如何解决ggplot2选择性标注特定散点时的报错问题？

数据样例 #

"Country","Continent","GDP","HCI"
Afghanistan,Asia,1991,0.40
Albania,Europe,19566,0.63
Algeria,Africa,13682,0.53
Angola,Africa,7077,0.36
Antigua and Barbuda,Americas,25449,0.60
Argentina,Americas,26506,0.60
...

报错代码 #

ctry_inf <- read.csv("Paises_PIB_ICH.csv")

ctry_inf_selec <- ctry_inf[ctry_inf$Continent == "Asia" | ctry_inf$Continent == "Africa",  ]

ggplot(ctry_inf_selec, aes(x=GDP, y=HCI)) + geom_point() + 
  scale_x_continuous(trans='log10') + 
  geom_text(aes(label = 
                  ifelse(ctry_inf$Country == "Namibia" | ctry_inf$Country == "United Arab Emirates", 
                         as.character(Country), "")))

报错信息 #

> ggplot(ctry_inf_selec, aes(x=GDP, y=HCI)) + geom_point() + 
+   scale_x_continuous(trans='log10') + 
+   geom_text(aes(label = ifelse(ctry_inf$Continent == "Asia", as.character(Country), "")))
Error in `geom_text()`:
! Problem while computing aesthetics.
ℹ Error occurred in the 2nd layer.
Caused by error in `check_aesthetics()`:
! Aesthetics must be either length 1 or the same as the data (91)
✖ Fix the following mappings: `label`
Run `rlang::last_trace()` to see where the error occurred.

问题原因 #

报错核心是**geom_text的label美学映射长度与当前图层数据集长度不匹配**：你在ifelse里引用了原始数据集ctry_inf$Country，而当前图层用的是筛选后的ctry_inf_selec（共91行），原始数据集行数多于91，导致生成的label向量长度和图层数据不一致，触发ggplot的美学检查报错。

解决方案（两种方法，均仅用ggplot2） #

方法1：修正ifelse的数据集引用 #

直接使用当前图层的数据集列名（ggplot会自动继承ctry_inf_selec），不需要加ctry_inf$前缀：

ctry_inf <- read.csv("Paises_PIB_ICH.csv")

ctry_inf_selec <- ctry_inf[ctry_inf$Continent == "Asia" | ctry_inf$Continent == "Africa",  ]

ggplot(ctry_inf_selec, aes(x=GDP, y=HCI)) + 
  geom_point() + 
  scale_x_continuous(trans='log10') + 
  geom_text(aes(label = 
                  ifelse(Country == "Namibia" | Country == "United Arab Emirates", 
                         as.character(Country), "")))

方法2：单独给geom_text指定目标数据集（更高效） #

只给需要标注的两个国家生成文本，避免生成大量空字符串：

ctry_inf <- read.csv("Paises_PIB_ICH.csv")

ctry_inf_selec <- ctry_inf[ctry_inf$Continent == "Asia" | ctry_inf$Continent == "Africa",  ]
# 筛选出需要标注的国家
label_data <- ctry_inf_selec[ctry_inf_selec$Country %in% c("Namibia", "United Arab Emirates"), ]

ggplot(ctry_inf_selec, aes(x=GDP, y=HCI)) + 
  geom_point() + 
  scale_x_continuous(trans='log10') + 
  geom_text(data = label_data, aes(label = Country))

两种方法都能解决长度不匹配的问题，方法2更轻量化，推荐使用。

内容的提问来源于stack exchange，提问作者Luís Frederico Santos