How Concentrated Is Air Travel Capacity in a Country’s Largest International Airport?
Ava Zhang
Research Question
How concentrated is air travel capacity in a country’s largest international airport?
I chose the Global Airports dataset from the World Bank because I was interested in how international air travel is distributed within different countries. The dataset includes airport locations with international travel and the total number of seats reported for each airport.
For this analysis, I wanted to see whether a country’s air travel capacity is spread across many international airports or mostly concentrated in its largest airport.
Setup
library(dplyr)
library(ggplot2)
library(readr)Import and Explore the Data
First, I imported the data and looked at the size, column names, and first few rows.
airport_data <- read_csv("airport_volume_airport_locations.csv")
print(dim(airport_data))## [1] 2173 6print(names(airport_data))## [1] "Orig" "Name" "TotalSeats"
## [4] "Country Name" "Airport1Latitude" "Airport1Longitude"print(head(airport_data))## # A tibble: 6 × 6
## Orig Name TotalSeats `Country Name` Airport1Latitude Airport1Longitude
## <chr> <chr> <dbl> <chr> <dbl> <dbl>
## 1 HEA Herat 22042. Afghanistan 34.2 62.2
## 2 JAA Jalalabad 6344. Afghanistan 34.4 70.5
## 3 KBL Kabul Inte… 1016197. Afghanistan 34.6 69.2
## 4 KDH Kandahar I… 39924. Afghanistan 31.5 65.8
## 5 MZR Mazar-e-Sh… 58327. Afghanistan 36.7 67.2
## 6 OAI Bagram Air… 18017. Afghanistan 34.9 69.3The dataset has 2,174 rows and 6 columns. The variables include the airport code, airport name, total seats, country, latitude, and longitude.
Data Preparation
Before doing the analysis, I checked each column for missing values.
missing_values <- colSums(is.na(airport_data))
print(missing_values)## Orig Name TotalSeats Country Name
## 0 0 0 0
## Airport1Latitude Airport1Longitude
## 0 0The result shows that one row contains missing values. I removed the incomplete row before continuing the analysis.
clean_airport_data <- na.omit(airport_data)
print(dim(clean_airport_data))## [1] 2173 6One row contained missing values and was removed, leaving 2,173 rows for analysis.
Insight 1: Number of International Airports by Country
First, I used count() to see how many international
airports are represented for each country.
airport_count <- clean_airport_data %>%
count(`Country Name`, sort = TRUE)
print(head(airport_count, 10))## # A tibble: 10 × 2
## `Country Name` n
## <chr> <int>
## 1 United States 291
## 2 China 102
## 3 Russian Federation 93
## 4 Canada 82
## 5 France 62
## 6 Japan 53
## 7 United Kingdom 52
## 8 Mexico 47
## 9 Italy 46
## 10 Spain 43The result shows the 10 countries with the most international airports represented in the dataset. The United States has the highest number, with 291 airports.
This made me wonder if having many international airports means that air travel capacity is spread across those airports, or if most of the capacity can still be concentrated in one major airport.
Insight 2: How Concentrated Is Air Travel Capacity?
Next, I used group_by() and summarize() to
calculate the number of airports, total reported seats, and seats at the
largest airport for each country.
country_summary <- clean_airport_data %>%
group_by(`Country Name`) %>%
summarize(
airport_count = n(),
total_seats = sum(TotalSeats, na.rm = TRUE),
largest_airport_seats = max(TotalSeats, na.rm = TRUE)
)
print(head(country_summary))## # A tibble: 6 × 4
## `Country Name` airport_count total_seats largest_airport_seats
## <chr> <int> <dbl> <dbl>
## 1 Afghanistan 7 1175318. 1016197.
## 2 Albania 1 2063323. 2063323.
## 3 Algeria 18 5796917. 3904926.
## 4 Angola 4 1235516. 1227093.
## 5 Anguilla 1 28511. 28511.
## 6 Antigua and Barbuda 2 631487. 630335.The result gives one row for each country. It shows the number of airports represented, the total reported seats, and the seats at the largest airport.
Next, I calculated the percentage of each country’s total reported seats that comes from its largest airport.
country_summary <- country_summary %>%
mutate(
largest_airport_share = largest_airport_seats / total_seats * 100
)
print(head(country_summary))## # A tibble: 6 × 5
## `Country Name` airport_count total_seats largest_airport_seats
## <chr> <int> <dbl> <dbl>
## 1 Afghanistan 7 1175318. 1016197.
## 2 Albania 1 2063323. 2063323.
## 3 Algeria 18 5796917. 3904926.
## 4 Angola 4 1235516. 1227093.
## 5 Anguilla 1 28511. 28511.
## 6 Antigua and Barbuda 2 631487. 630335.
## # ℹ 1 more variable: largest_airport_share <dbl>The largest_airport_share variable shows the percentage
of a country’s reported seats that comes from its largest airport. A
higher percentage means that more of the country’s air travel capacity
is concentrated in one airport.
For the comparison, I only included countries with at least 5 international airports represented in the dataset.
concentration_data <- country_summary %>%
filter(airport_count >= 5) %>%
arrange(desc(largest_airport_share))
print(head(concentration_data, 10))## # A tibble: 10 × 5
## `Country Name` airport_count total_seats largest_airport_seats
## <chr> <int> <dbl> <dbl>
## 1 Belarus 6 2753340. 2729413.
## 2 Chile 11 7094794. 7009498.
## 3 Israel 6 14048713. 13820203.
## 4 Estonia 5 1923393. 1888248.
## 5 Peru 7 7352555. 7214675.
## 6 Panama 9 10318333. 10113504.
## 7 Sudan 6 1945812. 1905692.
## 8 Senegal 5 1637042. 1578607.
## 9 Papua New Guinea 11 472358. 455316.
## 10 Azerbaijan 5 2783342. 2681377.
## # ℹ 1 more variable: largest_airport_share <dbl>This helps avoid comparisons based on countries with only one or two airport observations.
Finland was especially interesting to me. It has 22 international airports represented in the dataset, but about 95% of its reported seats come from its largest airport.
This surprised me because Finland has many international airports represented, but its reported air travel capacity is still highly concentrated in one major airport.
Visualization
For the visualization, I selected Finland and several other countries to compare the percentage of reported seats coming from their largest airport.
comparison_countries <- concentration_data %>%
filter(`Country Name` %in% c(
"Finland",
"Belarus",
"Chile",
"Germany",
"United Kingdom",
"Spain"
)) %>%
mutate(
`Country Name` = factor(
`Country Name`,
levels = c(
"Spain",
"Germany",
"United Kingdom",
"Chile",
"Belarus",
"Finland"
)
)
)
print(comparison_countries)## # A tibble: 6 × 5
## `Country Name` airport_count total_seats largest_airport_seats
## <fct> <int> <dbl> <dbl>
## 1 Belarus 6 2753340. 2729413.
## 2 Chile 11 7094794. 7009498.
## 3 Finland 22 12422282. 11762383.
## 4 United Kingdom 52 154411889. 50054383.
## 5 Germany 38 128494422. 40889303.
## 6 Spain 43 106829344. 26811681.
## # ℹ 1 more variable: largest_airport_share <dbl>The selected countries show very different patterns. Finland stands out because even though 22 international airports are represented in the dataset, about 95% of its reported seats come from its largest airport.
ggplot(comparison_countries,
aes(x = `Country Name`,
y = largest_airport_share)) +
geom_col() +
coord_flip() +
labs(
title = "Air Travel Capacity in Selected Countries",
subtitle = "Finland: 95% of seats concentrated in one airport",
x = "Country",
y = "Share of Seats at Largest Airport (%)"
)
The chart shows that having many international airports does not necessarily mean that air travel capacity is evenly distributed among them.
Conclusion
From this analysis, I found that the number of international airports in a country does not tell the whole story about how its air travel capacity is distributed.
Finland was the most interesting example to me. Out of 22 international airports represented in the dataset, about 95% of Finland’s reported seats come from just its largest airport.
This shows that even when a country has many international airports, its air travel capacity can still be highly concentrated in one major airport.