学ぶ Density and Violin Plots | Advanced Plot Types

Data Visualization in R with ggplot2

メニューを表示するにはスワイプしてください

Understanding the distribution of your data is a crucial step in data analysis, and density and violin plots are two powerful tools for this purpose. Density plots display the distribution of a continuous variable, making it easy to see the shape, spread, and modality (number of peaks) of the data. They are especially useful when you want to visualize how a variable is distributed across its range and to compare distributions between groups. Violin plots extend this idea by combining features of boxplots and density plots: they show the distribution of the data across different categories, revealing not only summary statistics like the median and quartiles but also the full shape of the distribution. Use density plots when you are interested in the overall shape of a single distribution or comparing a few distributions. Choose violin plots when you want a detailed comparison of distributions across categories, especially to highlight nuances such as multimodality or skewness that boxplots might miss.


              12345678910111213141516171819
            
library(ggplot2)

# Create a sample data frame
df <- data.frame(
  category = rep(c("A", "B", "C"), each = 50),
  value = c(
    rnorm(50, mean = 5, sd = 1),
    rnorm(50, mean = 7, sd = 1.5),
    rnorm(50, mean = 6, sd = 0.8)
  )
)

# Violin plot comparing distributions across categories
ggplot(df, aes(x = category, y = value, fill = category)) +
  geom_violin(trim = FALSE) +
  labs(title = "Violin Plot of Value by Category",
       x = "Category",
       y = "Value") +
  theme_minimal()

While both boxplots and violin plots are used to summarize distributions across categories, violin plots provide a richer view of the underlying data. In the code above, each violin shape represents the distribution of the value variable for a category. Unlike boxplots, which display only the median, quartiles, and potential outliers, violin plots reveal the full density curve, making it easy to spot features like bimodality, skewness, or the presence of multiple peaks. This additional information can be critical when comparing groups, as it helps you understand not just where most data points lie, but also how the data is spread and whether there are subgroups or unusual patterns within each category.