如何在R中并排查看列表？多数据集变量对比需求

Great question! You can absolutely build a clean, aligned comparison using purrr, janitor, and tidyr—here's a step-by-step implementation that handles multiple data frames, aligns common variables first, and pushes unique variables to the end:

Step 1: Setup and Sample Data #

First, let's adjust your sample data to ensure factors are properly defined (since your original code targets factor variables):

library(tidyverse)
library(janitor)

# Sample data with factor variables
df1 <- data.frame(
  var1 = factor(c("a", "b")),
  var2 = factor(c("A", "B")),
  var3 = factor(c("checking"))
)

df2 <- data.frame(
  var1 = factor(letters),
  var2 = factor(LETTERS),
  var4 = factor(c("testing"))
)

# Store data frames in a list (works for 2+ frames)
df_list <- lst(df1, df2)

Step 2: Define Ordered Variables (Common First, Unique Last) #

We'll first identify which factor variables are common across all data frames and which are unique to some, then order them accordingly:

# Get factor variables for each data frame
factor_vars_per_df <- df_list %>% 
  map(~select_if(.x, is.factor) %>% names())

# All unique factor variables across all frames
all_factor_vars <- factor_vars_per_df %>% flatten_chr() %>% unique()

# Variables present in every data frame (common)
common_factor_vars <- all_factor_vars %>% 
  keep(~every(factor_vars_per_df, ~.x %in% .))

# Variables present in only some data frames (unique)
unique_factor_vars <- setdiff(all_factor_vars, common_factor_vars)

# Final ordered list: common variables first, then unique
ordered_vars <- c(common_factor_vars, unique_factor_vars)

Step 3: Generate Aligned Comparison Tables #

For each ordered variable, we'll create a summary table that includes counts and percentages from every data frame (filling in missing values with NA for frames that don't have the variable):

compare_summaries <- ordered_vars %>% 
  map(function(var) {
    # For each data frame, get tabyl or empty placeholder if variable is missing
    df_list %>% 
      map(function(df) {
        if (var %in% names(df)) {
          df %>% 
            select(all_of(var)) %>% 
            tabyl() %>% 
            mutate(variable = var)
        } else {
          # Placeholder for missing variables
          tibble(
            variable = var,
            !!var := NA_character_,
            n = NA_integer_,
            percent = NA_real_
          )
        }
      }) %>% 
      bind_rows(.id = "data_frame") %>% 
      # Reshape to have each data frame's stats as separate columns
      pivot_wider(
        names_from = data_frame,
        values_from = c(n, percent),
        names_glue = "{data_frame}_{.value}"
      ) %>% 
      # Reorder columns for readability
      select(variable, all_of(var), starts_with("df1_"), starts_with("df2_"))
  })

Step 4: View the Results #

When you print compare_summaries, you'll get a list of tables where each table corresponds to a variable. Common variables appear first, and unique variables are at the end:

# Print all summaries
compare_summaries

Example Output Snippet:

For var1 (common to both frames):

# A tibble: 26 × 5
   variable var1  df1_n df1_percent df2_n df2_percent
   <chr>    <fct> <int>       <dbl> <int>       <dbl>
 1 var1     a         1         0.5     1       0.0385
 2 var1     b         1         0.5     1       0.0385
 3 var1     c        NA        NA        1       0.0385
...

For var3 (unique to df1):

# A tibble: 1 × 5
  variable var3     df1_n df1_percent df2_n df2_percent
  <chr>    <fct>    <int>       <dbl> <int>       <dbl>
1 var3     checking     1           1    NA          NA

Key Features: #

• Works with 2+ data frames (just add more to df_list)
• Aligns common variables at the top of the output
• Pushes variables unique to some frames to the end
• Includes counts and percentages (from tabyl) for each frame
• Fills in NA for variables missing from a data frame

内容的提问来源于stack exchange，提问作者user63230