fake-data.r

# reference: https://gist.github.com/andrewheiss/51f60fc2a410ca0d321d942153b955e7

library(tidyverse)
library(wakefield)  # For fancy data generation
library(truncnorm)  # Truncated normal distributions
library(ggridges)   # Ridge plots

# Make all the random draws consistent
set.seed(1234)

# Function to generate GPAs
gpa_centered_on <- function(mean){
  round(rtruncnorm(1, a = 1.0, b = 4.0, mean = mean, sd = .5), 2)
}
  
# Generate fake data
all_students <- 
  r_data_frame(n = 1600, 
               id, 
               race, 
               age(x = 20:30), 
               gender_inclusive) %>% 
  mutate(undergrad_gpa = gpa_centered_on(2.5),
         math_camp = c(rep(T, 800), rep(F, 800)),
         graduate_gpa = gpa_centered_on(3.0 + 0.5*math_camp))


ggplot(all_students, aes(x = graduate_gpa, y = math_camp, fill = math_camp)) +
  geom_density_ridges() +
  guides(fill = FALSE) +
  theme_bw()