Advanced data visualization techniques with ggplot2.

Key Visualization Techniques Used

1. Faceting (facet_wrap) 🧩

The core technique is faceting, which breaks the data into subsets and displays each subset in its own plot panel.

facet_wrap(~ island, ncol = 2): This creates a separate panel for each unique value in the island column (Torgersen, Biscoe, Dream). This allows for a quick visual comparison of the Flipper Length/Body Mass relationship across the different geographic colonies.

Note: While the prompt asked to facet by species, faceting by island while coloring by species provides a richer, three-way comparison, revealing how species distribution and relationships differ geographically.

2. Multi-Layer Geoms (geom_point and geom_smooth) 📊

Two distinct geometric layers are stacked to show both the raw data and the trend:

geom_point(alpha = 0.7, size = 3): Visualizes the raw data, mapping the aesthetic variables (color, shape) to the species column defined in the global aes(). The alpha setting ensures points that overlap are still visible.

geom_smooth(method = "lm", se = FALSE, linewidth = 1.2): Overlays a Linear Model (LM) regression line. This is powerful for quantitative risk analysis as it immediately visualizes the general trend: as flipper length increases, body mass also increases. Removing the standard error (se = FALSE) declutters the visual.

3. Custom Color Palette (scale_color_viridis_d) 🌈

Instead of the default ggplot2 colors, a custom scale is used:

scale_color_viridis_d(option = "D", ...): The Viridis color palettes are perceptually uniform, meaning the colors maintain the same perceived brightness across the spectrum. This is ideal for categorical data (the _d suffix indicates discrete data) as it ensures no single species appears artificially more prominent or lighter than the others.

4. Custom Themeing (theme_minimal and theme(...)) 🖼️

The plot starts with theme_minimal() for a clean, white background and then uses the theme() function for fine-grained control:

Titles: Setting plot.title = element_text(face = "bold", size = 16, hjust = 0.5) centers the title and makes it bold for emphasis.

Facet Labels: strip.text = element_text(face = "bold", size = 10) ensures the facet labels (the island names) stand out clearly within each panel.

Sample Answer

Advanced ggplot2 Visualization: Penguin Flipper Length vs. Body Mass

The following code uses the penguins dataset (available via the palmerpenguins package) and demonstrates faceting, geom layers, and custom aesthetics to enhance the visualization.

R Code for the Faceted Plot

# Assuming you have the palmerpenguins and ggplot2 libraries loaded
library(ggplot2)
library(palmerpenguins)

# Create the advanced faceted plot
ggplot(
  data = penguins,
  mapping = aes(
    x = flipper_length_mm,
    y = body_mass_g,
    color = species,        # Map species to color
    shape = species         # Map species to shape
  )
) +
  # 1. Add Scatter Plot Points
  geom_point(alpha = 0.7, size = 3) +

  # 2. Add Regression Line (Linear Model) for the Trend
  # The 'se = FALSE' removes the standard error ribbon for a cleaner look.
  geom_smooth(method = "lm", se = FALSE, linewidth = 1.2) +

  # 3. Faceting: Separate plots by 'island'
  # 'scales = "free_x"' allows the x-axis limits to vary for each island,
  # but we'll stick to 'fixed' for direct comparison here.
  facet_wrap(~ island, ncol = 2) +