R语言geom_jitter箱线图如何按Status字段设置两种不同点符号

问题原因 #

你现有代码的问题在于：

• 没有将Status字段映射到点的形状美学参数上，反而把shape固定为16，因此无法区分不同状态的点
• 你的Condition字段是字符类型，需要先转为有序因子才能正确在坐标轴上排序、绘制箱线图

步骤1：数据预处理 #

先将分类字段转为有序因子，方便后续绘图：

library(tidyverse)

# 构造示例数据（如果你已有Food数据可跳过这段）
Food <- tibble(
  Fruit = c("Apple","Mango","Apple","Apple","Mango","Mango","Mango","Apple","Apple","Apple","Mango","Mango","Apple"),
  Condition = c("Ripe","Unripe","Ripe","Unripe","Ripe","Unripe","Ripe","Ripe","Unripe","Unripe","Ripe","Ripe","Ripe"),
  Status = c("Yes","Maybe","Yes","Maybe","Maybe","Yes","Yes","Yes","Yes","Maybe","Yes","Yes","Yes")
)

# 将Condition转为有序因子，控制y轴排序
Food <- Food %>% 
  mutate(Condition = factor(Condition, levels = c("Unripe", "Ripe"), ordered = TRUE))

步骤2：修改绘图代码 #

核心是在geom_jitter的美学映射中绑定shape = Status，不要固定shape值：

ggplot(data = Food, aes(x = Fruit, y = Condition, color = Fruit)) +
  geom_boxplot()  +
  # 将点的形状映射到Status字段
  geom_jitter(aes(shape = Status), position = position_jitter(0.2), size = 3) +
  # 可选：自定义两种状态对应的点形状
  scale_shape_manual(values = c("Maybe" = 1, "Yes" = 16))

代码说明 #

• 把shape参数从geom_jitter的全局参数移到aes()内部，关联Status变量后ggplot会自动为不同状态的点分配不同形状
• scale_shape_manual可以自定义形状：上面的示例中Maybe对应空心圆圈、Yes对应实心圆点，你可以调整values里的数值更换形状
• 增加size=3是为了放大点方便区分形状，不需要可以直接删除

内容的提问来源于stack exchange，提问作者Ed_Gravy