#TidyTuesday World Mortality

mortality.png

title: "R Notebook for Tidy Tuesday" output: word_document: default html_document:

df_print: paged

This is an R Markdown Notebook for the #TidyTuesday challenge: a weekly social data project (in R), which builds off #makeovermonday style projects but aimed at the R ecosystem. An emphasis will be placed on understanding how to summarize and arrange data to make meaningful charts with ggplot2, tidyr, dplyr, and other tools in the tidyverse ecosystem.

This code is for Week 3, and my effort to recreate

this image, from ourwordindata.org, which generally looks like a colorful histogram tipped on its side

library(tidyverse, quietly = TRUE, warn.conflicts = FALSE) 
library(readxl) # To read in Excel data the tidyverse way
library(viridisLite) #awesome color-blind friendly palette

#import data

mortality <- read_excel("data/global_mortality.xlsx")
mortality

#clean up the df with labels that match the graphs

names(mortality) <- gsub(" (%)", "", names(mortality), fixed = TRUE)
#took a long time to figure this step out; still unsure why fixed must == true, but that was the trick to make it work.
names(mortality)

mortality

#Annual number of deaths by cause, World, 2016 First going to try to get to annual number of deaths across the world, filtered for 2016. Looking at data above, I'm notice two things:

  1. World is one of the 'country' observations, so won't need to add up them all (as I thought)
  2. This excel file only has %s, so going to need world populations if I want to make the first graphs, which has total numver of deaths/year. Not sure where to find this yet, so will move on to graph #2, which is "share of deaths, by cause, World, 2016"

#share of deaths, by cause, World, 2016 All of these data are available in the df, so will start here.

graph2 <- mortality %>% 
  filter(country == "World", year == 2016) %>%  #get selected observations
  mutate_at(vars(`Cardiovascular diseases`:Terrorism), funs(. / 100)) %>%  #convert to percent
  gather(key = disease, value = percent, `Cardiovascular diseases`:Terrorism) %>%  #I want to graph disease vs. percent
  arrange(desc(percent)) #probably don't need this
graph2 #check your work

#make plot

ggplot(graph2, aes(x = reorder(disease, percent), y = percent, fill = disease)) + #use reorder to rank these high to low.
  geom_col() +
  geom_text(aes(y = percent, label=paste(round(percent*100,2),"%")),
            hjust = -0.2, size = 3, color = "grey30") + # set up labels on bars, and round in aes!!! thanks @dylanjm_ds for the pro tip
  labs(title = "Share of deaths by cause, World, 2016", x = "", y = "") +
  coord_flip() + 
  scale_y_continuous(limits = c(0,0.35), breaks=c(0,.05,.10,.15,.20,.25,.30), minor_breaks = NULL, labels = scales::percent) +
  guides(fill = FALSE) +
  scale_fill_viridis(discrete=TRUE) +
  theme_minimal()

#save

ggsave("mortality.png", last_plot(), height = 8, width = 12, units = "in", dpi = 600)
#default is last plot, but can name objects here

#conclusions Ran out of time to make it prettier; would have added thicker vertical lines to the major breaks, maybe tweaked the colors to channel my inner rainbow, added subtitle and sub-subtitle (what do you call the text at the bottom left?), and maybe tweaked the aestehtics a bit, make the minor grids not intersect with the labels.

Surprised how long these still take! But learned:

  • paste() in aes
  • if not fill() in aes, then no color!