library(ggplot2)Let us make a data frame with two quantitative variables. We generate these variables using random numbers from normal/gaussian distribution.
set.seed(42) x <- rnorm(2000, mean=15, sd=20) y <- x+ rnorm(2000, mean=2, sd=30) df <- data.frame(x=x, y=y)Let us make a simple scatter plot to illustrate the problem of overplotting
df %>% ggplot(aes(x=x, y=y)) + geom_point()

df %>% ggplot(aes(x=x,y=y)) + geom_point(alpha=0.3)
