Lecture 2: Tasks Solutions

Your Turn #1

Question 1

library(tidyverse)

global_working_hours <- read.csv("https://www.dropbox.com/scl/fi/ln4ie7ik36opjx5juifqs/hours_worked_clean.csv?rlkey=ry52v43qyxhtr4ye461twkgnj&dl=1")

global_working_hours |> 
  summarise(mean_worked = mean(hours_worked),
            mean_worker = mean(hours_worker))

The large difference is driven by the fact that hours_worked corresponds to hours worked divided by the adult population, and a share of adults don’t work.

Question 2

global_working_hours |> 
  summarise(median_worked = median(hours_worked),
            median_worker = median(hours_worker))

The medians are quite similar to the means in question 1 above. This suggests the distribution of working hours across countries is relatively symmetric.

Question 3

quantile(global_working_hours$hours_worker)

      0%      25%      50%      75%     100% 
23.90819 35.52082 40.32315 43.68666 56.14137 

The 75th percentile is around 43.5.

Question 4

global_working_hours |> 
  summarise(iqr_men = IQR(hours_worker_men),
            iqr_women = IQR(hours_worker_women))

The dispersion in hours per worker is wider for women than for men.

Question 5

global_working_hours |> 
  summarise(cor(hours_worked, tax_labor, use = "complete.obs"))

The correlation is -0.5 which means that an increase in the labor tax rate is associated with a reduction in (average) hours worked.

Your Turn #2

Question 1

global_working_hours_panel <- read.csv("https://www.dropbox.com/scl/fi/aq6nlnuun9o8bk86h89a2/hours_worked_panel_clean.csv?rlkey=f6jqgl8swkvf6kgl1z9eqczmr&dl=1")

global_working_hours_panel |>
  count(year)

Question 2

global_working_hours_panel |> 
  filter(year == 2023 & hours_worker > 48)

Question 3

global_working_hours_panel |> 
  filter(hours_worker > 30 & hours_worker < 32 & region == "Latin America")

Question 4

global_working_hours_panel |> 
  filter(region %in% c("Western Europe and Anglosphere", "United States") &
           year == 2023) |> 
  summarise(max_hours_worker = max(hours_worker),
            .by = region)

Question 5

global_working_hours_panel |> 
  filter(year %in% c(1990, 2000, 2010, 2020)) |> 
  mutate(diff_men_women = hours_worked_men - hours_worked_women) |> 
  summarise(mean_diff = mean(diff_men_women),
            .by = year)

The difference in hours worked between men and women has decreased markedly over time (from ~ 15h to ~10h).

Your Turn #3

Question 1

global_working_hours_panel |> 
  filter(year == 2022) |> 
  ggplot(aes(x = hours_worker_women)) +
  geom_histogram(binwidth = 5, boundary = 30, color = "white", fill = "#785EF0")

• binwidth: width of each bin
• boundary: value at which one of the bins starts
• color: color of bar contours
• fill: color of bars

Question 2

global_working_hours_panel |> 
  summarise(mean_hours_worked = mean(hours_worked),
            .by = c(year, region)) |> 
  ggplot(aes(x = region, y = mean_hours_worked)) +
  geom_boxplot(color = "black", fill = "#785EF0")

Question 3

global_working_hours_panel |>
  ggplot(aes(x = hours_worker_men, y = hours_worker_women)) +
  geom_point(size = 2, alpha = 0.25, colour = "#d90502") +
  labs(x = "Hours per worker - Men",
       y = "Hours per worker - Women",
       title = "Relationship between male and female hours per worker")

Warning: Removed 61 rows containing missing values or values outside the scale range
(`geom_point()`).