THE 2022 Citaions component – https://twitter.com/ikashnitsky/status/1440676930135621643

the-rank.R

#===============================================================================
# 2021-09-17 -- THE ranking
# Citations component of the ranking
# Ilya Kashnitsky, ilya.kashnitsky@gmail.com, @ikashnitsky
#===============================================================================

library(tidyverse)
library(magrittr)
library(countrycode)
library(sf)
library(spData)
library(rmapshaper)
library(countrycode)
library(ggdark)
library(hrbrthemes)
library(showtext)
library(cowplot)
library(biscale)


# read in the raw data
raw <- read_csv("https://gist.githubusercontent.com/ikashnitsky/4feb1ebe817eefa7f49084010336a876/raw/a108116ea6a507572be9cf5541dfaff5e200a9a4/the-ranking-2022-raw.csv") %>%
    janitor::clean_names() %>%
    filter(!name == "Explore", ! name == "Country/Region")

# clean the data
the22 <- raw %>%
    drop_na() %>%
    mutate(
        country = raw %>% filter(rank %>% is.na) %>% pull(name),
        iso2 = country %>% countrycode("country.name", "iso2c"),
        iso3 = country %>% countrycode("country.name", "iso3c"),
        overall_exact = .3*teaching + .3*research + .3*citations +
            .075*international_outlook + .025*industry_income,
        overall_exact_without_citations = (
            .3*teaching + .3*research +
                .075*international_outlook + .025*industry_income
        ) %>% multiply_by(10/7)
    )

# calculate the ratio of citations to the overall ranking excl citations
ratio <- the22 %>%
    mutate(
        ratio = citations / overall_exact_without_citations,
        ratio_res = citations / research
    ) %>%
    arrange(ratio %>% desc)

# explore the top
ratio %>% transmute(rank, name, iso3, ratio = round(ratio, 1), overall = round(overall_exact, 1), teaching, research, citations, industry_income, international_outlook) %>% view()


# distribution of the components ------------------------------------------

the22 %>%
    ggplot()+
    geom_density(aes(overall_exact), color = "#fafafa", size = 1.5)+
    geom_density(aes(teaching), color = 4, size = 1)+
    geom_density(aes(research), color = 7, size = 1)+
    geom_density(aes(citations), color = 2, size = 1)+
    scale_x_continuous(position = "top")+
    dark_theme_minimal(base_family = font_rc)+
    labs(
        title = "THE 2022 ranking",
        subtitle = "Distributions of the overall value and the three main components",
        caption = "@ikashnitsky",
        y = "Density", x = NULL
    )+
    theme(
        plot.title = element_text(family = "Roboto Slab", face = 2),
        legend.position = c(.8, .6),
        plot.background = element_rect(color = NA)
    )+
    annotate(
        "text", label = c("Overall", "Teaching", "Research", "Citations"),
        color = c("#fafafa", 4, 7, 2), alpha = 3/4, hjust = 0, vjust = 1,
        family = "Roboto Slab", fontface = 2, size = 6,
        x = c(35, 25, 0, 55), y = c(.027, .057, .063, .017)
    )

ggsave(
    filename = "~/components-densities.png",
    width = 6, height = 4.2, type = "cairo"
)

# map ratio ---------------------------------------------------------------

# average ratio per country
ratio_avg <- ratio %>%
    group_by(country, iso2) %>%
    summarise(
        n = n(),
        avg_ratio = ratio %>% mean(),
        avg_ratio_res = ratio_res %>% mean()
    ) %>%
    ungroup()

# get world map outline (you might need to install the package)
world_outline <- spData::world %>%
    st_as_sf() %>%
    # drop Antarctica
    filter(!iso_a2 == "AQ")

# let's use a fancy projection
world_outline_robinson <- world_outline %>%
    st_transform(crs = "ESRI:54030")

# produce borders layer
country_borders <- world_outline_robinson %>%
    rmapshaper::ms_innerlines()

# map avg ratio
world_outline_robinson %>%
    left_join(ratio_avg, by = c("iso_a2" = "iso2")) %>%
    ggplot()+
    geom_sf(aes(fill = avg_ratio), color = NA)+
    geom_sf(data = country_borders, size = .1, color = "#ffffff")+
    coord_sf(datum = NA)+
    scale_fill_viridis_b(
        option = "B", begin = .15, breaks = c(1, 2, 4),
        guide = guide_coloursteps(barwidth = 15)
    )+
    dark_theme_minimal(base_family = font_rc)+
    labs(
        title = "THE 2022 ranking citation leaders",
        subtitle = "Ratio of citations component of the ranking to the overall score without citations",
        caption = "@ikashnitsky",
        fill = NULL
    )+
    theme(
        plot.title = element_text(family = "Roboto Slab", hjust = .5, face = 2),
        plot.subtitle = element_text(hjust = .5),
        plot.background = element_rect(fill = "#222222", color = NA),
        axis.text = element_blank(),
        legend.position = "top"
    )

ggsave(
    filename = "~/map-avg-ratio.png",
    width = 6, height = 4.2, type = "cairo"
)

the-ranking-2022-raw.csv