You want to add marginal rugs to a scatter plot.

Use geom_rug(). For this example (Figure 5-28), we’ll use the faithful data set, which contains data about the Old Faithful geyser in two columns—eruptions, which is the length of each eruption, and waiting, which is the length of time to the next eruption:

ggplot(faithful, aes(x=eruptions, y=waiting)) + geom_point() + geom_rug()

Figure 5-28. Marginal rug added to a scatter plot

A marginal rug plot is essentially a one-dimensional scatter plot that can be used to visualize the distribution of data on each axis.

In this particular data set, the marginal rug is not as informative as it could be. The resolution of the waiting variable is in whole minutes, and because of this, the rug lines have a lot of overplotting. To reduce the overplotting, we can jitter the line positions and make them slightly thinner by specifying size (Figure 5-29). This helps the viewer see the distribution more clearly: