如何在R的ggplot2中按指定类别排序李克特量表图

问题 #

我在R中有一个名为df的李克特量表响应数据框，已经用ggplot2绘制了对应的李克特图。现在需要调整图表排序：将图中的item先按「Very Dissatisfied」类别占比从低到高排序，若该占比相同则按「Dissatisfied」类别占比从低到高排序。

数据预览 #

df
# A tibble: 90 × 3
# Groups:   item [18]
   item  Response          Percentage
   <chr> <fct>                  <dbl>
 1 A     Very Dissatisfied       25  
 2 A     Dissatisfied            25  
 3 A     Average                 33.3
 4 A     Satisfied               11.1
 5 A     Very Satisfied          55.6
 6 B     Very Dissatisfied       25  
 7 B     Dissatisfied            25  
 8 B     Average                 44.4
 9 B     Satisfied               25  
10 B     Very Satisfied          55.6
# ℹ 80 more rows
# ℹ Use `print(n = ...)` to see more rows

现有绘图代码 #

response_mapping <- c("Very Dissatisfied" = 1,
                      "Dissatisfied" = 2,
                      "Average" = 3,
                      "Satisfied" = 4,
                      "Very Satisfied" = 5)

# Apply the mapping and calculate the sign
data_f_sum <- df %>% 
  ungroup() %>% 
  mutate(res.sgn = sign(response_mapping[as.character(Response)] - 3)) %>% 
  summarise(sum.prcnt = sum(Percentage),
            .by = c(item, res.sgn))

data_f_sum
likert_levels =  c("Very Dissatisfied", 
                   "Dissatisfied" ,
                   "Average" ,
                   "Satisfied", 
                   "Very Satisfied")

df = df%>%
  mutate(Response = factor(Response , levels = likert_levels))


ggplot(data = df, 
       aes(Percentage, item, fill = Response)) +
  geom_col(position = position_likert()) +
  scale_x_continuous(labels = ggstats::label_percent_abs()) +
  geom_label(data = data_f_sum,
             aes(label = sprintf("%.1f", sum.prcnt), y = item, x = res.sgn),
             alpha = 0.3, inherit.aes = FALSE) +
  coord_cartesian(xlim = c(-1, 1)) +
  scale_fill_brewer(type = "div", palette = "RdYlGn") +
  theme_bw()+ 
  theme(legend.position = "bottom") 

数据结构 #

structure(list(item = c("A", "A", "A", "A", "A", "B", "B", "B", 
"B", "B", "C", "C", "C", "C", "C", "D", "D", "D", "D", "D", "E", 
"E", "E", "E", "E", "F", "F", "F", "F", "F", "G", "G", "G", "G", 
"G", "H", "H", "H", "H", "H", "I", "I", "I", "I", "I", "J", "J", 
"J", "J", "J", "K", "K", "K", "K", "K", "L", "L", "L", "L", "L", 
"M", "M", "M", "M", "M", "N", "N", "N", "N", "N", "O", "O", "O", 
"O", "O", "P", "P", "P", "P", "P", "Q", "Q", "Q", "Q", "Q", "R", 
"R", "R", "R", "R"), Response = structure(c(1L, 2L, 3L, 4L, 5L, 
1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 
2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 
3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 
4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 
5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 
1L, 2L, 3L, 4L, 5L), levels = c("Very Dissatisfied", "Dissatisfied", 
"Average", "Satisfied", "Very Satisfied"), class = "factor"), 
    Percentage = c(25, 25, 33.3, 11.1, 55.6, 25, 25, 44.4, 25, 
    55.6, 25, 25, 22.2, 33.3, 44.4, 25, 25, 33.3, 11.1, 55.6, 
    25, 22.2, 11.1, 11.1, 55.6, 25, 25, 44.4, 11.1, 44.4, 25, 
    25, 11.1, 33.3, 55.6, 25, 25, 33.3, 22.2, 44.4, 25, 25, 11.1, 
    33.3, 55.6, 25, 25, 22.2, 22.2, 55.6, 25, 25, 11.1, 11.1, 
    77.8, 25, 25, 11.1, 33.3, 55.6, 25, 25, 33.3, 25, 66.7, 25, 
    25, 33.3, 11.1, 55.6, 25, 11.1, 25, 33.3, 55.6, 25, 25, 22.2, 
    22.2, 55.6, 25, 22.2, 22.2, 11.1, 44.4, 25, 11.1, 22.2, 11.1, 
    55.6)), class = c("grouped_df", "tbl_df", "tbl", "data.frame"
), row.names = c(NA, -90L), groups = structure(list(item = c("A", 
"B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", 
"O", "P", "Q", "R"), .rows = structure(list(1:5, 6:10, 11:15, 
    16:20, 21:25, 26:30, 31:35, 36:40, 41:45, 46:50, 51:55, 56:60, 
    61:65, 66:70, 71:75, 76:80, 81:85, 86:90), ptype = integer(0), class = c("vctrs_list_of", 
"vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -18L), .drop = TRUE))

解决方案 #

要实现指定的排序逻辑，核心是把item转换为有序因子，按照「Very Dissatisfied」占比升序、「Dissatisfied」占比升序的规则定义因子水平顺序。具体步骤如下：

1. 从原数据中提取每个item对应的「Very Dissatisfied」和「Dissatisfied」的占比
2. 按照排序规则对item进行排序，得到目标顺序
3. 将原数据中的item转换为因子，指定排序后的水平
4. 用修改后的数据绘图即可

修改后的完整代码 #

library(tidyverse)
library(ggplot2)
library(ggstats)

# 定义响应映射和李克特水平
response_mapping <- c("Very Dissatisfied" = 1,
                      "Dissatisfied" = 2,
                      "Average" = 3,
                      "Satisfied" = 4,
                      "Very Satisfied" = 5)

likert_levels =  c("Very Dissatisfied", 
                   "Dissatisfied" ,
                   "Average" ,
                   "Satisfied", 
                   "Very Satisfied")

# 处理数据：提取排序所需的占比，重新排序item为有序因子
df_sorted <- df %>%
  ungroup() %>%
  mutate(Response = factor(Response, levels = likert_levels)) %>%
  # 提取每个item的Very Dissatisfied和Dissatisfied占比
  pivot_wider(names_from = Response, values_from = Percentage) %>%
  # 按照指定规则排序：先Very Dissatisfied升序，再Dissatisfied升序
  arrange(`Very Dissatisfied`, `Dissatisfied`) %>%
  # 提取排序后的item顺序
  pull(item) %>%
  # 将原数据的item转换为因子，指定排序后的水平
  {mutate(df, item = factor(item, levels = .))} %>%
  # 重新分组（可选，保持原数据结构）
  group_by(item)

# 计算数据总和标签（原逻辑不变）
data_f_sum <- df_sorted %>% 
  ungroup() %>% 
  mutate(res.sgn = sign(response_mapping[as.character(Response)] - 3)) %>% 
  summarise(sum.prcnt = sum(Percentage),
            .by = c(item, res.sgn))

# 绘制排序后的李克特图
ggplot(data = df_sorted, 
       aes(Percentage, item, fill = Response)) +
  geom_col(position = position_likert()) +
  scale_x_continuous(labels = ggstats::label_percent_abs()) +
  geom_label(data = data_f_sum,
             aes(label = sprintf("%.1f", sum.prcnt), y = item, x = res.sgn),
             alpha = 0.3, inherit.aes = FALSE) +
  coord_cartesian(xlim = c(-1, 1)) +
  scale_fill_brewer(type = "div", palette = "RdYlGn") +
  theme_bw()+ 
  theme(legend.position = "bottom") 

关键部分解释 #

• 使用pivot_wider把每个item的各类响应占比转换为宽格式，方便提取「Very Dissatisfied」和「Dissatisfied」的数值
• arrange(Very Dissatisfied, Dissatisfied)实现了先按前者升序、再按后者升序的排序逻辑
• 通过factor(item, levels = .)把item转换为有序因子，ggplot会严格按照因子水平的顺序绘制y轴项目

内容的提问来源于stack exchange，提问作者Homer Jay Simpson