The Ideology of the Head of Government: A Glimpse from Latin American

Data Visualization
Ideology
Latin America
In this post, we take a look at the ideology of heads of government in Latin America
Author

Lucas Couto

Published

March 14, 2024

Ideology

This month’s visualisation is slightly different from what we’ve done so far. Unlike previous months, I did not seek inspiration from submissions to Tidy Tuesday. Specifically, I want to try to do something I miserably failed in recent years. Thus far, I’ve never got to make an animated graph =(. However, this is the month when I’ll actually manage to do it!

To this end, I’ll rely on The Global Leader Ideology dataset from Bastian Herre. From a worldwide perspective, the data identify the ideological position of heads of government and leaders on an economic scale. More specifically, these political actors are classified as “leftist”, “centrist”, or “rightist”. A thorough explanation behind this classification can be further accessed on the dataset’s page. Importantly, a discussion about the coding of each head of government is also available.

As Latin America has been the focus of research, I chose it as the region for which we’ll see the ideological preferences of heads of government. Without further ado, let’s get into the plot.

Animated Map

A few comments are noteworthy:

  1. I changed the colour scheme for the ideological classification when compared to the original graphs, mainly to show rightist leaders in darker blue and leftist leaders in darker red.

  2. Some small Caribbean states are not legible, but this is due to the adopted scale. This can be changed with the parameter alpha within geom_sf.

  3. To make the animated map, I gladly followed many steps from Rodrigo Zamith and Mel Moreno and Mathieu Basille. There are other great guides on the internet, a quick search will make you find a plethora of them!

Regional Trends

Some periods are marked by governments throughout the region leaning towards either the left (the Pink Tide in the early 2000s) or the right, as had been the case lately. A book examining the emergence of the new right in the region is scheduled to be released later this year. If you are interested, save the name:

Borges, André; Lloyd, Ryan; Vommaro, Gabriel, (eds.) The Recasting of the Latin American Right: Polarization and Conservative Reactions. Cambridge University Press.

The Ideology of Heads of Government in Latin America Over the Years

Code

Show the code 
##before you proceed, you should download the data from: https://github.com/bastianherre/global-leader-ideologies


#Load packages
library(gganimate)
library(ggtext)
library(haven)
library(ragg)
library(rmapshaper)
library(rnaturalearth)
library(rnaturalearthdata)
library(showtext)
library(sf)
library("tidylog", warn.conflicts = FALSE)
library(tidyverse)

#load data
ideologue <- read_dta("global_leader_ideologies.dta")

#data wrangling
LA <- ideologue %>%
      filter(country_name %in% c("Argentina", "Bahamas", "Barbados", "Belize", "Bolivia", "Brazil",
                                 "Chile", "Colombia", "Costa Rica", "Cuba", "Dominican Republic", "Ecuador",
                                 "El Salvador", "Guatemala", "Guyana", "Haiti", "Honduras", "Jamaica", "Mexico", "Nicaragua", "Panama", 
                                 "Paraguay", "Peru", "Suriname", "Trinidad and Tobago", "Uruguay", "Venezuela")) %>%
      select("name" = country_name, year, hog, hog_ideology) %>%
      mutate(name = ifelse(name == "Dominican Republic", "Dominican Rep.", name)) %>%
      mutate(hog_ideology = ifelse(hog_ideology == "none", "no ideology", hog_ideology)) %>%
      mutate(hog_ideology = ifelse(hog_ideology == "not applicable", "no information", hog_ideology)) %>%
      mutate(hog_ideology = factor(hog_ideology, levels = c("rightist", "centrist", 
                                                            "leftist", "no ideology", "no information")))

#Maps
world <- ne_countries(scale = "medium", 
                      returnclass = "sf")

sousou <- world %>%
          filter(region_wb == "Latin America & Caribbean") 

sousou_points <- st_centroid(sousou) %>%
                 cbind(st_coordinates(st_centroid(sousou$geometry)))

frier <- sousou_points %>%
          st_drop_geometry() %>%
          select(name, X, Y) %>%
          left_join(LA) %>%
          left_join(sousou) 

#font
font_add(family = "regular", "Outfit-Regular.ttf")
font_add(family = "bold", "Outfit-Bold.ttf")
showtext_auto() 

#plot
frier %>% arrange(year) %>%
          #filter(year == 2015) %>% #if you want to see the data in a particular point in time
          ggplot() +
          geom_sf(data = frier, aes(geometry = geometry), inherit.aes = FALSE) +
          geom_sf(data = frier %>% filter(!is.na(hog_ideology)), 
                  aes(fill = hog_ideology, geometry = geometry), 
                      size = .15, color = "white") +
          geom_sf(data =  frier %>% filter(is.na(hog_ideology)) %>% 
                  mutate(hog_ideology = ifelse(is.na(hog_ideology), "no information", hog_ideology)),
                  aes(fill = hog_ideology, geometry = geometry),
                  size = .15, color = "white") +
          coord_sf(expand = FALSE, clip = "off") +
          scale_fill_manual(values = c("#0000ff", "#A865C9",  "#ff0000", "#000000", "#bebebe"),
                            labels = c("Rightist", "Centrist", "Leftist", "No Ideology", "No Information")) +
          xlab("") + ylab("") +
          labs(title = "Ideology of the Head of Government - Year: {current_frame}",
               caption = "Lucas Couto | Data: Herre (2023)") +
          transition_manual(year) +
          ease_aes('linear') +
          theme(##Background
                panel.border=element_blank(),
                panel.grid.major = element_blank(),
                panel.grid.minor = element_blank(),
                plot.background = element_rect(fill = "#ffffff", linetype = 'blank'),
                #plot.margin = margin( , 0.5, , , "cm"),
                panel.background = element_rect(fill = "#ffffff"),
                plot.caption = element_text(family="regular", size=13, hjust=1.1, vjust=1.5),
                plot.caption.position = "plot",
                ##Axes
                axis.text = element_blank(),
                axis.ticks = element_blank(),
                ##Legend
                legend.text = element_text(size=15, family="regular"),
                legend.position = c(0.1, 0.2),
                legend.key.size = unit(1.0, "cm"),
                legend.spacing.y = unit(0, "cm"),
                 ##Titles
                plot.title=element_text(family = "bold", hjust = 0.0, vjust = 0.0, size = 15, color = "black"),
                plot.title.position = "plot") +
                guides(fill = guide_legend(title = "")) -> p


#The code below will create a folder with several figures in a png format. Combine them in specific sites to transform them into an animated map in GIF!
#animate(p, width=1050, height=750, duration=15, end_pause=5, rrenderer=file_renderer(dir="anim_img/"))