R Notebook

rm(list = ls())

Sys.setenv(LANG = "en")
library(xlsx)
library(readr)
library(readxl)
library(tidyverse)
library(stringr)
library(stringi)
library(gghighlight)
library(countrycode)
library(countries)
library(janitor)
library(sf)

library(janitor)
library(sf)
library(ggthemes)
library(here)
library(viridis)
library(ggplot2)
library(formatR)
library(gtools)
library(nnet) 
library(patchwork)
library(ggh4x)

library(plm)
library(broom)
library(ggalluvial)

library(pals)

Load data —————————————————————

Large datasets (move load to different notebook)

aggregates <- read_csv("../aggregated_data/aggregates.csv")
gd_world_or <- read_sf("../aggregated_data/TM_WORLD_BORDERS-0.3")

Figure 2A. Map

gd_world <- gd_world_or %>%
  select(ISO3, LON, LAT) %>% 
  filter(ISO3 != "ATA")

corpus_shape_sel <- left_join(aggregates, gd_world, by = join_by(ISO3))

(map_repr_s <- corpus_shape_sel %>% 
    ggplot(aes(geometry = geometry)) +
    geom_sf(aes()) +
    coord_sf(datum = NA) + 
    geom_sf(aes(fill = repr_s)) +
    theme_map() +
    theme(legend.title = element_text(size = 10),
          legend.text = element_text(size = 9),
          legend.position = "bottom",
          plot.title = element_text(hjust = 0.5, size = 14),
          plot.subtitle = element_text(hjust = 0.5, size = 12)) +
   labs(fill = "Research representation index") +
    scale_fill_manual(values = c("#d1bfa3", "#6baed6", "#e34a33"), na.value = "gray"))

ggsave("../output_figures/map_repr_s.png", width = 50, height = 25, units = "cm", dpi = 300, scale = 1, plot = last_plot())

p_legend <- aggregates %>% 
    ggplot(aes(x = log(emig_immig), y = log(art_country))) +
    geom_point(aes(color = repr_s), alpha = .6) +
    scale_color_manual(values = c("#d1bfa3", "#6baed6", "#e34a33"), na.value = "gray") +
    labs(color = "Research representation") +
    theme_minimal() +
    theme(
      legend.title = element_text(size = 10),
      legend.text = element_text(size = 9),
      legend.key.size = unit(0.6, "lines"),
      legend.spacing = unit(0.3, "lines"),
      legend.box.spacing = unit(0.3, "lines"),
      legend.background = element_blank(),
      legend.box.background = element_blank())

only_legend <- cowplot::get_legend(p_legend)

Figure 2B. Scatter

World and subregions

(scat_plot <- aggregates %>% 
    filter(!is.na(repr_s), mig_rel_index <= 100) %>% 
    ggplot(aes(x = log(emig_immig), y = log(art_country))) +
    geom_point(aes(color = repr_s), alpha = .6) +
    scale_color_manual(values = c("#d1bfa3", "#6baed6", "#e34a33")) +
    labs(color = "Research representation",
         size = "Relative migration index",
         x = "Immigration + emigration stock (log)",
         y = "Research salience (log)") + 
  theme_minimal() +
  theme(legend.position = "bottom",
        legend.title = element_text(size = 10)))

ggsave("../output_figures/scatter_world_repr_s.png", width = 50, height = 25, units = "cm", dpi = 300, scale = 1, plot = last_plot())

Continents. Selected version

design = "
#A
BC
DE
"

(scatter_continents_repr_s <- scat_plot + 
  geom_text(data = subset(aggregates, repr_s == "Under" & continent != "Europe"),
            aes(x = log(emig_immig), y = log(art_country), label = ISO3),
            size = 3.5, check_overlap = T) +
    geom_text(data = subset(aggregates, repr_s == "Under" & continent == "Europe"),
            aes(x = log(emig_immig), y = log(art_country), label = ISO3),
            size = 3, check_overlap = T) +
    geom_text(data = subset(aggregates, ISO3 == "USA"),
              aes(x = log(emig_immig), y = log(art_country), label = ISO3),
              size = 3.5, check_overlap = T) +
  ggh4x::facet_manual(~continent, design = design)) 

ggsave("../output_figures/scatter_continents_repr_s.png", width = 50, height = 25, units = "cm", dpi = 300, scale = 1, plot = last_plot())

Combine figures 2A and 2B

# Remove legend before combining plots.
(map_repr_s_new <- map_repr_s + theme(legend.position = "none"))

scatter_continents_repr_s <- scatter_continents_repr_s + theme(legend.position = "none")

combined_map_scatter <- map_repr_s_new / scatter_continents_repr_s

combined_map_scatter + 
  inset_element(
    only_legend,
    left = 0.02, bottom = 0.9, right = 0.2, top = .9, 
    align_to = 'panel'
  )

ggsave("../output_figures/figure_2.png", width = 50, height = 25, units = "cm", dpi = 500, scale = 1, plot = last_plot())

Appendix

Figure S2. Relative migration.

design = "
#A
BC
DE
"

(scat_plot_rel <- aggregates %>% 
    filter(!is.na(repr_rel_s)) %>% 
    ggplot(aes(x = log(mig_rel_index), y = log(art_country))) +
    geom_point(aes(color = repr_rel_s)) +
   geom_text(data = subset(aggregates, repr_rel_s == "Under"),
            aes(x = log(mig_rel_index), y = log(art_country), label = ISO3),
            check_overlap = T) +
    scale_color_manual(values = c("#f0e6d2", "#6baed6", "#e34a33")) +
    labs(color = "Research representation",
         x = "Relative immigration + emigration (log)",
         y = "Research salience (log)") + 
    theme_minimal() +
    theme(legend.position = c(0.2, 0.9)) +
  ggh4x::facet_manual(~continent, design = design)) 

ggsave("../output_figures/scatter_continents_repr_rel_s.png", plot = last_plot())