ggplot2中X轴含多变量交互时的标签优化及scale_fill_manual报错解决方法

Hey there! Let's tackle your two ggplot2 problems one by one—they're both straightforward fixes, so you’ll have your plot looking perfect in no time.

Problem 1: Simplifying X-axis Labels to Only Race Names #

Since you’re already faceting by inc_level, you don’t need to use interaction(race_code, inc_level) for your x-axis at all! Mapping x = race_code directly will automatically show only race names (white, black, etc.) on each facet’s x-axis, which is exactly what you want. This avoids the messy combined labels entirely and keeps your plot cleaner.

If for some reason you need to keep using the interaction (though it’s unnecessary here), you can override the labels with scale_x_discrete to strip off the income level suffix:

scale_x_discrete(labels = function(x) gsub("\\..*", "", x))

This regex removes everything from the first dot onward, leaving just the race name.

Problem 2: Fixing the "Continuous value supplied to discrete scale" Error #

Even though your YEAR column only has three values, if it’s stored as a numeric/integer type (not a factor), ggplot treats it as a continuous variable. But scale_fill_manual is designed for discrete categorical scales, hence the mismatch error.

You have two simple fixes:

• Option 1: Convert YEAR to a factor directly in the aes() call:

  aes(x = race_code, y = percent_lwpr, group = YEAR, fill = factor(YEAR))
  

• Option 2: Convert the YEAR column to a factor in your data frame first (this makes all future uses of YEAR in plots discrete by default):

  data1$YEAR <- factor(data1$YEAR)
  

Full Modified Code (Both Issues Fixed) #

Here’s the complete code using the cleanest approach—direct x-axis mapping + factor conversion in the aesthetic:

ggplot(data = data1, aes(x = race_code, y = percent_lwpr, group = YEAR, fill = factor(YEAR))) +
  geom_bar(stat="identity", position=position_dodge(width=0.8)) +
  scale_fill_manual(values=c("#ffa600", "#ff6361", "#bc5090")) +
  facet_grid(~inc_level, switch = "x", scales = "free_x", space = "free_x") +
  theme(panel.spacing = unit(0, "lines"), 
        strip.background = element_blank(), 
        strip.placement = "outside")

内容的提问来源于stack exchange，提问作者upoma