What defines internet usage for a country?
Do different regions of the world have different internet usage?
What is the impact of democracy, education, economy, and health on internet use?
Fall 2018
Objectives
Introduction
Datasets Used:
- Population (CIA World Factbook)
- Internet Users (CIA World
Factbook)
- Democracy Index & Category (Wikipedia)
- Education Expenditures (CIA World
Factbook)
- GDP per Capita (CIA World Factbook)
- Life Expectancy (CIA World Factbook)
-
Health Expenditures (CIA World Factbook)
- World Regions (Wikimedia)
Data Cleaning
## Classes 'tbl_df', 'tbl' and 'data.frame': 224 obs. of 12 variables: ## $ country : chr "China" "India" ... ## $ internet_users: num 7.31e+08 3.74e+08 ... ## $ population : num 1.38e+09 1.28e+09 ... ## $ world_region : chr "Asia & Pacific" "Asia & Pacific" ... ## $ global_ns : chr "Global South" "Global South" ... ## $ dem_score : num 3.1 7.23 7.98 6.86 7.88 ... ## $ dem_cat : chr "Authoritarian" "Flawed democracy" ... ## $ gdp.edu : num NA 3.2 5.4 5.8 3.8 ... ## $ gdppp : num 16700 7200 59500 15600 42800 ... ## $ gdp.health : num 5.5 4.7 17.1 8.3 10.2 ... ## $ years : num 75.7 68.8 80 74 85.3 ... ## $ in_percent : num 53 29.2 ...
Defining Internet Usage
We chose to define internet usage for a country as the number of its internet users as a percentage of the country’s total population.
This is similar to the United Nation’s measurement of internet usage (internet users per 100 population).
Internet Usage Map
Internet Usage (Histogram)
Internet Usage by Region (Summary)
| World Region | Q1 | Median | Q3 | IQR |
|---|---|---|---|---|
| Africa | 9.212222 | 17.62190 | 25.17869 | 15.96646 |
| Arab States | 20.663345 | 49.67855 | 76.20705 | 55.54370 |
| Asia & Pacific | 25.171843 | 42.54363 | 68.11942 | 42.94758 |
| Europe | 69.355164 | 76.76947 | 87.98873 | 18.63356 |
| North America | 79.888475 | 85.25627 | 91.27361 | 11.38514 |
| South/Latin America | 44.234467 | 55.60399 | 68.17442 | 23.93996 |
Internet Usage by Region
Internet Usage by Global N/S
High & Low Internet Usage
Adding World Region to a Linear Model
## [1] "R^2 of Internet Usage ~ Life Expectancy:"
## [1] 0.6165133
## [1] "R^2 of Internet Usage ~ Life Expectancy + World Region:"
## [1] 0.6827802
Potential Internet Usage Factors
- Democracy: The country’s democracy index and category; data is compiled by the Economist Intelligence Unit (EIU).
- Education: The percentage of a country’s GDP spent on education.
- Economy: The country’s GDP per Capita.
- Health: The percentage of a country’s GDP spent on education & and the average life expectancy of the given country.
Impact of Economy on Internet Use
Impact of Democracy on Internet Use
Democracy Types and Internet Use
Impact of Health on Internet Use
Impact of Health (cont)
Linear Model of Internet Usage ~ Health Expenditures
## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 29.033772 5.0325347 5.769215 3.251257e-08 ## gdp.health 2.877285 0.6870142 4.188101 4.329408e-05
\(H_0: \beta_1 = 0\)
\(H_A: \beta_1 \neq 0\)
The p-value for this test is 4.329408e-05, which is
very close to 0.
We reject \(H_0\).
## [1] "R^2 of Internet Usage ~ % of GDP Spent on Health:"
## [1] 0.08575427
Impact of Health (cont)
Impact of Education on Internet Use
Impact of Education (cont)
Conclusion
- There is a difference in internet usage for different regions
- GDPPP, life expectancy, and democracy index/category are good indicators for internet usage
- Education and health expenditures (as a percentage of GDP) are poor indicators for internet usage