用ggplot2订购刻面堆积的条形图

Question

我有一个如下所示的数据集，我正在尝试获得一个有序的多面堆叠条形图。 我已经看了一些关于SO的答案，特别是这个问题来提出我的情节，但我不确定为什么这不适用于我堆叠的条形图的一个特定的条形图

structure(list(reg = c("J", "J", "J", "J", "J", "J", "J", "J", 
"KA", "KA", "KA", "KA", "KA", "KA", "KA", "KA", "KA", "KA", "RI", 
"RI", "RI", "RI", "SU", "SU", "SU", "SU", "SU", "SU", "SA", "SA", 
"SA", "SA", "SA", "SA", "SA", "SA", "SA", "SA", "SA", "SA", "SA", 
"SA", "SA", "SA"), org = structure(c(1L, 1L, 3L, 3L, 4L, 4L, 
2L, 2L, 1L, 1L, 3L, 3L, 4L, 4L, 2L, 2L, 8L, 8L, 2L, 2L, 1L, 1L, 
1L, 1L, 2L, 2L, 5L, 5L, 1L, 1L, 3L, 3L, 4L, 4L, 2L, 2L, 6L, 6L, 
5L, 8L, 7L, 7L, 9L, 9L), .Label = c("WR", "MS", "SM", "AA", "AAL", 
"PH", "P3", "SD", "HS"), class = "factor"), level = structure(c(2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 1L), .Label = c("E", 
"R"), class = "factor"), share = c(6.94, 1.51, 3.01, 0.42, 1.58, 
0.24, 2.85, 0.19, 4.7, 0.5, 2.16, 5.32, 1.08, 1.26, 0.82, 1.11, 
1.8, 1.73, 2, 3.52, 5, 2, 1.29, 0.38, 0.72, 0.46, 0.44, 2.27, 
15.79, 13.74, 5.12, 16.21, 6.54, 11.16, 8.19, 10.91, 4.71, 5.32, 
0.45, 0.14, 1.03, 0.33, 5.04, 3.03), reg_l = c("J_R", "J_E", 
"J_R", "J_E", "J_R", "J_E", "J_R", "J_E", "KA_R", "KA_E", "KA_R", 
"KA_E", "KA_R", "KA_E", "KA_R", "KA_E", "KA_R", "KA_E", "RI_R", 
"RI_E", "RI_R", "RI_E", "SU_R", "SU_E", "SU_R", "SU_E", "SU_R", 
"SU_E", "SA_R", "SA_E", "SA_R", "SA_E", "SA_R", "SA_E", "SA_R", 
"SA_E", "SA_R", "SA_E", "SA_E", "SA_E", "SA_R", "SA_E", "SA_R", 
"SA_E")), row.names = c(NA, -44L), class = c("tbl_df", "tbl", 
"data.frame"))

这是我现在的代码

data$level <- as.factor(as.character(data$level))
data$reg_l <- with(data, paste(reg,level,sep = "_"))

data$org <- factor(data$org,levels=c("WR","MS","SM","AA","AAL","PH","P3","SD","HS"))

# plot
reg_plot <- ggplot(data, aes(x = reorder(reg_l,share),y = share,group=org)) +
  coord_flip()+
  facet_wrap(level~.,ncol = 1,scales="free") +
  geom_bar(stat="identity",aes(fill=org),colour=NA,width=0.75) +
  scale_x_discrete(breaks = data$reg_l, labels = data$reg) +
  xlab("\n") + 
  ylab("\n") +
  scale_y_continuous(limits = c(0,100)) +
  ggtitle("\n")

reg_plot

下面的图片是我现在得到的情节，但由于某种原因， RI似乎条不遵循的顺序high到low值

任何有关如何解决这个问题的建议都将受到高度赞赏

Answer 1

在使用聚合函数sum时，使用reorder似乎不尊重facet_wrap 。 使用forcats::fct_reorder()为我工作：

reg_plot <- ggplot(data, aes(x = forcats::fct_reorder(reg_l, share, .fun = sum),y = share,group=org)) +
  coord_flip()+
  facet_wrap(level~.,ncol = 1,scales="free") +
  geom_bar(stat="identity",aes(fill=org),colour=NA,width=0.75) +
  scale_x_discrete(breaks = data$reg_l, labels = data$reg) +
  xlab("\n") + 
  ylab("\n") +
  scale_y_continuous(limits = c(0,100)) +
  ggtitle("\n")

reg_plot