Exploring a data frame

Often the most interesting feature of your data are the relationships between the variables. If there are only a handful of variables saved as columns in a data frame, it is possible to visualize all of these relationships neatly in a single plot.

Base R offers a fast plotting function pairs(), which draws all possible scatter plots from the columns of a data frame, resulting in a scatter plot matrix. Libraries GGally and ggplot2 together offer a slow but more detailed look at the variables, their distributions and relationships.

This exercise is part of the course

Helsinki Open Data Science

View Course

Exercise instructions

Draw a scatter matrix of the variables in learning2014 (other than gender)
Adjust the code: Add the argument col to the pairs() function, defining the colour with the 'gender' variable in learning2014.
Draw the plot again to see the changes.
Access the ggpot2 and GGally libraries and create the plot p with ggpairs().
Draw the plot. Note that the function is a bit slow.
Adjust the argument mapping of ggpairs() by defining col = gender inside aes().
Draw the plot again.
Adjust the code a little more: add another aesthetic element alpha = 0.3 inside aes().
See the difference between the plots?

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# learning2014 is available

# draw a scatter plot matrix of the variables in learning2014.
# [-1] excludes the first column (gender)
pairs(learning2014[-1])

# access the GGally and ggplot2 libraries
library(GGally)
library(ggplot2)

# create a more advanced plot matrix with ggpairs()
p <- ggpairs(learning2014, mapping = aes(), lower = list(combo = wrap("facethist", bins = 20)))

# draw the plot

Edit and Run Code