如何在R语言中基于指定列的值进行数据抽样？

Hey there! Let's walk through how to perform sampling operations based on specific column values using your dataset. First, let's fix the truncated code to get the full reproducible dataset:

df <- data.frame(
  ID = c("A", "A", "A", "B", "B", "B", "C", "C", "C", "D", "D", "D", "E", "E", "E"), 
  X = c(2L, 4L, 4L, 5L, 3L, 1L, 5L, 2L, 5L, 2L, 3L, 3L, 5L, 4L, 3L), 
  Y = c(2L, 3L, 3L, 3L, 4L, 5L, 3L, 3L, 3L, 3L, 2L, 2L, 3L, 3L, 4L),
  Z = c(5L, 2L, 2L, 1L, 2L, 3L, 1L, 4L, 1L, 4L, 4L, 4L, 1L, 2L, 2L),
  Average = c(3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L)
)

Below are common sampling scenarios tailored to your data:

1. Grouped Sampling (e.g., by ID column) #

If you want to sample a fixed number or fraction of rows per group (like each ID), the dplyr package makes this straightforward:

Sample a fixed number of rows per group #

library(dplyr)

# Randomly select 1 row for each ID
grouped_sample_n <- df %>%
  group_by(ID) %>%
  sample_n(size = 1) %>%  # Change size to your desired number
  ungroup()

# View the result
print(grouped_sample_n)

Sample a fraction of rows per group #

# Select 2/3 of rows for each ID (since each ID has 3 rows, this picks 2 rows)
grouped_sample_frac <- df %>%
  group_by(ID) %>%
  sample_frac(size = 2/3) %>%  # Adjust fraction as needed
  ungroup()

2. Conditional Sampling (based on column value filters) #

If you want to sample rows that meet a specific condition (e.g., X > 3), combine filtering with sampling:

# First filter rows where X > 3, then randomly sample 5 rows from the result
conditional_sample <- df %>%
  filter(X > 3) %>%
  sample_n(size = 5)

# Or if you want all rows matching the condition, just use filter():
all_matching_rows <- df %>% filter(X > 3)

3. Global Sampling (filter first, then sample across the dataset) #

Even though all your Average values are 3, here's how you'd sample from rows matching a column value globally:

# Randomly sample 10 rows where Average equals 3
global_sample <- df %>%
  filter(Average == 3) %>%
  sample_n(size = 10)

4. Base R Alternative (no packages needed) #

If you prefer not to use dplyr, you can achieve grouped sampling with base R functions:

# Randomly select 1 row per ID using split + lapply + rbind
base_r_grouped_sample <- do.call(
  rbind,
  lapply(split(df, df$ID), function(group) group[sample(nrow(group), 1), ])
)

内容的提问来源于stack exchange，提问作者Geet