Lab 04: Tidy Summaries with Gapminder Data

BMI 5/625

1 Overview

Both challenges are due by the end of the day via Sakai on Wednesday April 29th. For the first challenge, focused on data-tidying, you’ll want to refer back to our slides. For the second challenge, you’ll want to refer to the reference lab.

2 Challenge 1: Data-Tidying

Tidy the data/gapminder_broadband_per_100.xlsx file (Tip: use the readxl package’s read_excel() function to import from Excel, and use janitor::clean_names() immediately after import to make life easier)

• The data are the fixed broadband internet subscribers (per 100 people) for different countries by year: “Fixed broadband subscribers are users of the Internet who subscribe to paid high-speed access to the public Internet. High-speed access is at least 256 kilobits per second in one or both directions. Source: International Telecommunication Union, World Telecommunication Development Report and database, and World Bank estimates. Note: Please cite the International Telecommunication Union for third-party use of these data.”
• Read more about the numbers here
• You’ll know you’re done when the dataset is “tidy” according to the definition/example from the slides:
  • Each row corresponding to a single data point
  • In the case of this dataset, a datapoint is a (country, year, count) triple

3 Challenge 2: Gapminder Exercise

Install and load the gapminder data package (already installed on Posit Cloud).

install.packages("gapminder")
library(gapminder)
?gapminder

Pick at least two of the tasks below from the task menu and approach each with both a table (containing the appropriately-wrangled data) and a companion figure.

• dplyr should be your main data manipulation tool
• ggplot2 should be your main visualization tool

Make observations about what your tables/figures show and about the process. If you want to do something comparable but different, i.e. swap one quantitative variable for another- go for it!

You do not have to use tidyr or otherwise worry about reshaping your tables. Many of your tables may not be formatted perfectly in the report. Simply printing dplyr tabular output is fine. For all things, graphical and tabular, if you’re dissatisfied with a result, discuss the problem, what you tried to do to fix it, and move on.

Note: The dataset is chronological, in that it contains data over multiple years for the same country and quantity. Make sure that your analysis takes this into account! In other words, make sure not to blindly treat e.g. datapoints about a country’s GDP from 1950 as being comparable another country’s GDP in 2000.

3.1 Task menu

• Get the maximum and minimum of GDP per capita for all continents.

• Look at the spread of GDP per capita across countries within the continents.

• How does life expectancy vary across different continents?

• Report the absolute and/or relative abundance of countries with low life expectancy over time by continent: Compute some measure of worldwide life expectancy - you decide - a mean or median or some other quantile or perhaps your current age. Then determine how many countries on each continent have a life expectancy less than this benchmark, for each year.

• Make up your own!

3.2 Companion graphs

For each table, make sure to include a relevant figure. One tip for starting is to draw out on paper what you want your x- and y-axis to be first and what your geom is; that is, start by drawing the plot you want ggplot to give you. Your figure does not have to depict every single number present in the table. Use your judgement. It just needs to complement the table, add context, and allow for some sanity checking.

Notice which figures are easy/hard to make, and whether the visualization adds clarity, detracts from, or is completely redundant (and therefore probably unnecessary) with respect to the tabular display.

4 Report your process

You’re encouraged to reflect on what was hard/easy, problems you solved, helpful tutorials you read, etc.

---

Gapminder EDA ideas from Jenny Bryan, author and creator of the Gapminder package.