25 Our World in Data

25.1 The package `owidR`

We study the R package owidR for importing data from Our World in Data.

Warnings: The package does not work properly though there was no problem in AY2022.

25.2 References

The package official site contains other links. When you quote the package, use the link to the official site.

Package Official Site: https://CRAN.R-project.org/package=owidR
- ReadMe: https://cran.r-project.org/web/packages/owidR/readme/README.html
- Manual: https://cran.r-project.org/web/packages/owidR/owidR.pdf
- Vignette: Create and Analyse a Dataset
owidR: Import Data from Our World in Data
- Man page and source codes

In general, README gives a short introduction to the package, a Manual, the comprehensive descriptions of each function, and a Vignette, a practical introduction containing examples and applications.

25.3 Introduction

This package acts as an interface to Our World in Data datasets, allowing for an easy way to search through data used in over 3,000 charts and load them into the R environment.

25.3.1 Setup

25.3.1.1 Installation

Run the following for the first time

install.packages("owidR")

25.3.1.2 Load the package

library(owidR)
library(tidyverse) 
#> ── Attaching core tidyverse packages ──── tidyverse 2.0.0 ──
#> ✔ dplyr     1.1.3     ✔ readr     2.1.4
#> ✔ forcats   1.0.0     ✔ stringr   1.5.0
#> ✔ ggplot2   3.4.4     ✔ tibble    3.2.1
#> ✔ lubridate 1.9.3     ✔ tidyr     1.3.0
#> ✔ purrr     1.0.2     
#> ── Conflicts ────────────────────── tidyverse_conflicts() ──
#> ✖ dplyr::filter() masks stats::filter()
#> ✖ dplyr::lag()    masks stats::lag()
#> ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

The package automatically load a part of tidyverse, e.g., dplyr, ggplot, …. Since it works well with the schemetidyverse, it is better to load tidyverse with it.

The creator of this package also suggests loading packages plm for panels of data, and texreg for displaying models, but let us start without them until we actually use them. For panel data, see, for example, the site](https://www.aptech.com/blog/introduction-to-the-fundamentals-of-panel-data/).

25.3.2 Core functions

25.3.2.1 List of core functions

In this package, chart is close to data, and chart id is a data indicator.

owid: Get a dataset used in an OWID chart
owid_covid: Get the Our World in Data covid-19 dataset
owid_search: Search the data sources used in OWID charts
owid_source: A function to get source information from an OWID dataset and display it in the R console.
pal_owid: Colour palettes based on the colours used by Our World in Data
view_chart: A function that opens the original OWID chart in your browser
world_map_data: Function that returns a simple feature collection of class sf. Map data is from naturalearthdata.com. Designed to be used internally.

25.3.2.2 Usage

25.3.2.2.1 `owid_search`

Search the data sources used in OWID charts

owid_search(term)

Example

Since the output is long, I cut it off to the first six rows using head().

owid_search("emissions") %>% head()
#>      titles                                               
#> [1,] "Methane emissions from agriculture"                 
#> [2,] "Nitrous oxide emissions from agriculture"           
#> [3,] "Per capita nitrous oxide emissions from agriculture"
#> [4,] "Air pollutant emissions"                            
#> [5,] "Emissions of air pollutants"                        
#> [6,] "Emissions of air pollutants"                        
#>      chart_id                              
#> [1,] "methane-emissions-agriculture"       
#> [2,] "nitrous-oxide-agriculture"           
#> [3,] "per-capita-nitrous-oxide-agriculture"
#> [4,] "air-pollutant-emissions"             
#> [5,] "emissions-of-air-pollutants"         
#> [6,] "emissions-of-air-pollutants-oecd"

A matrix is returned. If the list is long, it is easier to see the pairs of the titles and chart_ids by adding as_tibble().

owid_search("emissions") %>% as_tibble()
#> # A tibble: 187 × 2
#>    titles                                           chart_id
#>    <chr>                                            <chr>   
#>  1 Methane emissions from agriculture               methane…
#>  2 Nitrous oxide emissions from agriculture         nitrous…
#>  3 Per capita nitrous oxide emissions from agricul… per-cap…
#>  4 Air pollutant emissions                          air-pol…
#>  5 Emissions of air pollutants                      emissio…
#>  6 Emissions of air pollutants                      emissio…
#>  7 Emissions of particulate matter                  emissio…
#>  8 Global sulphur dioxide (SO₂) emissions by world… so-emis…
#>  9 Sulphur dioxide (SO₂) emissions                  so2-emi…
#> 10 Sulphur oxide (SO₂) emissions                    sulphur…
#> # ℹ 177 more rows

If the list is not long, you do not need to add as_tibble(). However, note that you need to keep in mind that the title and the chart_id consists of a pair, and you need to use the chart_id to download the data using owid.

owid_search("human rights")
#>      titles                                                                             
#> [1,] "Human rights index vs. electoral democracy index"                                 
#> [2,] "Confirmed killings of human rights defenders, journalists and trade unionists"    
#> [3,] "Countries with accredited independent national human rights institutions"         
#> [4,] "Distribution of human rights index"                                               
#> [5,] "Human rights index"                                                               
#> [6,] "Human rights index"                                                               
#> [7,] "Human rights index vs. GDP per capita"                                            
#> [8,] "Share of countries with accredited independent national human rights institutions"
#>      chart_id                                                                   
#> [1,] "human-rights-index-vs-electoral-democracy-index"                          
#> [2,] "cases-of-killed-human-rights-defenders-journalists-trade-unionists"       
#> [3,] "countries-with-independent-national-human-rights-institution"             
#> [4,] "distribution-human-rights-index-vdem"                                     
#> [5,] "human-rights-index-vdem"                                                  
#> [6,] "human-rights-index-population-weighted"                                   
#> [7,] "human-rights-index-vs-gdp-per-capita"                                     
#> [8,] "share-countries-accredited-independent-national-human-rights-institutions"

25.3.2.2.2 `owid`

Get a dataset used in an OWID chart

owid(chart_id = NULL, rename = NULL, tidy.date = TRUE, ...)

chard_id: The chart_id as returned by owid_search, which is combined with ‘-’. Don’t mix up with the chart titles.

rename: Rename the value column. Currently only works if their is just one value col- umn.

Example

emissions <- owid("per-capita-ghg-emissions")

emissions

rights <- owid("human-rights-scores")

rights

Note.

You can use rename to change column names. For example,

owid("per-capita-ghg-emissions", rename = "ghgPcap")

Since until after importing the data, you never know the original column name, and how many columns are for indicators. It is natural to change column names using dpyr::rename. In the next example, I used Total including LUCF. However, ‘Total including LUCF’ and “Total including LUCF” work as well.

emissions %>% rename(ghgPcap = `Per-capita greenhouse gas emissions in CO2 equivalents`)

If there are more than one variables to rename, use vector notation as follows. Here top_n(1) is same as slice(1), and gives the first row only.

owid("electoral-democracy") %>% top_n(1)

owid("electoral-democracy", rename = c("electoral_democracy", "vdem_high", "vdem_low"))

You can use dplyr::rename, and keep the record of renaming column names.

(democracy <- owid("electoral-democracy"))

democracy <- democracy %>% 
  rename(`electoral_democracy` = `Electoral democracy`,  
         `vdem_high` = `electdem_vdem_high_owid`, 
         `electdem_vdem_low_owid` = `electdem_vdem_low_owid`)
democracy

25.3.2.2.3 `owid_source`

A function to get source information from an OWID dataset and display it in the R console.

owid_source(data)

Example

owid_source(emissions)

owid_source(rights)

25.3.2.2.4 `view_chart`

A function that opens the original OWID chart in your browser

view_chart(x)

x Either a tibble returned by owid(), or a chart_id.
Example

The first one uses the chart, i.e., the tibble returned by owid(), and the second, chart_id. You can also embed in your R Markdown file by copying Embed iframe clink from Share botton at the bottom right corner.

firearm_suicide <- owid("suicide-rate-by-firearm")
    view_chart(firearm_suicide)

view_chart("electoral-democracy")

view_chart("share-of-individuals-using-the-internet")

25.3.2.2.5 `pal_owid`

Colour palettes based on the colours used by Our World in Data

25.3.2.2.6 `owid_covid`

owid_covid: Get the Our World in Data covid-19 dataset

owid_covid()

See the detail at the GitHub site.

Example

covid <- owid_covid()

covid %>% filter(location == "Japan")
#> # A tibble: 1,428 × 67
#>    iso_code continent location date       total_cases
#>    <chr>    <chr>     <chr>    <date>           <dbl>
#>  1 JPN      Asia      Japan    2020-01-03          NA
#>  2 JPN      Asia      Japan    2020-01-04          NA
#>  3 JPN      Asia      Japan    2020-01-05          NA
#>  4 JPN      Asia      Japan    2020-01-06          NA
#>  5 JPN      Asia      Japan    2020-01-07          NA
#>  6 JPN      Asia      Japan    2020-01-08          NA
#>  7 JPN      Asia      Japan    2020-01-09          NA
#>  8 JPN      Asia      Japan    2020-01-10          NA
#>  9 JPN      Asia      Japan    2020-01-11          NA
#> 10 JPN      Asia      Japan    2020-01-12          NA
#> # ℹ 1,418 more rows
#> # ℹ 62 more variables: new_cases <dbl>,
#> #   new_cases_smoothed <dbl>, total_deaths <dbl>,
#> #   new_deaths <dbl>, new_deaths_smoothed <dbl>,
#> #   total_cases_per_million <dbl>,
#> #   new_cases_per_million <dbl>,
#> #   new_cases_smoothed_per_million <dbl>, …

25.4 Examples

The following is based on the presentation and the first two R Notebook files created by Professor Kaizoji.

25.4.1 Human Rights- modified from README

Lets use the core functions to get data on how human rights have changed over time. First by searching for charts on human rights.

owid_search("human rights") %>% as_tibble()
#> # A tibble: 8 × 2
#>   titles                                            chart_id
#>   <chr>                                             <chr>   
#> 1 Human rights index vs. electoral democracy index  human-r…
#> 2 Confirmed killings of human rights defenders, jo… cases-o…
#> 3 Countries with accredited independent national h… countri…
#> 4 Distribution of human rights index                distrib…
#> 5 Human rights index                                human-r…
#> 6 Human rights index                                human-r…
#> 7 Human rights index vs. GDP per capita             human-r…
#> 8 Share of countries with accredited independent n… share-c…

Let’s use the human rights protection dataset.

rights <- owid("human-rights-protection")
rights

ggplot2 makes it easy to visualise our data.

rights %>% 
  filter(entity %in% c("United Kingdom", "France", "United States", "Japan")) %>%  
  ggplot(aes(year, `Human rights protection`, colour = entity)) +
  geom_line()

25.4.2 Internet - modified from vignette

owid_search("internet") %>% as_tibble()
#> # A tibble: 8 × 2
#>   titles                                            chart_id
#>   <chr>                                             <chr>   
#> 1 Landline Internet subscriptions                   landlin…
#> 2 Landline Internet subscriptions by download speed landlin…
#> 3 Landline Internet subscriptions per 100 people    broadba…
#> 4 Landline Internet subscriptions per 100 people, … landlin…
#> 5 Number of people using the Internet               number-…
#> 6 Share of US adults who use the Internet, by age   share-u…
#> 7 Share of primary schools with access to the Inte… primary…
#> 8 Share of the population using the Internet        share-o…

Get a dataset used in an OWID chart.

owid("share-of-individuals-using-the-internet") %>% top_n(1)

internet <- owid("share-of-individuals-using-the-internet", rename = "internet_use")
internet

Get source information on an OWID dataset

owid_source(internet)

A function that opens the original OWID chart in your browser.

view_chart(internet)

Plot an owid dataset. The first is the simplest, and the second uses oied theme.

internet %>% filter(entity == "World") %>%
  ggplot(aes(year, internet_use))+ geom_line()

internet %>% filter(entity == "World") %>%
  ggplot(aes(year, internet_use))+ geom_line() +
  labs(title = "Share of the World Population \nusing the Internet",
       y = "Individuals using the Internet \n(% of population)") +
  scale_y_continuous(limits = c(0, 100))

internet %>% 
  filter(entity %in% c("United Kingdom", "Spain", "Russia", "Egypt", "Nigeria")) %>% 
  ggplot(aes(year, internet_use, color =  entity)) + geom_line() +
  labs(title = "Share of Population with Using the Internet",
       y = "Individuals using the Internet \n(% of population)",
       color = "country") +
  scale_y_continuous(limits = c(0, 100), labels = scales::label_number(suffix = "%"))

Creating a choropleth map.

owid_map(internet, year = 2017) +
  labs(title = "Share of Population Using the Internet, 2017")

25.4.3 Democracy - modified from vignette

owid_search("democrac") %>% as_tibble()
#> # A tibble: 93 × 2
#>    titles                                           chart_id
#>    <chr>                                            <chr>   
#>  1 Child mortality rate vs. electoral democracy     child-m…
#>  2 People living in democracies and non-democracies world-p…
#>  3 Age of democracy                                 age-of-…
#>  4 Age of democracy                                 age-of-…
#>  5 Age of democracy                                 age-of-…
#>  6 Age of electoral democracy                       age-of-…
#>  7 Age of electoral democracy                       age-of-…
#>  8 Age of liberal democracy                         age-of-…
#>  9 Citizen satisfaction with democracy              citizen…
#> 10 Citizen support for democracy                    citizen…
#> # ℹ 83 more rows

owid("electoral-democracy") %>% top_n(1)

democracy <- owid("electoral-democracy", rename = c("electoral_democracy", "vdem_high", "vdem_low"))
democracy

owid_source(democracy)

owid_map(democracy, year = 2015, palette = "YlGn") +
  labs(title = "Electoral Democracy")

democracy %>% 
  filter(entity %in% c("United Kingdom", "Spain", "Russia", "Egypt", "Nigeria")) %>%
  ggplot(aes(year, electoral_democracy, color = entity)) + geom_line() +
  labs(title = "Electoral Democracy", y = "", color = "country") +
  scale_y_continuous(limits = c(0, 1), labels = scales::label_number(suffix = "%"))

gdp <- owid("gdp-per-capita-worldbank", rename = "gdp")

gov_exp <- owid("total-gov-expenditure-gdp-wdi", rename = "gov_exp")

age_dep <- owid("age-dependency-ratio-of-working-age-population", rename = "age_dep")

unemployment <- owid("unemployment-rate", rename = "unemp")

Mutating joins

left_join(): includes all rows in x.

References

See https://dplyr.tidyverse.org/reference/mutate-joins.html.
Posit Primers - Tidy your data: https://posit.cloud/learn/primers/4
- Join Data Sets: https://posit.cloud/learn/primers/4.3

data <- internet %>% 
  left_join(democracy) %>% 
  left_join(gdp) %>% 
  left_join(gov_exp) %>% 
  left_join(age_dep) %>% 
  left_join(unemployment)

Drawing scatter plot

data %>% 
  filter(year == 2015) %>% 
  ggplot(aes(internet_use, electoral_democracy)) +
  geom_point(colour = "#57677D", na.rm = TRUE) +
  geom_smooth(method = "lm", colour = "#DC5E78", na.rm = TRUE) +
  labs(title = "Relationship Between Internet Use \nand electoral_democracy", x = "Internet Use", y = "electoral_democracy")

data %>% 
  filter(year == 2015) %>% 
  ggplot(aes(gdp, internet_use)) +
  geom_point(colour = "blue") +
  geom_smooth(method = "gam", colour = "red", level = 0.0) +
  labs(title = "Relationship Between Internet Use and GDP", x = "GDP", y = "Internet Use")

model1 <- lm(electoral_democracy ~ internet_use, data)
summary(model1)

model2 <- lm(gdp ~ internet_use, data)
summary(model2)

Creating a table of the results of the regression analysis using texreg. For the first time, install the pacage texreg.

install.packages("texreg")

library(texreg)
#> Version:  1.38.6
#> Date:     2022-04-06
#> Author:   Philip Leifeld (University of Essex)
#> 
#> Consider submitting praise using the praise or praise_interactive functions.
#> Please cite the JSS article in your publications -- see citation("texreg").
#> 
#> Attaching package: 'texreg'
#> The following object is masked from 'package:tidyr':
#> 
#>     extract

models <- list("Model 1" = model1,
               "Model 2" = model2)
screenreg(models, stars = NULL)

24 World Bank

26 OECD