Comparison of the distributions of QS university scores among a number of countries.
- Make sure you have the following libraries:
library(tidyverse)
library(jsonlite)
library(janitor)
library(countrycode)
library(here)
library(conflicted)Optionally, sort out the most likely conflicts:
conflict_prefer('flatten', 'purrr')
conflict_prefer('filter', 'dplyr')
conflict_prefer('lag', 'dplyr')df <- read_csv(here("data", "qs.csv"))
country_summary <- read_csv(here("data", "country_summary.csv"))country_summary %>%
filter(count_score >= 5) %>%
arrange(desc(median_score)) %>%
select(country, `n(scores)` = count_score, `med(score)` = median_score) %>%
head(10) %>%
knitr::kable()| country | n(scores) | med(score) |
|---|---|---|
| HK | 6 | 75.05 |
| CH | 9 | 54.90 |
| NL | 13 | 53.40 |
| DK | 5 | 51.00 |
| SE | 8 | 48.85 |
| JP | 16 | 48.15 |
| MY | 8 | 48.05 |
| US | 86 | 45.95 |
| GB | 49 | 45.90 |
| BE | 8 | 40.55 |
df %>%
filter(country %in% c("CH", "FR", "NL", "US", "GB", "DE")) %>%
ggplot(aes(score, country)) +
labs(
title = "QS University Score Distribution Per Country",
subtitle = "For CH, DE, FR, GB, NL, and the US") +
geom_boxplot(varwidth = TRUE)unirankings is released under the MIT License.