【菜鸟入门R语言】独立t检验

t检验的目的是，比较两组数据有无显著性差异。

知识要点：

• 根据是否成对、是否符合零假设选择合适的检验方法
• t检验的统计学原理
• 数据保存为文件 write.csv()
• t.test()官方文档

我的数据是两种环境下的植物体的生理性状，不是成对的，所以选择独立t检验，并不需要检验方差同质性(homogeneity of variance) 的前提假设，比如Levene’s test，别问我为什么说这么常识性的话。

mtcars <- read.csv(file.choose())
res<-c()
res[1]<-t.test(mtcars$D1_1,mtcars$D2_1,paired = FALSE)$p.value
res[2]<-t.test(mtcars$D1_2,mtcars$D2_2,paired = FALSE)$p.value
res[3]<-t.test(mtcars$D1_3,mtcars$D2_3,paired = FALSE)$p.value
res[4]<-t.test(mtcars$D1_4,mtcars$D2_4,paired = FALSE)$p.value
res[5]<-t.test(mtcars$F1_1,mtcars$F2_1,paired = FALSE)$p.value
res[6]<-t.test(mtcars$F1_2,mtcars$F2_2,paired = FALSE)$p.value
res[7]<-t.test(mtcars$F1_3,mtcars$F2_3,paired = FALSE)$p.value
res[8]<-t.test(mtcars$F1_4,mtcars$F2_4,paired = FALSE)$p.value
res[9]<-t.test(mtcars$L1_1,mtcars$L2_1,paired = FALSE)$p.value
res[10]<-t.test(mtcars$L1_2,mtcars$L2_2,paired = FALSE)$p.value
res[11]<-t.test(mtcars$L1_3,mtcars$L2_3,paired = FALSE)$p.value
res[12]<-t.test(mtcars$L1_4,mtcars$L2_4,paired = FALSE)$p.value
res[13]<-t.test(mtcars$Y1_1,mtcars$Y2_1,paired = FALSE)$p.value
res[14]<-t.test(mtcars$Y1_2,mtcars$Y2_2,paired = FALSE)$p.value
res[15]<-t.test(mtcars$Y1_3,mtcars$Y2_3,paired = FALSE)$p.value
res[16]<-t.test(mtcars$Y1_4,mtcars$Y2_4,paired = FALSE)$p.value
res[17]<-t.test(mtcars$M1_1,mtcars$M2_1,paired = FALSE)$p.value
res[18]<-t.test(mtcars$M1_2,mtcars$M2_2,paired = FALSE)$p.value
res[19]<-t.test(mtcars$M1_3,mtcars$M2_3,paired = FALSE)$p.value
res[20]<-t.test(mtcars$M1_4,mtcars$M2_4,paired = FALSE)$p.value
res[21]<-t.test(mtcars$N1_1,mtcars$N2_1,paired = FALSE)$p.value
res[22]<-t.test(mtcars$N1_2,mtcars$N2_2,paired = FALSE)$p.value
res[23]<-t.test(mtcars$N1_3,mtcars$N2_3,paired = FALSE)$p.value
res[24]<-t.test(mtcars$N1_4,mtcars$N2_4,paired = FALSE)$p.value
res[25]<-t.test(mtcars$T1_1,mtcars$T2_1,paired = FALSE)$p.value
res[26]<-t.test(mtcars$T1_2,mtcars$T2_2,paired = FALSE)$p.value
res[27]<-t.test(mtcars$T1_3,mtcars$T2_3,paired = FALSE)$p.value
res[28]<-t.test(mtcars$T1_4,mtcars$T2_4,paired = FALSE)$p.value

res_m<-matrix(res,4,7);
write.csv(res_m,"d:/doing/res.csv");