# use suppressPackageStartupMessages() to clean up output when loading packages
suppressPackageStartupMessages(library(tidyverse))
suppressPackageStartupMessages(library(gapminder))
suppressPackageStartupMessages(library(tsibble))
suppressPackageStartupMessages(library(DT))

This worksheet contains exercises aimed for practice with dplyr.

    1. What’s the minimum life expectancy for each continent and each year? (b) Add the corresponding country to the tibble, too. (c) Arrange by min life expectancy.
# whatever you group by in group_by, is what summarise will summarise.
# when you make columns with summarize, they can be immediately used in the same summarise function
gapminder %>% 
  group_by(continent, year) %>% 
  summarize(min_life = min(lifeExp),
            country = country[lifeExp == min_life]) %>%
    arrange(min_life)
## # A tibble: 60 x 4
## # Groups:   continent [5]
##    continent  year min_life country     
##    <fct>     <int>    <dbl> <fct>       
##  1 Africa     1992     23.6 Rwanda      
##  2 Asia       1952     28.8 Afghanistan 
##  3 Africa     1952     30   Gambia      
##  4 Asia       1957     30.3 Afghanistan 
##  5 Asia       1977     31.2 Cambodia    
##  6 Africa     1957     31.6 Sierra Leone
##  7 Asia       1962     32.0 Afghanistan 
##  8 Africa     1962     32.8 Sierra Leone
##  9 Asia       1967     34.0 Afghanistan 
## 10 Africa     1967     34.1 Sierra Leone
## # … with 50 more rows
  1. Calculate the growth in population since the first year on record for each country by rearranging the following lines, and filling in the FILL_THIS_IN. Here’s another convenience function for you: dplyr::first().
mutate(rel_growth = FILL_THIS_IN) %>% 
arrange(FILL_THIS_IN) %>% 
gapminder %>% 
DT::datatable()
group_by(country) %>% 
# use first to get the population for the first year on record (current order, can use order_by to specify a different order than the current one)
gapminder %>%
    group_by(country) %>%
    arrange(year) %>%
    mutate(rel_growth = pop - first(pop)) %>%
    DT::datatable()
  1. Determine the country that experienced the sharpest 5-year drop in life expectancy, in each continent, sorted by the drop, by rearranging the following lines of code. Ensure there are no NA’s. Instead of using lag(), use the convenience function provided by the tsibble package, tsibble::difference():
drop_na() %>% 
ungroup() %>% 
arrange(year) %>% 
filter(inc_life_exp == min(inc_life_exp)) %>% 
gapminder %>% 
mutate(inc_life_exp = FILL_THIS_IN) %>% 
arrange(inc_life_exp) %>% 
group_by(country) %>% 
group_by(continent) %>% 
knitr::kable()
# difference() takes vector minus its lag, equivalent to lifeExp - lag(lifeExp)
# arrange works INDEPENDENTLY OF GROUPS
# have to use .group_by = TRUE to take group into account
gapminder %>%
    group_by(country) %>% 
    arrange(year) %>%
    mutate(inc_life_exp = difference(lifeExp)) %>%
    drop_na() %>% 
    ungroup() %>% 
    group_by(continent) %>% 
    filter(inc_life_exp == min(inc_life_exp)) %>% 
    arrange(inc_life_exp) %>%
    knitr::kable()
country continent year lifeExp pop gdpPercap inc_life_exp
Rwanda Africa 1992 23.599 7290203 737.0686 -20.421
Cambodia Asia 1977 31.220 6978607 524.9722 -9.097
El Salvador Americas 1977 56.696 4282586 5138.9224 -1.511
Montenegro Europe 2002 73.981 720230 6557.1943 -1.464
Australia Oceania 1967 71.100 11872264 14526.1246 0.170