02. t检验与箱线图

两个样本差异比较的方法。

比如两个向量集：

> head(boy)
[1] 177.97 171.89 170.29 158.69 177.36 182.50
> head(girl)
[1] 151.11 158.74 155.47 158.62 176.03 153.65

直接t 检验好了：

> t.test(boy, girl)
  Welch Two Sample t-test
data:  boy and girl
t = 6.3941, df = 97.71, p-value = 5.552e-09
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
  7.506641 14.263359
sample estimates:
mean of x mean of y 
 170.0252  159.1402

如果想要检测单边的，比如 x 是否显著大于 y，加个参数 alternative（greater 与 less） ：

> t.test(boy, girl, alternative = 'greater')
  Welch Two Sample t-test
data:  boy and girl
t = 6.3941, df = 97.71, p-value = 2.776e-09
alternative hypothesis: true difference in means is greater than 0
95 percent confidence interval:
 8.058101      Inf
sample estimates:
mean of x mean of y 
 170.0252  159.1402

如果想要做一般的t 检验（student-t-test）：

> t.test(boy, girl, alternative = 'greater', var.equal= T)
  Two Sample t-test
data:  boy and girl
t = 6.3941, df = 98, p-value = 2.753e-09
alternative hypothesis: true difference in means is greater than 0
95 percent confidence interval:
 8.058181      Inf
sample estimates:
mean of x mean of y 
 170.0252  159.1402

比较两个组之间的差异，箱线图是最好的可视化方法之一了：

ggplot(data = clinical_info_final) +
  geom_boxplot(mapping = aes(x = condition_brca1, y = Age_at_surgery, fill = condition_brca1))
ggplot(data = clinical_info_final) +
  geom_boxplot(mapping = aes(x = condition_brca2, y = Age_at_surgery, fill = condition_brca2))
condition_brca2