Color and shape aesthetics in facet_grid when margins presented

Question

Consider the following data:

n <- 1000  
data <- data.frame(id = 1:n,
                 a = sample(c("a1", "a2"), n, replace = T),
                 b = sample(c("b1", "b2", "b3"), n, replace = T),
                 x = rnorm(n),
                 y = rnorm(n))

I am creating a scatter plot in a grid for each combination of a and b with margins.

library(ggplot2)
  ggplot(data, aes(x = x, y = y)) +
  geom_jitter(aes(color = b, shape = a)) +
  facet_grid(a ~ b, margins = T)

This creates an extra level for both factors called (all), which for me is pointless.

I would like to obtain the effect where points are distinguished by color and shape even on marginal plots.

Answer 1

A solution is to create a variable to do the coloring and faceting independently:

n <- 1000  
data <- data.frame(id = 1:n,
                   a = sample(c("a1", "a2"), n, replace = T),
                   b = sample(c("b1", "b2", "b3"), n, replace = T),
                   x = rnorm(n),
                   y = rnorm(n)) %>%
  mutate(a_ = factor(a),
         b_ = factor(b))

library(ggplot2)
ggplot(data, aes(x = x, y = y)) +
  geom_jitter(aes(color = b, shape = a)) +
  facet_grid(a_ ~ b_, margins = T)