Stata：对同一变量的两个减法进行配对t检验答案

【问题标题】：Stata: Perform a paired ttest on two substracts of the same variableStata：对同一变量的两个减法进行配对t检验
【发布时间】：2022-12-01 10:00:29
【问题描述】：

我有一个看起来像这样的数据集：

var1	var2	var3	attribute
1	0.93	0.88	1
1	0.76	0.20	1
1	0.40	0.18	0
0	0.34	0.91	1
0	0.09	0.51	0
...	...	...	...

对于每个变量（变种1,变量2,变量3等），我需要在测试时比较其平均值属性= 1 vs 整个变量的平均值。我怎样才能做到这一点？

【问题讨论】：

标签： for-loop foreach stata t-test

【解决方案1】：

免责声明：我不知道这在统计上是否正确。

input var1 attribute
0.93 1
0.76 1
0.40 0
0.34 1
0.09 0
end

gen var2 = var1 if attribute == 1

mean var1
mean var2

ttesti 5 0.504 0.1510166 3 0.6766667 0.1753409 // using output from -mean- (see -help ttest-)

Two-sample t test with equal variances
------------------------------------------------------------------------------
         |     Obs        Mean    Std. err.   Std. dev.   [95% conf. interval]
---------+--------------------------------------------------------------------
       x |       5        .504    .0675367    .1510166    .3164881    .6915119
       y |       3    .6766667    .1012331    .1753409    .2410958    1.112238
---------+--------------------------------------------------------------------
Combined |       8      .56875    .0610348    .1726325    .4244257    .7130744
---------+--------------------------------------------------------------------
    diff |           -.1726667    .1165096               -.4577554     .112422
------------------------------------------------------------------------------
    diff = mean(x) - mean(y)                                      t =  -1.4820
H0: diff = 0                                     Degrees of freedom =        6

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0944         Pr(|T| > |t|) = 0.1889          Pr(T > t) = 0.9056

【讨论】：