Distribution plots are great for numeric variables, but we need a different type of plot for categorical variables. Fortunately, we can use the exact same bal.plot() function from cobalt with no need to specify the variable type. By updating the arguments for x and var.name to use graduate, we will get a bar plot to examine balance for the categorical variable graduate.

# import library
library(cobalt)
# plot distributions for stress variable
bal.plot(
  x = meditate ~ graduate, #new formula
  data = sleep_data, #dataset
  var.name = "graduate", #new variable
  colors = c("#E69F00", "#009E73") #set fill colors
)

From this plot, we see that the ratio of undergraduates to graduates is much larger for the meditation group (green) than for the non-meditation group (orange).

Both plots so far suggest that there are differences between the treatment and control groups with respect to the stress and graduate variables. However, balance plots don’t precisely quantify the degree of imbalance in the dataset. To get a more detailed picture, we can check balance numerically.