## ----------------------------------------------------
dir_path <- "../../slides/datamanaging/data/"
list.files(dir_path)
file_path <- paste0(dir_path, "daily_show_guests.csv")
library(tidyverse)
read_lines(file_path, n_max=10)
daily_show <- read.csv(file_path, skip=4)
str(daily_show)
head(daily_show)

daily_show <- read_csv(file_path, skip=4)
spec(daily_show) ; problems(daily_show)
str(daily_show) ; glimpse(daily_show) ; head(daily_show)

daily_show <- rename(daily_show,
                     year = YEAR,
                     job = GoogleKnowlege_Occupation,
                     date = Show,
                     category = Group,
                     guest_name = Raw_Guest_List)

select(daily_show, job, date, category, guest_name) # or
select(daily_show, -year)

select(daily_show, last_col()) # or
daily_show[,ncol(daily_show)]

arrange(daily_show, category)

arrange(daily_show, desc(date)) 
#arrange(daily_show, desc(year), desc(date)) 

daily_show %>% select(category) %>% arrange(category) %>% unique # or
daily_show %>% pull(category) %>% sort %>% unique # or
daily_show %>% distinct(category) %>% arrange(category) 

scientists <- filter(daily_show, category == "Science")
head(scientists)

scientists %>% count(job, sort=T)

scientists <- scientists %>% filter(job%in%c("neurosurgeon", "surgeon", "Surgeon"))
head(scientists) 




## ------------------------------------------------------
file_path <- paste0(dir_path, "nobel.csv")
read_lines(file_path, n_max=10)
nobel <- read_csv(file_path)
glimpse(nobel)
head(nobel)

nobel %>% distinct(category) %>% arrange(category)

nobel_stem <- nobel %>% filter(category %in% c("Physics", "Medicine", "Chemistry", "Economics"))
nobel_nonstem <- nobel %>% filter(!(category %in% c("Physics", "Medicine", "Chemistry", "Economics")))

nrow(nobel_stem)+nrow(nobel_nonstem)==nrow(nobel)

write_csv(nobel_stem, file=paste0(dir_path, "nobel-stem.csv"))

options(digits=3)

nobel %>% count(gender) %>% mutate(prop = n/sum(n)*100)
# alternatively
nobel %>% group_by(gender) %>% summarize(n = n())

# What is "org"?
nobel %>% filter(gender=="org")
nobel_mf <- nobel %>% filter(gender!="org")

# What is the distribution of "share" between males and females?
nobel_mf %>% count(gender, share) 
nobel_mf %>% group_by(gender) %>% count(share) %>% mutate(prop = n/sum(n)*100)

# What is the distribution of "category" between males and females?
nobel_mf %>% group_by(gender) %>% count(category, sort=T) %>% mutate(prop = n/sum(n)*100, cum_sum = cumsum(prop))
nobel_mf %>% mutate(stem = ifelse(category %in% c("Physics", "Medicine", "Chemistry", "Economics"), "stem", "nonstem")) %>% group_by(gender) %>% count(stem) %>% mutate(prop = n/sum(n)*100)


# Do femal nobel winners increase over time?
nobel %>% filter(gender=="female") %>% count(year) %>% arrange(n)
nobel %>% filter(gender=="female") %>% group_by(year) %>% summarize(n = n())

# ... by age ...