基于双列条件过滤：移除特定条件下含NA值城市的所有行（dplyr实现）

Got it, let's solve this problem using dplyr—here's a straightforward approach that fits your requirements perfectly:

First, make sure you have the dplyr package loaded:

library(dplyr)

Then, use this code to clean your data frame:

clean_df <- df %>%
  group_by(city) %>%
  # Filter out cities where ANY row with crime = "total" has an NA value
  filter(!any(crime == "total" & is.na(value))) %>%
  ungroup()

How this works: #

• group_by(city): Groups the data by each city, so we can evaluate conditions across all rows for a single city.
• filter(!any(crime == "total" & is.na(value))):
  • The any() function checks if there's at least one row in the city's group where crime is "total" and value is NA.
  • We use ! (NOT) to invert this check—so we keep only cities where this condition never occurs.
• ungroup(): Resets the grouping to return a standard tibble/data frame.

Result: #

When you run this on your sample data, Amsterdam gets removed entirely (since its 2017 "total" crime value is NA), leaving only Rotterdam's rows:

clean_df
#> # A tibble: 3 × 4
#>   city      year crime  value
#>   <chr>    <int> <chr>  <int>
#> 1 Rotterdam  2015 total   4901
#> 2 Rotterdam  2016 total   4830
#> 3 Rotterdam  2017 total   4659

This logic also works if your data has other crime types beyond "total"—any city with a NA in its "total" crime values will have all its rows removed, regardless of other crime categories.

内容的提问来源于stack exchange，提问作者Tdebeus