}

其中X<sub>1</sub>bar和X<sub>2</sub>bar分别代表两个独立样本的均值，而S<sub>p</sub>代表的是对总体标准差的估计值。

Where X<sub>1</sub>bar and X<sub>2</sub>bar represent the mean of two independent samples, and S<sub>p</sub> represents the estimate of the standard deviation of the population.

ES越小，代表重叠程度越大，效应越小；ES越大，重叠程度越小，效应越大。

The smaller the ES, the greater the overlap and the smaller the effect. While the larger the ES, the smaller the overlap and the larger the effect.



----

==== 2. 独立样本t检验的置信区间 (Confidence interval of the independent sample t test)====

独立样本t检验的置信水平为1-α的置信区间为：

The confidence interval for the independent sample t test with a confidence level of 1-α is

{{ ::独立样本t检验的置信区间.jpg?nolink&250 |}}

其中S<sub>ω</sub>是对总体标准差的估计值。

Where S<sub>ω</sub> represents the estimate of the standard diavation of the population.

----
==== 3. 独立样本t检验的效力 (Power of the independent sample t test)====
效力的大小和显著性水平、样本容量和检验方向有关。

The magnitude of efficacy is related to the level of significance, sample size, and test direction.

  * 样本量越大，效力越高；
     The larger the sample size, the higher the power.


  * 效应越大，效力越高；
     The greater the effect, the higher the power.


  * 相同条件下双尾检验的效力比单尾检验效力高。
     The power of the two-tail test is higher than that of the single-tail test under the same conditions.