R语言合并含互补NA值的两列为无NA单列并删除原列

现有如下结构的R数据框（tidytable/data.table类型）：

structure(list(RFM_Score_Meaning.x = c("Dropped", "Dropped", 
"Dropped", "Dropped", "Dropped", "Dropped", "Dropped", NA, NA, 
NA, NA, NA, NA, NA, NA, NA), RFM_Score_Meaning.y = c(NA, NA, 
NA, NA, NA, NA, NA, "Need Encouragement", "Need Encouragement", 
"Need Encouragement", "Need Encouragement", "Need Encouragement", 
"Need Encouragement", "Need Encouragement", "Need Encouragement", 
"Need Encouragement")), row.names = c(NA, -16L), class = c("tidytable", 
"data.table", "data.frame"), .internal.selfref = <pointer: 0x000002470fabbfe0>)

两列的NA位置完全互补：RFM_Score_Meaning.x的NA行对应RFM_Score_Meaning.y的有效值，反之亦然。需要删除这两列，生成一列包含所有有效值且无NA的新列。

以下是几种实用的实现方法：

方法1：使用tidyverse/dplyr的coalesce() #

coalesce()函数会逐行取第一个非NA的值，完美适配这种互补列的合并场景：

library(dplyr)

# 合并列并删除原列
df <- df %>%
  mutate(RFM_Score_Meaning = coalesce(RFM_Score_Meaning.x, RFM_Score_Meaning.y)) %>%
  select(-RFM_Score_Meaning.x, -RFM_Score_Meaning.y)

方法2：Base R原生实现 #

无需加载额外包，用ifelse判断每行的NA情况：

# 生成新列
df$RFM_Score_Meaning <- ifelse(is.na(df$RFM_Score_Meaning.x), 
                               df$RFM_Score_Meaning.y, 
                               df$RFM_Score_Meaning.x)
# 删除原两列
df <- df[, !names(df) %in% c("RFM_Score_Meaning.x", "RFM_Score_Meaning.y")]

方法3：data.table/tidytable原生语法 #

针对你的数据框类型，用data.table的fcoalesce()更高效：

library(data.table) # 或 tidytable

# 生成新列并删除原列
df[, RFM_Score_Meaning := fcoalesce(RFM_Score_Meaning.x, RFM_Score_Meaning.y)][, 
  c("RFM_Score_Meaning.x", "RFM_Score_Meaning.y") := NULL]

内容的提问来源于stack exchange，提问作者hachiko