looking into Tidy Tuesday Mystery

tidyTuesdayMystery

library(dplyr)
library(ggplot2)

setwd("~/Desktop/tidytuesday/data/2020/2020-06-30")

comic_bechdel <- read.csv("comic_bechdel.csv", stringsAsFactors = F)

brks <- c(0, 0.25, 0.5, 0.75, 1)

dim(comic_bechdel) ## 308 9

comic_bechdel2 <- comic_bechdel %>%
  filter(!is.na(pass_bechdel)) %>%
  group_by(series) %>%
  mutate(count = n()) %>%
  group_by(series, pass_bechdel) %>%
  summarise(count_group = n(), prop = count_group / count, percent = round(prop * 100))

dim(comic_bechdel2) ## 307 5 this doesn't seem right

ggplot(comic_bechdel2, aes(x = series, y = count_group, fill = pass_bechdel)) + geom_bar(position = "fill", stat = "identity") + scale_y_continuous(breaks = brks) + coord_flip()

comic_bechdel2b <- comic_bechdel %>%
  filter(!is.na(pass_bechdel)) %>%
  group_by(series) %>%
  mutate(count = n()) %>%
  group_by(series, pass_bechdel) %>%
  summarise(count_group = n(), prop = count_group /count[1], percent = round(prop * 100))
  ## note count[1] all of count will be the same number within a group

dim(comic_bechdel2b) ## 13 5 this is what we expect

ggplot(comic_bechdel2b, aes(x = series, y = count_group, fill = pass_bechdel)) + geom_bar(position = "fill", stat = "identity") + scale_y_continuous(breaks = brks) + coord_flip() ## fixed!