用于追踪组成的堆积条形图

Question

我需要在堆积的条形图中绘制一组数据。数据来自具有三种不同成分（M1，M1BIS，M2，M2BIS，M3和M3BIS）的六个样品，并且已经在三个不同点（表面（S），中心（M）和底部（G））中取得。和两次（0＆amp; 32天）。

我需要在堆积的条形图中绘制这些数据（每个样本和时间一个条）以查看三个点中的盐浓度，并且所有条形必须具有相同的长度（以百分比表示）

    rdata <-
structure(list(Sample = c("M1", "M1", "M1", "M2", "M2", "M2", 
"M3", "M3", "M3", "M1 BIS", "M1 BIS", "M1 BIS", "M2 BIS", "M2 BIS", 
"M2 BIS", "M3 BIS", "M3 BIS", "M3 BIS", "M1", "M1", "M1", "M2", 
"M2", "M2", "M3", "M3", "M3", "M1 BIS", "M1 BIS", "M1 BIS", "M2 BIS", 
"M2 BIS", "M2 BIS", "M3 BIS", "M3 BIS", "M3 BIS"), Composition = c(1L, 
1L, 1L, 2L, 2L, 2L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 3L, 3L, 
3L, 1L, 1L, 1L, 2L, 2L, 2L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 
3L, 3L, 3L), t = c(0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 32L, 32L, 32L, 32L, 32L, 32L, 32L, 
32L, 32L, 32L, 32L, 32L, 32L, 32L, 32L, 32L, 32L, 32L), Value = c("1,72", 
"1,3", "4,12", "3,7", "2,88", "3,52", "1,38", "2,57", "6,89", 
"1,28", "0,85", "4,15", "0,71", "1,47", "4,77", "1,17", "1,37", 
"4,76", "4,16", "3,65", "3,39", "3,62", "4,04", "4,37", "2,77", 
"2,6", "3,23", "3,1", "3,34", "3,64", "3,68", "2,82", "3,12", 
"4,03", "3,87", "4,27"), Point = c("G", "M", "S", "G", "M", "S", 
"G", "M", "S", "G", "M", "S", "G", "M", "S", "G", "M", "S", "S", 
"M", "G", "S", "M", "G", "S", "M", "G", "S", "M", "G", "S", "M", 
"G", "S", "M", "G")), .Names = c("Sample", "Composition", "t", 
"Value", "Point"), row.names = c(NA, -36L), class = "data.frame")

Answer 1

我希望这样的事情。

library(reshape2)
rdata$Value <- gsub(",", ".", rdata$Value)
rdata$Value <- as.numeric(rdata$Value)

rdataWide <- dcast(data=rdata, formula = Point ~ Sample + t, value.var = "Value")
rdataWide[,-1] <- t(t(rdataWide[, -1])/colSums(rdataWide[, -1]))

par(xpd=T)
barplot(as.matrix(rdataWide[, -1]), col=2:4, space = 0.25, width=0.5, las=2)
legend(x = 0.25, y=1.25, legend = rdataWide[,1], col = 2:4, pch = 15, bty = "n", ncol = 3)

编辑：

它应该是par(xpd=T)而不是ppar(xpd=T)，这允许在绘图空间之外渲染。设置为false时，图例会被剪裁。

我还添加了library(reshape2)电话。我为疏忽道歉。