Stratified Boxplot in R

Stratified boxplots, also known as grouped or side-by-side boxplots, allow you to compare the distributions of a numerical variable across different levels of a categorical variable. In R, the boxplot() function can be used to generate stratified boxplots. This tutorial will walk you through the process:

1. Basic Setup

First, let's set up some sample data and load necessary libraries:

# Generating sample data
set.seed(123)
group1 <- rnorm(50, mean=50, sd=10)
group2 <- rnorm(50, mean=60, sd=15)
group3 <- rnorm(50, mean=55, sd=12)

data <- data.frame(
  Value = c(group1, group2, group3),
  Group = factor(rep(1:3, each=50))
)

# Load necessary libraries
library(ggplot2)

2. Basic Stratified Boxplot

Using base R's boxplot() function:

boxplot(Value ~ Group, data = data, main="Stratified Boxplot", xlab="Group", ylab="Value")

Using ggplot2:

ggplot(data, aes(x=Group, y=Value)) + 
  geom_boxplot() + 
  labs(title="Stratified Boxplot", x="Group", y="Value")

3. Customizing Appearance

Using base R:

boxplot(Value ~ Group, data = data, main="Stratified Boxplot", xlab="Group", ylab="Value",
        col=c("red", "green", "blue"))

Using ggplot2:

ggplot(data, aes(x=Group, y=Value, fill=Group)) + 
  geom_boxplot() + 
  labs(title="Stratified Boxplot", x="Group", y="Value") + 
  scale_fill_manual(values=c("red", "green", "blue"))

4. Adding Data Points

Sometimes it's useful to overlay individual data points on the boxplot:

Using ggplot2:

ggplot(data, aes(x=Group, y=Value, fill=Group)) + 
  geom_boxplot(alpha=0.7) + 
  geom_jitter(width=0.2) + 
  labs(title="Stratified Boxplot with Data Points", x="Group", y="Value") + 
  scale_fill_manual(values=c("red", "green", "blue"))

5. Horizontal Boxplots

You might prefer horizontal boxplots in some cases:

Using ggplot2:

ggplot(data, aes(y=Group, x=Value, fill=Group)) + 
  geom_boxplot() + 
  labs(title="Horizontal Stratified Boxplot", y="Group", x="Value") + 
  scale_fill_manual(values=c("red", "green", "blue"))

6. Tips:

Base R's boxplot() function provides quick and simple boxplots, but for more intricate customizations, ggplot2 is more flexible.
When adding individual data points with geom_jitter(), the width argument can control the spread of the jitter.

Conclusion

Stratified boxplots provide a clear way to visualize and compare distributions across different groups. Whether you use base R or ggplot2, the key is to ensure your visualizations effectively communicate the underlying data patterns.

Creating Grouped Boxplots in R:
- Use the boxplot() function to create grouped boxplots.
```
# Example with two groups
boxplot(value ~ group, data=my_data)
```
R Boxplot by Group or Category:
- Group data by a categorical variable in a boxplot.
```
boxplot(value ~ category, data=my_data)
```
Stratified Boxplot with ggplot2 in R:
- Use ggplot2 for more customization in stratified boxplots.
```
library(ggplot2)
ggplot(my_data, aes(x=category, y=value)) + geom_boxplot()
```
Adding Colors to Stratified Boxplots in R:
- Enhance visual appeal by adding colors to boxplots.
```
ggplot(my_data, aes(x=category, y=value, fill=category)) + geom_boxplot()
```
Customizing Boxplot Appearance in R:
- Customize appearance using parameters like notch, outline, and width.
```
boxplot(value ~ group, data=my_data, notch=TRUE, outline=FALSE, width=0.5)
```
Multiple Boxplots on One Graph in R:
- Compare multiple groups or categories on a single graph.
```
boxplot(value ~ group, data=my_data, add=TRUE, col="lightblue")
```
R Boxplot by Factor Levels:
- Create boxplots based on factor levels.
```
boxplot(value ~ factor_variable, data=my_data)
```
Grouped Boxplot with Notched Boxes in R:
- Add notches to visualize confidence intervals.
```
boxplot(value ~ group, data=my_data, notch=TRUE)
```
Interactive Stratified Boxplot in R:
- Use interactive plotting libraries like plotly for dynamic exploration.
```
library(plotly)
plot_ly(my_data, x=~category, y=~value, type="box")
```
Comparing Distributions with Stratified Boxplots in R:
- Compare distributions across different categories.
```
ggplot(my_data, aes(x=category, y=value, color=category)) + geom_boxplot()
```
Combining Boxplots and Scatter Plots in R:
- Combine boxplots with scatter plots for a comprehensive view.
```
ggplot(my_data, aes(x=category, y=value)) + geom_boxplot() + geom_jitter()
```
R Boxplot Outliers by Group:
- Identify and highlight outliers in boxplots.
```
boxplot(value ~ group, data=my_data, outline=TRUE)
```
Boxplot Summary Statistics in R:
- Display summary statistics using the summary() function.
```
summary_boxplot <- boxplot(value ~ group, data=my_data)
summary(summary_boxplot)
```
Handling Missing Data in Stratified Boxplots in R:
- Address missing data using functions like na.omit().
```
ggplot(na.omit(my_data), aes(x=category, y=value, fill=category)) + geom_boxplot()
```