Star Wars Survey Coding Challenge

Show the code

#load your libraries in this chunk
library(tidyverse)
dat_multi <- read_csv("https://github.com/byuidatascience/data_fivethirtyeight_starwars/raw/master/survey_multiresponse_questions.csv")
dat_single <- read_csv("https://github.com/byuidatascience/data_fivethirtyeight_starwars/raw/master/survey_oneresponse_questions.csv")
dat_wide <- read_csv("https://github.com/byuidatascience/data_fivethirtyeight_starwars/raw/master/survey_wide_all.csv")

The challenge

In a timed coding challenge, I had to clean FiveThirtyEight’s Star Wars fan survey and recreate two charts from their article. The survey ships as three messy files: multi-response questions, single-response questions, and a wide table with long, machine-generated column names. The work was to rename and select the useful fields, filter to respondents who had seen all six films, derive each person’s favorite film from their rankings, and rebuild the published charts.

Cleaning the survey

Show the code

single <- dat_single %>% 
  rename(
    ID = respondentid,
    exposure = seen_any6,
    fan = fan_yn,
    expanded = fan_expanded,
    startrek = fan_startrek) %>% 
  select(c(1,7,8,9,10,11)) %>% 
  drop_na(shot_first)
wide <- dat_wide %>% 
  rename(
    ID = respondentid,
    exposure = seen_any6,
    fan = fan_yn,
    E1_Rank = "rank_1f6w__StarWars:EpisodeIThePhantomMenace",
    E2_Rank = "rank_1f6w__StarWars:EpisodeIIAttackoftheClones",
    E3_Rank = "rank_1f6w__StarWars:EpisodeIIIRevengeoftheSith",
    E4_Rank = "rank_1f6w__StarWars:EpisodeIVANewHope",
    E5_Rank = "rank_1f6w__StarWars:EpisodeVTheEmpireStrikesBack",
    E6_Rank = "rank_1f6w__StarWars:EpisodeVIReturnoftheJedi") %>% 
  select(c(1,2,3,10,11,12,13,14,15)) %>% 
  filter(exposure == "Yes") %>% 
  drop_na()

Deriving each respondent’s favorite film

Show the code

best <- wide %>% 
    mutate(best = case_when(
      E1_Rank == 1 ~ "The Phantom Menace",
      E2_Rank == 1  ~ "Attack of the Clones",
      E3_Rank == 1 ~ "Revenge of the Sith",
      E4_Rank == 1 ~ "A New Hope",
      E5_Rank == 1  ~ "The Empire Strikes Back",
      E6_Rank == 1 ~ "Return of the Jedi"))

Recreated charts

Show the code

ggplot(best, mapping = aes(best))+
  geom_bar(fill = "blue")+
  coord_flip()+
  labs(title = "What's the Best 'Star Wars' Movie?",
       subtitle = "Of 471 respondents who have seen all six films")

Show the code

ggplot(single, mapping = aes(shot_first))+
  geom_bar(fill = "blue")+
  coord_flip()+
  labs(title = "Who Shot First?",
         subtitle = "According to 834 respondents")