R-ggplot2：如何将柱状图stop_out变量置于y轴负侧、stop_in保留在正侧

你只需要在数据整理步骤中将stop_out对应的数值转为负数，即可实现下车人数展示在y轴负半轴、上车人数保留在正半轴的效果，也可以额外调整y轴刻度标签让负半轴显示绝对值，更符合业务阅读习惯。

修改后完整代码如下：

library(tidyr)
library(dplyr)
library(ggplot2)

tramway_df <- data.frame(
  stop_order = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17),
  stop_id = c(9943, 5041, 5361, 2902, 2903, 5872, 5802, 5803, 5804, 5805, 
               5710, 5711, 5712, 1643, 5714, 521, 529),
  stop_name = c("DA VINCI", "DA VINCI", "BORDET STATION", "VAN CUTSEM", "FONSON", "PAIX", 
            "TILLEUL", "HELMET", "FOYER SCHAERBEEK.", "WAELHEM", "VERBOEKHOVEN", "PAVILLON",
            "RUBENS", "LIEDTS", "THOMAS", "GARE DU NORD", "ROGIER"),
  stop_in = c(27, 102, 153, 141, 190, 317, 379, 438, 328, 214, 278, 298, 154, 144, 162, 185, 0),
  stop_out = c(0, 7, 3, 10, 22, 37, 81, 122, 143, 168, 291, 230, 149, 301, 412, 707, 821),
  stop_charge = c(27, 121, 271, 402, 571, 851, 1149, 1465, 1650, 1696, 1682, 1745, 1750, 1592, 1342, 821, 0)
)

tramway_df %>% 
  gather(movement, value, stop_in:stop_out) %>% 
  mutate(stop_id = factor(stop_id, levels = unique(stop_id)),
         # 核心修改：将下车人数转为负值，自动渲染到y轴负半轴
         value = ifelse(movement == "stop_out", -value, value)) %>% 
ggplot(aes(x=stop_id)) +
  geom_col(aes(y=value, fill = movement), width = 0.7) +
  geom_line(aes(y=stop_charge, group=1), color = "black") + 
  scale_x_discrete(breaks = tramway_df$stop_id,
                   labels = tramway_df$stop_name) +
  # 优化：y轴刻度显示绝对值，避免负号误导
  scale_y_continuous(labels = abs, name = "人数") +
  # 优化：调整图例名称，适配中文阅读
  scale_fill_discrete(labels = c("上车人数", "下车人数"), name = "客流类型") +
  # 优化：x轴标签倾斜避免重叠
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

修改说明 #

• 核心逻辑是通过条件判断将stop_out的数值转为负数，ggplot会自动把负值柱状图绘制在y轴负半轴
• 如果需要上下车的柱子在x轴同一位置左右并列，只需把geom_col的参数改回position = "dodge"即可
• 后面的y轴、图例、主题调整都是优化项，如果不需要可以直接删除，不影响核心效果

内容的提问来源于stack exchange，提问作者salwa