wind2599 比如: AB=data.frame(A=LETTERS,B=rnorm(26)); A1=AB[AB$B>0,] levels(A1$A) [1] "A" "B" "C" "D" "E" "F" "G" "H" "I" "J" "K" "L" "M" "N" "O" "P" "Q" "R" [19] "S" "T" "U" "V" "W" "X" "Y" "Z" > levels(AB$A) [1] "A" "B" "C" "D" "E" "F" "G" "H" "I" "J" "K" "L" "M" "N" "O" "P" "Q" "R" [19] "S" "T" "U" "V" "W" "X" "Y" "Z" 这是为什么啊？
wind2599 A=data.frame(a=rep(LETTERS,each=6),b=c(rep(1:26,3),rep(-2,78))) barplot(tapply(A$b,A$a,mean)) barplot(tapply(A[A$b>0,]$b,A[A$b>0,]$a,mean)) [attachment=226455,1497] [attachment=226455,1498]
wind2599 回复 第6楼 的 波波头一头：回复 第6楼 的 波波头一头：是的，如果在分类变量前加个factor就好了,barplot(tapply(A[A$b>0,]$b,factor(A[A$b>0,]$a),mean))，主要想知道原理是什么。
