Unlocking the power of data visualization with R

library(tidyverse)
tuesdata <- tidytuesdayR::tt_load(2023, week = 04)

survivalists <- tuesdata$survivalists
loadouts <- tuesdata$loadouts
episodes <- tuesdata$episodes
seasons <- tuesdata$seasons

survivalists%>%head

episodes <- episodes%>%
  select(season,air_date,viewers)

viewers <- seasons%>%
  select(lon,lat,location,country,n_survivors,season)%>%
  inner_join(survivalists,by=c("season","country"))%>%
  inner_join(episodes,by=c("season"))%>%
  select(age,n_survivors,viewers,air_date,location,country,city,state)%>%
  # select(lon,lat,location,country,city,state,
  #        name,age,gender,profession,n_survivors,season,
  #        air_date,viewers)%>%
  distinct()%>%#DataExplorer::profile_missing()
  drop_na()
viewers

library(owidR)
# df <- owidR::owid_covid()
# save(df,file="df.RData")
load("df.RData")
df_short <- df%>%
  select(continent,country=location,
         date,total_cases,median_age,life_expectancy)%>%
  filter(country=="Canada")

df_short%>%count(country)
viewers%>%count(country)

viewers%>%
  inner_join(df_short%>%filter(country=="Canada"),
             by=c("air_date"="date","country"))%>%
  select(air_date,n_survivors,viewers,total_cases,median_age,age)%>%
  #filter(!is.na(total_cases))%>%
  distinct()%>%
  count(air_date)

my_df <- viewers%>%
  inner_join(df_short%>%filter(country=="Canada"),
             by=c("air_date"="date","country"))%>%
  select(n_survivors,viewers,total_cases,median_age,age)%>%
  distinct() %>%
  scale()%>%
  as.data.frame()%>%
  select(viewers,total_cases)
my_df

v <- my_df$viewers
c <- my_df$total_cases
cor(v,c)

my_df%>%
  ggplot()+
  geom_density(aes(viewers))+
  geom_density(aes(total_cases),color="red")

library(tidyverse)
library(ggdist)
library(distributional)
library(extrafont)
library(showtext)
showtext_auto()
sysfonts::font_add_google(name="Chelsea Market",
                          family="Chelsea Market")


my_df%>%
  pivot_longer(cols = everything(), 
               names_to = "type" ,values_to = "values")%>%
  mutate(type=ifelse(type=="viewers",
                     "TV series Alone - Viewers",
                     "OWId Covid19 Total Cases in Canada"))%>%
  ggplot(aes(y=type,x=values,
             #xdist=dist_normal(values, sd(values)),
         #xdist = dist_normal(0, 1),
         layout = "weave",
         fill = stat(x > 0)))+
  stat_dots(aes(xdist = dist_normal(0, 1)))+
  ggnewscale::new_scale_fill()+
  stat_dist_slabinterval(aes(x=values),alpha=0.8)+
  geom_vline(xintercept = 0, alpha = 0.25) +
   tvthemes::scale_fill_hilda()+
  labs(title="Canada: TV Series ALONE Viewers and OWId Covid19 Total Cases",
       subtitle="Distributions of standardized values",
       caption = "Correlation doesn't imply causation! These are 11 weeks of Covid-19 (2020-06-11 - 2020-08-20) and ALONE Viewers\nDataSource: #TidyTuesday 2023 week4 Alone data\nDataViz: Federica Gazzelloni #30DayChartChallenge 2023 Day6 - OWId\n")+
  tvthemes::theme_avatar(text.font = "Chelsea Market")+
  theme(text=element_text(),
        plot.title = element_text(hjust = 1),
        plot.caption = element_text(hjust = 1),
        legend.position = "none",
        axis.title = element_blank())

my_df%>%
  mutate(day=seq_along(1:11))%>%
  pivot_longer(cols = c("viewers","total_cases"), 
               names_to = "type" ,values_to = "values")%>%
  mutate(type=ifelse(type=="viewers",
                     "TV series Alone - Viewers",
                     "OWId Covid19 Total Cases in Canada"))%>%
  ggplot(aes(x=day,y=values,group=type,color=type))+
    geom_line()+
  geom_smooth(method = "lm",se=F)+
  tvthemes::scale_color_hilda()+
  theme_bw()+
  theme(legend.position = "top")