为正负水平条形图添加颜色：营收差异图表颜色渲染异常排查

Got it, let's sort out that color issue you're facing! The problem with your current code is that you’re setting the fill value directly in geom_bar() instead of mapping it to your data within aes(). ggplot needs to associate the color with the Difference values, so we need to adjust how we handle the fill mapping.

Here's the corrected approach: #

First, we’ll map the fill to a categorical indicator (whether the difference is positive or negative) inside aes(), then use scale_fill_manual() to assign your desired colors.

Option 1: Map directly in the plot (no data modification)

ggplot(appts2, aes(x = Provider, y = Difference)) +
  # Map fill to a dynamic indicator of positive/negative difference
  geom_bar(aes(fill = ifelse(Difference < 0, "Negative", "Positive")), 
           stat = "identity") +
  # Assign your custom colors to the categories
  scale_fill_manual(values = c("Negative" = "red", "Positive" = "black")) +
  coord_flip() +
  # Correct way to add a title in ggplot (replace the old `main` parameter)
  ggtitle("Difference in Projected vs Actual Revenue") +
  # Optional: Hide the legend if you don't need it
  guides(fill = "none")

Option 2: Preprocess the data (cleaner for complex plots)

If you prefer a more readable workflow, add a new column to your data frame first to label positive/negative differences:

# Using dplyr to add a new column (install dplyr if you haven't: install.packages("dplyr"))
library(dplyr)
appts2 <- appts2 %>%
  mutate(Diff_Category = ifelse(Difference < 0, "Negative", "Positive"))

# Now plot with the pre-defined category
ggplot(appts2, aes(x = Provider, y = Difference, fill = Diff_Category)) +
  geom_bar(stat = "identity") +
  scale_fill_manual(values = c("Negative" = "red", "Positive" = "black")) +
  coord_flip() +
  ggtitle("Difference in Projected vs Actual Revenue") +
  guides(fill = "none")

Why your original code didn't work: #

When you set fill = ifelse(appts2$Difference < 0, "red", "black") directly in geom_bar() (outside aes()), ggplot treats this as a single fixed value instead of mapping it to each row of your data. By moving the fill logic into aes(), you tell ggplot to assign colors based on the value of Difference for each provider.

内容的提问来源于stack exchange，提问作者regents