在 ggplot2 中使用 stat="bin" 标记堆积条形图

Question

似乎 ggplot2 应该能够使用 geom_text 标记堆积条形图，并在 geom_bar 中对数据框的子集执行计数统计，而无需专门为标签重新创建新的数据框。这可能吗？

这是我的数据框的样子：

> head(df.m2)
Community        Categories               Indicators  Code Scores  Condition
1  Zaragoza         Landscape Landscape \n Composition f01.1      0   Marginal
2  Zaragoza         Landscape               Windbreaks f01.1      6 Acceptable
3  Zaragoza         Landscape        Field \n Location f01.1      6 Acceptable
4  Zaragoza         Landscape     Soil \n Conservation f01.1     12    Optimal
5  Zaragoza Farmer Management         Crop \n Rotation f01.1     12    Optimal
6  Zaragoza Farmer Management        Crop \n Varieties f01.1      6 Acceptable

这是我用来创建绘图的代码（注意用于汇总分类变量的 stat="bin"）：

p <- ggplot(df.m2, aes(Indicators, fill=Condition))
p + coord_cartesian(ylim=c(-.3,12.3)) +
geom_bar(stat="bin", colour="gray", alpha=.7) +
scale_fill_manual(values = c("green","yellow","red")) +
theme(axis.text.x = element_text(angle = 90,vjust= .5,hjust=1)) +
labs(x = "Farmers' Indicators", y = "Number of Rankings by Farmers") +
facet_grid(Community ~ Categories, scales = "free_x")

我想包含生成的情节，但我的声誉等级不允许。我想要的是将计数打印并集中在每个相应的条上。

Answer 1

我会为此使用data.table 包。它仍然不适用于仅ggplot 功能。通常最好先做数据匹配，然后使用ggplot。

require(data.table)
dt.m2 <- data.table(df.m2)
dt.m2[, count:=.N, by=list(Indicators, Categories)] # create the count index 
# almost the same as before
p <- ggplot(dt.m2, aes(Indicators, fill=Condition))
p + coord_cartesian(ylim=c(-.3,12.3)) +
  geom_bar(stat="bin", colour="gray", alpha=.7) +
  geom_text(aes(y=count+1, label=paste0("n=", count))) + # add text above bar
  # you may have to change the y position for this to look good
  scale_fill_manual(values = c("green","yellow","red")) +
  theme(axis.text.x = element_text(angle = 90,vjust= .5,hjust=1)) +
  labs(x = "Farmers' Indicators", y = "Number of Rankings by Farmers") +
  facet_grid(Community ~ Categories, scales = "free_x")