r-over-time.R

library(tidyverse)
library(lubridate)

questions <- readr::read_csv("~/Repositories/stacklite/questions.csv.gz")
question_tags <- readr::read_csv("~/Repositories/stacklite/question_tags.csv.gz")

r_questions <- question_tags %>%
  filter(Tag == "r")

r_questions_by_month <- questions %>%
  semi_join(r_questions, by = "Id") %>%
  mutate(Month = round_date(CreationDate, "month"))

r_questions_by_month %>%
  count(Deleted = !is.na(DeletionDate), Month) %>%
  mutate(Percent = n / sum(n)) %>%
  ggplot(aes(Month, Percent, color = Deleted)) +
  geom_line()