Exercises for Week 3

Data and Computing Fundamentals

Narrow Data Exercises

In the Gapminder\TotalPopulation.csv data …

• How many countries altogether? 260 or so
• How many years altogether? 310
• How many countries for each year? Starts at 64 and goes up to almost all 260
• How many years for each country? Not sure how I would find this. It seems that there is not data from every year, especially earlier on.
• Are there any countries that are listed before 1950 but not since then? No, but the countries that don't exist anymore (Czechoslovakia, etc.) simply do not have data past certain points.

In the Gapminder\LandArea.csv data …

• Are there countries that have changed in size over the years? Hint: Use groupBy() with an operation of min(LandArea) != max(LandArea) Yes there are.
• What countries have changed in area by more than 10%? Ifelse does not seem to work…

Narrow to Wide

Use subset() and join() to create a new data frame that has the country's population in 1990 as one variable and the population in 2010 as another variable.

pop90 = subset(pop, Year == 1990)

## Error: object 'pop' not found

pop10 = subset(pop, Year == 2010)

## Error: object 'pop' not found

allpop = join(pop90, pop10)

## Error: object 'pop10' not found

Quantitative to Categorical

• Use transform() and cut() to create a prewar variable indicating whether data are from before World War II. Use levels: “pre” and “post”
• Use transform() and cut() to create a decade variable coded as “1980s”, “1990s”, “2000s”
• Use subset(), transform() and ntiles() to divide up countries into 5 categories according to their GDP in 2010.
  • Use groupBy() to find the range of GDP in each of the categories. Hint: max() and min()
• Use groupBy(), transform() and ntiles() to divide up countries into 5 categories according to how much their GDP has grown from 1990 to 2010.
• Use groupBy(), transform() and ntiles() to divide up countries into 5 categories according to how much their population has grown from 1990 to 2010.
• Make a bar plot that shows the relationship between the GDP growth categories and population growth categories.
• Use cut() to divide countries up into those that have shrank in population from 1990 to 2010, those that grew by less than 1%, less than 5%, less than 10%, and so on.

Map Exercises

In each R chunk, include the commands needed to pre-process the data, e.g., groupBy, join, and transform. Then, use mWorldMap() to generate a map command interactively. Then add the map command to your R chunk. Don't use mWorldMap() in this document.

Make a map of per-country CO_2 emissions.

Make a map of per-country CO_2 emissions per capita.

Make a map of per-country CO_2 emissions per unit of GDP.

Make a map showing which countries have more than doubled in population since 1960.

Make a bubble map showing population as the size of the bubble and fertility as the color.

Pick a Gapminder variable. Make a map that displays which countries have data over a wide range of years and which have data only recently. Hint: min(Year)