如何用dplyr基于某行值删除多行数据

First, let's recreate your sample dataset so we're working with the same starting point:

library(dplyr)

# Original dataset
pet_data <- tibble(
  Pet = c("Cow", "Cow", "Cow", "Dog", "Dog", "Dog", "Tiger", "Tiger", "Tiger"),
  Day = c("Monday", "Tuesday", "Wednesday", "Monday", "Tuesday", "Wednesday", "Monday", "Tuesday", "Wednesday"),
  Bite = c("No", "No", "No", "No", "No", "No", "No", "Yes", "No")
)

To remove all pets that have ever bitten (even once), we can use a concise combination of group_by() and filter() in dplyr:

# Filter out pets with any history of biting
safe_pets <- pet_data %>%
  group_by(Pet) %>%  # Group all records by individual pet
  filter(all(Bite == "No")) %>%  # Keep only groups where EVERY Bite entry is "No"
  ungroup()  # Reset grouping to return a standard data frame

How this works: #

• group_by(Pet): Clusters all rows for each pet together, so we can evaluate their entire biting history as a group.
• filter(all(Bite == "No")): For each pet group, we only keep it if every entry in the Bite column is "No". If a pet has even one "Yes" (like the Tiger), this condition fails, and the entire group is excluded.
• ungroup(): Removes the pet grouping so the final safe_pets is a regular, ungrouped data frame.

Running this code will give you exactly your target dataset—only cows and dogs, with no trace of the tiger.

Alternative Explicit Approach #

If you prefer a more step-by-step method, you can first identify safe pets, then filter the original data:

# Step 1: Get a list of pets that never bit
safe_pet_names <- pet_data %>%
  group_by(Pet) %>%
  summarise(never_bitten = all(Bite == "No")) %>%
  filter(never_bitten) %>%
  pull(Pet)  # Extract just the pet names as a vector

# Step 2: Filter original data to keep only safe pets
safe_pets <- pet_data %>%
  filter(Pet %in% safe_pet_names)

This achieves the same result, but the first method is more efficient and concise for most use cases.

内容的提问来源于stack exchange，提问作者peekay