R GGPLOT2 other uses

Load primary packages

# Load required packages
library(tidyverse)
library(ggplot2)
library(wesanderson)

Streamgraphs

Streamgraphs is a type of chart that’s like a stacked area chart. Instead of plotting values against a conventional y axis, streamgraphs make the starting point of each section balanced in the middle of the chart(symmetrical around the x axis).

Advantages:

• Streamgraphs is good for showing trends and patterns in data, especially when you have many categories to compare.
• Symmetrical layout of streamgraphs makes it simpler to notice shifts in the data and understand the - overall flow of information.
• Relative Trends: Display how different parts contribute to the whole.
• Aesthetic Appeal: Create visually engaging and artistic representations.

Uses:

• Time-based Data: Illustrate data changes over time.
• Multivariate Data: Display multiple categories’ trends simultaneously.
• Narrative Visualization: Tell stories with data’s dynamic evolution.

# remotes::install_github("davidsjoberg/ggstream")
library(ggstream)

head(blockbusters)

ggplot(blockbusters, 
       aes(year, box_office, fill = genre)) +
  geom_stream() +
  scale_fill_manual(values = wes_palette("Darjeeling2")) +
  theme_minimal()

Ridgeline plots

Ridgeline plots, sometimes called joyplots, are like lines that overlap a bit and look like a range of mountains.

Advantages:

• Compact Comparison: Show multiple distributions in a small space.
• Visual Trend Analysis: Quickly spot changes over time or space.
• Group Comparisons: Compare distributions among different groups.

Uses:

• Time Changes: Display how things change over time.
• Spatial Patterns: Visualize variations across different places.
• Group Comparisons: Compare distributions of categories.

# install.packages("ggridges")
library(ggridges)

head(blockbusters)

ggplot(blockbusters, 
       aes(x = box_office, y = genre, fill = genre)) +
  geom_density_ridges(scale = 4) +
  scale_fill_manual(values = wes_palette("Darjeeling2")) +
  theme_minimal()

Sankey diagrams

Sankey diagrams are like maps that show how things move from one place to another. They help us understand where things come from and where they go. These things can be stuff we use, like materials, or money we spend.

library(devtools)
# devtools::install_github("davidsjoberg/ggsankey")
library(ggsankey)

example_dat <-
  mtcars %>%
  make_long(cyl, vs, am, gear, carb) # function in ggsankey to format data correctly

head(example_dat)

ggplot(example_dat,
       aes(x = x, 
           next_x = next_x, 
           node = node, 
           next_node = next_node,
           fill = factor(node))) +
  geom_sankey(flow.alpha = .6) +
  theme_minimal()

Advantages:

• Easy Understanding: Simplify complex information.
• Visual Representation: Clear visual display of resource flows.
• Comparison: Easily compare different resource flows.
• Identify Inefficiencies: Spot wastage or inefficiencies.

Uses:

• Energy Analysis: Show energy usage and losses.
• Material Flows: Track materials in industries, reduce waste.
• Environmental Impact: Display pollution or environmental effects.
• Cost Analysis: Visualize cost distribution and optimization.
• Water Management: Demonstrate water sourcing and use.
• Supply Chain: Understand goods and services flow.

# install.packages("ggalluvial")
library(ggalluvial)

head(UCBAdmissions)

## , , Dept = A
## 
##           Gender
## Admit      Male Female
##   Admitted  512     89
##   Rejected  313     19
## 
## , , Dept = B
## 
##           Gender
## Admit      Male Female
##   Admitted  353     17
##   Rejected  207      8
## 
## , , Dept = C
## 
##           Gender
## Admit      Male Female
##   Admitted  120    202
##   Rejected  205    391
## 
## , , Dept = D
## 
##           Gender
## Admit      Male Female
##   Admitted  138    131
##   Rejected  279    244
## 
## , , Dept = E
## 
##           Gender
## Admit      Male Female
##   Admitted   53     94
##   Rejected  138    299
## 
## , , Dept = F
## 
##           Gender
## Admit      Male Female
##   Admitted   22     24
##   Rejected  351    317

ggplot(as.data.frame(UCBAdmissions),
       aes(y = Freq, axis1 = Gender, axis2 = Dept)) +
  geom_alluvium(aes(fill = Admit), width = 1/12) +
  scale_fill_manual(values = wes_palette("Darjeeling2")) +
  theme_minimal()

Bump charts

Bump Chart is a special form of a line plot. This chart is well-suited for exploring changes in rank over time.

Advantages:

• Rank Insight: Clearly show changes in rank order over time.
• Focused Comparison: Highlight individual entities’ performance shifts.
• Simplicity: Visualize complex rank changes in a straightforward way.

Uses:

• Temporal Changes: Explore how rankings change over time.
• Competitive Analysis: Compare entities’ performance shifts.
• Identifying Trends: Spot trends in rank changes for specific items.

# devtools::install_github("davidsjoberg/ggbump")
library(ggbump)

blockbusters2 <-
  blockbusters %>% 
  filter(genre %in% c("Action", "Comedy", "Drama")) %>% 
  group_by(year) %>% 
  mutate(rank = rank(box_office))

head(blockbusters2)

ggplot(blockbusters2, 
       aes(year, rank, color = genre)) +
  geom_point(size = 7) +
  geom_bump() +
  scale_color_manual(values = wes_palette("Darjeeling2")) +
  theme_minimal()

Waffle charts

Waffle charts are similar to pie charts, but they use squares instead of slices. It’s like a big square made of 100 smaller squares in a 10-by-10 pattern. We color these squares to show different proportions, just like we color slices in a pie chart.

Advantages:

• Visual Simplicity: Easily show proportions using colored squares.
• Comparison: Quickly compare proportions between categories.
• Space-Efficient: Fit a lot of information in a small space.

Uses:

• Simple Percentages: Illustrate simple percentage breakdowns.
• Visualizing Goals: Display progress toward a goal in a visual way.
• Small Data Sets: Effective for smaller datasets with a few categories.

# install.packages("waffle", repos = "https://cinc.rud.is")
# remotes::install_github("hrbrmstr/waffle")
library(ggplot2)
library(waffle)
expenses <- c(`Infants: <1(16467) `=16467, `Children: <11(30098) `=30098,
              `Teens: 12-17(20354)`=20354, `Adults:18+(12456) `=12456,
              `Elderly: 65+(12456) `=12456)
head(expenses)

##   Infants: <1(16467)  Children: <11(30098)    Teens: 12-17(20354) 
##                 16467                 30098                 20354 
##    Adults:18+(12456)   Elderly: 65+(12456)  
##                 12456                 12456

waffle(expenses/1000, rows=5, size=0.6, 
       colors=c("#44D2AC", "#E48B8B", "#B67093", 
                "#3A9ABD", "#CFE252"), 
       title="Age Groups bifurcation", 
       xlab="1 square = 1000 persons")

Beeswarm charts

Beeswarm plot is a type of scatter plot that is used for representing categorical values.

Advantages:

• Category Clarity: Clearly display individual data points within categories.
• No Overlaps: Prevent overlap of data points for better visibility.
• Space-Efficient: Efficiently represent data points without clutter.

Uses:

• Categorical Data: Display data for different categories.
• Individual Points: Show data points distinctly without merging.
• Sparse Data: Effective for datasets with limited data points.

# install.packages("ggbeeswarm")
library(ggbeeswarm)

head(blockbusters)

ggplot(blockbusters, 
       aes(x = genre, y = box_office, color = genre)) + 
  geom_quasirandom() +
  theme_minimal() +
  scale_color_manual(values = wes_palette("Darjeeling2")) +
  theme_minimal()

Mosaic charts

Mosaic plot is like a special stacked bar chart. It’s used to show how things are divided into groups, using percentages. This plot is a picture of a table that compares different things.

Advantages:

• Visual Comparison: Easily compare proportions in different groups.
• Contingency Insight: Show relationships between two categorical variables.
• Space-Efficient: Convey complex information using a compact layout.

Uses:

• Categorical Comparison: Compare percentages across different categories.
• Association Analysis: Visualize relationships between two categorical factors.
• Table Representation: Represent contingency tables graphically.

# devtools::install_github("haleyjeppson/ggmosaic")
library(ggmosaic)

head(UCBAdmissions)

## , , Dept = A
## 
##           Gender
## Admit      Male Female
##   Admitted  512     89
##   Rejected  313     19
## 
## , , Dept = B
## 
##           Gender
## Admit      Male Female
##   Admitted  353     17
##   Rejected  207      8
## 
## , , Dept = C
## 
##           Gender
## Admit      Male Female
##   Admitted  120    202
##   Rejected  205    391
## 
## , , Dept = D
## 
##           Gender
## Admit      Male Female
##   Admitted  138    131
##   Rejected  279    244
## 
## , , Dept = E
## 
##           Gender
## Admit      Male Female
##   Admitted   53     94
##   Rejected  138    299
## 
## , , Dept = F
## 
##           Gender
## Admit      Male Female
##   Admitted   22     24
##   Rejected  351    317

ggplot(as.data.frame(UCBAdmissions)) +
  geom_mosaic(aes(x = product(Admit, Dept), fill = Gender, weight = Freq)) +
  scale_fill_manual(values = wes_palette("Darjeeling2")) +
  theme_minimal()