The
In the Data and Computing Fundamentals course, we use these data for several purposes:
Eventually, we plan to write interface software to get data directly from the Gapminder site, but for the present we have downloaded a subset of data and provided access through the fetchData() and fetchGapminder() functions.
The list of available files is here, along with instructions for adding new files.
The Gapminder data are originally stored with countries being the cases and years the “variables.” For example, here is the data on cell phone subscriptions (only part of the data is shown):
phones <- fetchData("Gapminder/CellPhoneTotal.csv")
## Retrieving from
## http://www.mosaic-web.org/go/datasets/Gapminder/CellPhoneTotal.csv
## Mobile.cellular.subscriptions..total.number X2009 X2010 X2011
## 1 Abkhazia NA NA NA
## 2 Afghanistan 10500000 13000000 17558265
## 3 Akrotiri and Dhekelia NA NA NA
## 4 Albania 2463741 2692372 3100000
## 5 Algeria 32729824 32780165 35615926
## 6 American Samoa NA NA NA
This is a “wide” format. In this presentation, there is a separate variable for each year — 2009, 2010, and 2011 are shown. Under that year, the value gives the number of cell phone subscriptions.
Although the wide format is reasonable, it has disadvantages. In particular, it can be hard to do comparisons across countries, except for the same year.
The “narrow” format of these same data has three rows: country, cell phone number, and year. Converting between wide and narrow formats is not difficult, but to simplify things, we have presented a function that does the conversion automatically:
phones <- fetchGapminder("Gapminder/CellPhoneTotal.csv")
## Retrieving from
## http://www.mosaic-web.org/go/datasets/Gapminder/CellPhoneTotal.csv
## Country Year CellPhoneTotal
## 12102 Afghanistan 2009 10500000
## 12104 Albania 2009 2463741
## 12105 Algeria 2009 32729824
## 12111 Argentina 2009 52482780
## 12113 Aruba 2009 128000
## 12377 Afghanistan 2010 13000000
## 12379 Albania 2010 2692372
## 12380 Algeria 2010 32780165
## 12386 Argentina 2010 53700000
## 12388 Aruba 2010 131800
## 12652 Afghanistan 2011 17558265
## 12654 Albania 2011 3100000
## 12655 Algeria 2011 35615926
## 12661 Argentina 2011 55000000
To a human reader, this format is not necessarily easier than the wide format. But with grouping and joining operations, it's considerably easier.
For instance, suppose you're interested in the fraction of the population with cell phone subscriptions. You'll want to read in both the phone and the population data, sorting them out year-by-year for each country. Then divide subscriptions by population to get an index, and perhaps display this as a map. But not every country has a reading for every year, so perhaps best to look at the maximum number reported for the period since 2005.
phones <- fetchGapminder("Gapminder/CellPhoneTotal.csv")
phonemax <- groupBy(phones, by = Country, phones = max(CellPhoneTotal))
pop <- fetchGapminder("Gapminder/TotalPopulation.csv")
# Bad formatting by Gapminder
pop <- transform(pop, pop = as.numeric(gsub(",", "", TotalPopulation)))
popmax <- groupBy(pop, by = Country, pop = max(pop, na.rm = TRUE))
both <- join(popmax, phonemax)
both <- subset(both, phones > 0)
both <- transform(both, phoneDensity = phones/pop)
Mapping out the number of subscriptions per population shows some expected and some perhaps unexpected patterns, such as the large density in countries such as Chile, Argentina, Libya, South Africa, Zimbabwe, etc.
worldMap(phoneDensity ~ Country, data = both)
## 190 codes from your data successfully matched countries in the map
## 5 codes from your data failed to match with a country code in the map
## 53 codes from the map weren't represented in your data