技术问询：使用lapply函数绘制多图时移除各图未使用的X轴因子

Got it, let's tackle this problem step by step. You've got a wide-format dataframe with demographic details and subscale scores, and when using lapply to generate plots for each subscale, you want to get rid of those unused group factors on the X-axis (like groups that have no valid scores for a specific subscale). Here's a straightforward solution using tidyverse tools (I'll also include a base R option if you prefer):

First, let's confirm we're working with the full dataset you provided:

id <- c(1, 2, 3, 4, 5)
gender <- c("Male", "Female", "Male", "Female", "Male")
group <- c("A", "A", "B", "C", "B")
subscale_1 <- c(NA, NA, 3, 2, 3)
subscale_2 <- c(3, 3, NA, 2, NA)
subscale_3 <- c(3, 2, 5, NA, 1)
subscale_4 <- c(1, NA, 3, NA, 5)
subscale_5 <- c(NA, 5, NA, 8, NA)
df <- data.frame(id, gender, group, subscale_1, subscale_2, subscale_3, subscale_4, subscale_5)

Step 1: Reshape the Data to Long Format #

Wide format is great for data entry, but long format is way easier for plotting and grouping by subscale. We'll use pivot_longer from the tidyr package to restructure the data:

library(tidyr)
library(ggplot2) # For cleaner, more flexible plots

# Convert to long format and filter out rows with missing scores
df_long <- pivot_longer(df, 
                        cols = starts_with("subscale_"), 
                        names_to = "subscale", 
                        values_to = "score") %>%
  filter(!is.na(score))

Step 2: Use lapply to Generate Plots with Clean X-Axes #

The key trick here is re-setting the factor levels for group within each subscale's data—this tells R to only use groups that have valid scores for that specific subscale.

ggplot2 Version (Recommended)

This creates a list of boxplots (swap in geom_point, geom_bar, etc., based on your needs):

# Get unique subscale names to iterate over
subscale_list <- unique(df_long$subscale)

# Generate plots with lapply
plot_list <- lapply(subscale_list, function(sub) {
  # Grab only data for the current subscale
  current_data <- df_long[df_long$subscale == sub, ]
  
  # Reset group factor levels to ONLY those present in the current subscale
  current_data$group <- factor(current_data$group, levels = unique(current_data$group))
  
  # Build the plot
  ggplot(current_data, aes(x = group, y = score)) +
    geom_boxplot(fill = "#63B8FF") +
    labs(title = paste("Score Distribution:", sub),
         x = "Group", y = "Score") +
    theme_minimal()
})

# Print all plots to view them
lapply(plot_list, print)

Base R Version

If you prefer base R plotting, here's how to achieve the same result:

plot_list_base <- lapply(subscale_list, function(sub) {
  current_data <- df_long[df_long$subscale == sub, ]
  valid_groups <- unique(current_data$group)
  
  boxplot(score ~ factor(group, levels = valid_groups), 
          data = current_data,
          main = paste("Score Distribution:", sub),
          xlab = "Group", ylab = "Score",
          col = "#63B8FF")
})

# Print all base R plots
lapply(plot_list_base, print)

Why This Works #

By default, if group is a factor (or gets coerced to one during plotting), R will show all original factor levels on the X-axis—even if there's no data for some groups. By explicitly setting levels = unique(current_data$group), we tell R to only display groups that have valid scores for the current subscale, eliminating those unused factors.

内容的提问来源于stack exchange，提问作者DTYK