How to Make a Scatter Plot Matrix in R

Scatter plot matrix is a plot that generates a grid of pairwise scatter plots for multiple numeric variables. Creating a scatter plot matrix can be a useful way to visually explore relationships between several numeric variables quickly.

Code used in this code clip:

library(tidyverse)
library(GGally)

data ＜- mtcars

# Scatter plot matrix in base R
pairs(~ mpg + hp + cyl + wt, data = data,

# Scatter plot matrix with GGally
data %＞% ggpairs(columns = c("mpg", "hp", "cyl", "wt"),
upper = list(continuous = wrap('cor', size = 8)))

# Scatter plot matrix with GGally, coloring by cyl
data %＞% mutate(cyl = factor(cyl)) %＞%
ggpairs(columns = c("mpg", "hp", "wt", "cyl"),
aes(color = cyl),
upper = list(continuous = wrap('cor', size = 5)),
lower = list(combo = wrap("facethist", bins = 30)),
diag = list(continuous = wrap("densityDiag", alpha = 0.5)))

Code Clips are basic code explanations in 3 minutes or less. They are intended to be short reference guides that provide quick breakdowns and copy/paste access to code needed to accomplish common data science tasks. Think Stack Overflow with a video explanation.

* Note: YouTube does not allow greater than or less than symbols in the text description, so the code above may not be exactly the same as the code shown in the video! For R that means I may use = for assignment and the special Unicode large ＜ and ＞ symbols in place of the standard sized ones for dplyr pipes and comparisons. These special symbols should work as expected for R code on Windows, but may need to be replaced with standard greater than and less than symbols for other operating systems.

[JapaneseQuest]

I am liking your thumbnails lately -- so colorful. :)

[cristobaljullianfigueroa]

very simple and success!, thanks from Chile!

[josephbrockly-anderson]

This is exactly what I needed to find to finish a project of mine. Thank you for the info and the breakdown. Is there any way to change the title of a variable once it's in the matrix? So maybe instead of "cyl" it appears on the graph as "Cylinders" or something like that?

[katarinahocevar]

Hello, thanks for a very helpful video. I am using last code (Scatter plot matrix with GGally), for my data, coloring by variable with two levels. Is it possible to change a color of levels (e.g. from default option red and blue to green and purple)? Best regards.

[sabamajeed]

The last code is not visible after alpha, can you please specify what is written. Thank you for explaining the over all code it is very helpful

[dr.md.sabibulhaque]

how to put legends in this scatter plot?

[alexdufur]

Hello, I tried to run
# Scatter plot matrix with GGally
data %＞% ggpairs(columns = c("mpg", "hp", "cyl", "wt"),
upper = list(continuous = wrap('cor', size = 8))) but it gives me a weird error "could not find function "ggpairs". I installed all the packages needed.

[dilshadsaeed]

duhok %>% mutate(rain = factor(rain))%>%
+ ggpairs(columns = c("rain", "Pre", "Min", "Max", "Hum", "Win"),
+ aes(color = rain),
+ upper = list(continuous = wrap('cor', size = 5)),
+ lower = list(combo = warp("facethist", bins = 30)),
+ diag = list(continuous = wrap("densityDiag", alpha = 0.5)))
Error in stop_if_high_cardinality(data, columns, cardinality_threshold) :
Column 'rain' has more levels (432) than the threshold (15) allowed.
Please remove the column or increase the 'cardinality_threshold' parameter. Increasing the cardinality_threshold may produce long processing times.

Please how I can fix it. thankyou