在R语言中删除所有变量值均缺失的行：现有代码无效求解决方案

Hey there! Let's get this sorted out. You're looking to remove only rows where every single variable is missing, while keeping rows that have at least one non-missing value—this is a common task, and there are two reliable ways to do it in R: using base R or the tidyverse/dplyr package.

1. Base R Approach #

This method uses rowSums() to count non-missing values per row, then filters to keep rows with at least one non-missing entry:

# Create a test dataframe (replace with your actual data)
df <- data.frame(
  var1 = c(NA, 5, NA, 9),
  var2 = c(NA, NA, 3, NA),
  var3 = c(NA, NA, NA, 7)
)

# Remove rows where all values are missing
clean_df_base <- df[rowSums(!is.na(df)) > 0, ]

# View the result
clean_df_base

How this works: #

• !is.na(df) converts your dataframe to a boolean matrix where TRUE means non-missing, FALSE means missing.
• rowSums() adds up the TRUE values (treated as 1s) for each row.
• We keep only rows where this sum is greater than 0—meaning at least one value is present.

2. Tidyverse (dplyr) Approach #

If you prefer the tidyverse syntax, use filter() with if_any() to check for at least one non-missing value across all columns:

library(dplyr)

# Using the same test dataframe
df <- data.frame(
  var1 = c(NA, 5, NA, 9),
  var2 = c(NA, NA, 3, NA),
  var3 = c(NA, NA, NA, 7)
)

# Remove rows with all missing values
clean_df_tidy <- df %>% 
  filter(if_any(everything(), ~!is.na(.x)))

# View the result
clean_df_tidy

How this works: #

• everything() targets all columns in your dataframe.
• if_any() checks if any column in the row meets the condition !is.na(.x) (i.e., has a non-missing value).
• Rows that pass this check are kept.

Common Pitfall to Avoid #

A common mistake is using na.omit(df)—this removes all rows with any missing values, which is not what you want. Stick to the methods above to preserve rows with partial missing data.

内容的提问来源于stack exchange，提问作者IanM