Edward R. Tufte- The Marginal Histogram Scatter Plot
1) This week is a bit of a weird one in my opinion. I have learned about Edward R. Tufte posts about using different visuals in R and found some different ways to visualize data, as well as experimenting with them. He covers a lot of different visuals that you can create with R, so if you want to check out what he has then, the link is here. Our goal for this week is to recreate one of his graphics in R and review it. Without further ado lets get to it.
2) The graphic that I am goign to cover from Tufte's webpage is one of the marginal histogram scatter plots. Specifically the one with ggMarginal. I have never seen a data visual like that before so it should be interesting for me, and hopefully for you as well.
> #LIS 4317 Module 11 Assingment
>
> #setting things up
> #Install packages
> #install.packages(c("CarletonStats", "devtools", "epanetReader", "fmsb", "ggplot2", "ggthemes", "latticeExtra", "MASS", "PerformanceAnalytics", "psych", "plyr", "prettyR", "plotrix", "proto", "RCurl", "reshape", "reshape2"))
> #install.packages("ggExtra")
>
>
>
>
>
> #Load packages
> library(CarletonStats)
> library(devtools)
> library(epanetReader)
> library(fmsb)
> library(ggplot2)
> library(ggthemes)
> library(latticeExtra)
> library(MASS)
> library(PerformanceAnalytics)
> library(psych)
> library(plyr)
> library(prettyR)
> library(plotrix)
> library(proto)
> library(RCurl)
> library(reshape)
> library(reshape2)
> library(ggExtra)
>
> # Example Data
> x <- 1967:1977
> y <- c(0.5,1.8,4.6,5.3,5.3,5.7,5.4,5,5.5,6,5)
>
> # Create the Marginal Histogram scatter plot
> scatter_plot <- ggplot(data.frame(x, y), aes(x = x, y = y)) +
+ geom_point(color = "blue", size = 3) +
+ theme_minimal() +
+ xlab("Year") +
+ ylab("Per capita budget expenditures") +
+ ggtitle("Per capita budget expenditures in constant dollars")
>
> # Add marginal histograms
> scatter_plot_marginal <- ggExtra::ggMarginal(scatter_plot, type = "histogram", fill = "lightblue")
>
> # Print the plot
> print(scatter_plot_marginal)
| |
|
3) I had install and load a ton of packages for this one, but the code came out relatively compact for what it achieved. The marginal histogram scatter plot created here is actually a powerful visualization tool because it combines the strengths of a scatter plot with marginal histograms. It provides more insight than a regular scatter plot, showing trends, outliers, and distributions. While boxplots or density plots can also show distributions, the marginal histogram is more intuitive and integrated, achieving a seamless exploration of data, though it may not suit very large datasets or complex statistical needs. Which is why I gave it a pretty small amount of data. With a large data set the graphic would be too crowded and become too noisy to interpret. Overall, this graphic is pretty cool, but not scalable, and very niche.
Comments
Post a Comment