如何用Base R或Tidyverse移除除id列外全为空的行？

原始数据 #

df <- structure(list(id = c("A", "B", "C", "D", "E", "F", "G", "H", 
"I", "J", "K", "L"), name = c("Sam", "Dave", NA, "Alan", "Ted", 
"Ana", NA, NA, "Max", NA, "Walt", NA), age = c(28, 32, NA, NA, 
23, NA, 21, NA, NA, 22, 21, NA), height = c(NA, 161, NA, 168, 
167, NA, 166, NA, 158, 171, NA, NA)), class = c("tbl_df", "tbl", 
"data.frame"), row.names = c(NA, -12L))

需求说明 #

需要移除除id列外所有单元格均为NA的行，最终得到如下数据框：

final <- structure(list(id = c("A", "B", "D", "E", "F", "G", "I", "J", 
"K"), name = c("Sam", "Dave", "Alan", "Ted", "Ana", NA, "Max", 
NA, "Walt"), age = c(28, 32, NA, 23, NA, 21, NA, 22, 21), height = c(NA, 
161, 168, 167, NA, 166, 158, 171, NA)), class = c("tbl_df", "tbl", 
"data.frame"), row.names = c(NA, -9L))

Base R 实现 #

通过rowSums和is.na组合判断每行除id外的NA数量，筛选出至少有一个非NA值的行：

final_base <- df[rowSums(is.na(df[, -match("id", names(df))])) != (ncol(df) - 1), ]

• df[, -match("id", names(df))]：提取除id列外的所有列
• is.na(...)：将列转换为逻辑矩阵，NA值标记为TRUE
• rowSums(...)：计算每行的NA数量
• ncol(df)-1：除id外的总列数，当该行NA数量不等于该值时，说明存在非NA值，保留该行

tidyverse 实现 #

使用dplyr的filter和if_all函数，简洁实现筛选逻辑：

library(dplyr)

final_tidy <- df %>%
  filter(!if_all(-id, is.na))

• -id：指定除id列外的所有列
• if_all(-id, is.na)：判断该行除id外的所有列是否全为NA
• !：取反，筛选出不满足全NA的行

两种方法得到的结果均与目标数据框final一致。

内容的提问来源于stack exchange，提问作者Shubham