使用R语言替换DF空列及随机分配性别值的问题

Hey there! I’ve run into this exact frustration before when cleaning demographic data—nothing’s more annoying than only half your empty cells getting filled. Let’s break down why this might be happening and fix it step by step.

Common Reasons for Partial Assignment #

First, let’s rule out the most likely culprits:

• You’re not targeting all empty values: Your data might have a mix of NA values and empty strings (""), but your filter only catches one type.
• Incorrect indexing: Using df$gender == NA instead of is.na(df$gender) (since comparing anything to NA returns NA, which gets ignored in indexing).
• Mismatched sample size: If you hardcode a size in sample() instead of using the actual number of empty cells, you’ll only fill that many values.

Step-by-Step Solutions #

1. First, Identify All Empty Values #

Start by checking what kind of empty values you’re dealing with:

# Count NA values in gender column
sum(is.na(df$gender))
# Count empty strings in gender column (ignore NAs for this check)
sum(df$gender == "", na.rm = TRUE)

2. Fix the Assignment #

Pick the method that matches your data (or combine them if you have both empty types):

Case 1: Empty Values are NA

# Get indices of all rows with NA gender
na_gender_rows <- which(is.na(df$gender))
# Randomly assign "Male" or "Female" (adjust categories to fit your needs)
df$gender[na_gender_rows] <- sample(
  x = c("Male", "Female"),
  size = length(na_gender_rows),
  replace = TRUE
)

Case 2: Empty Values are Empty Strings ("")

# Get indices of all rows with empty string gender
empty_gender_rows <- which(df$gender == "")
# Randomly assign values to these rows
df$gender[empty_gender_rows] <- sample(
  x = c("Male", "Female"),
  size = length(empty_gender_rows),
  replace = TRUE
)

Case 3: Mix of NA and Empty Strings

Standardize all empty values to NA first, then handle them in one go:

# Convert empty strings to NA for consistency
df$gender[df$gender == ""] <- NA
# Get indices of all NA rows
na_indices <- which(is.na(df$gender))
# Assign random values to all empty cells
df$gender[na_indices] <- sample(c("Male", "Female"), length(na_indices), replace = TRUE)

3. Full Test Example #

Let’s verify this works with a sample data frame:

# Create test data with mixed empty values
df <- data.frame(
  id = 1:10,
  gender = c("Male", NA, "Female", "", NA, "Male", "", NA, "Female", "")
)

# Clean and assign random genders
df$gender[df$gender == ""] <- NA
na_rows <- which(is.na(df$gender))
df$gender[na_rows] <- sample(c("Male", "Female"), length(na_rows), replace = TRUE)

# View the complete result
print(df)

This should fill every empty gender field with a random value, no partial assignments left.

内容的提问来源于stack exchange，提问作者add-semi-colons