Life expectancy vs. GDP per capita over time

AE 08

Setup

library(tidyverse)
library(tidymodels)
library(distributional)
library(ggdist)
library(ggthemes)

Data

Load the data and filter to every 10 years starting in 1955. Also, reorder the factor levels of continent so that they are in a more intuitive order (Africa, Asia, Europe, North America, South America, Oceania).

# add code here

Models over years

First, in a single pipeline, fit a linear model of life expectancy as a function of log GDP per capita for each continent and year. Then, extract the slope estimates and their standard errors using broom::tidy(). Finally, calculate the 95% confidence intervals for the slope estimates.

Print the resulting object.

# add code here

Then, plot the slope estimates and their 95% confidence intervals using ggplot2::geom_pointrange(). Use position_dodge() to avoid overlapping error bars. Add a horizontal dashed line at y = 0 to indicate where the slope is not significantly different from zero. Reorder the legend so that it is in the same order as the continents in the plot. Finally, use ggthemes::scale_color_colorblind() to make the colors more accessible.

# add code here

Focus on 2025

Filter the estimates to only include the year 2025 and reorder the factor levels of continent based on the slope estimates for that year, so that the continents with higher slope estimates will be at the top of the plots.

# add code here

Half-eye plots

Create half-eye plots of the slope estimates and their uncertainty using ggdist::stat_halfeye().

Hint

Use the dist_normal() function from the distributional package to specify the normal distribution for each estimate, with the mean equal to the slope estimate and the standard deviation equal to the standard error.

# add code here

Gradient interval

Create gradient interval plots of the slope estimates and their uncertainty using ggdist::stat_gradientinterval().

# add code here

Dots interval

Create dots interval plots of the slope estimates and their uncertainty using ggdist::stat_dotsinterval().

# add code here