Exercise 3: dplyr

After working through Exercise 3, you’ll…

  • have assessed how well you know dplyr
  • know what dplyr functions and concepts you might want to repeat again
  • have managed to apply the dplyr concepts to data

Task 1

Below you will see multiple choice questions. Please try to identify the correct answers. 1, 2, 3 and 4 correct answers are possible for each question.

1. What are the main characteristics of tidy data?

  • Every cell contains values.
  • Every cell contains a variable.
  • Every observation is a column.
  • Every observation is a row.

2. What are dplyr functions?

  • summary()
  • describe()
  • mutate()
  • manage()

3. How can you sort the eye_color of Star Wars characters from Z to A?

  • starwars_data %>% arrange(desc(eye_color))
  • starwars_data %>% arrange(eye_color)
  • starwars_data %>% select(arrange(eye_color))
  • starwars_data %>% select(eye_color) %>% arrange(desc(eye_color))

4. Imagine you want to recode the height of these characters. You want to have three categories from small and medium to tall. What is a valid approach?

  • starwars_data %>% mutate(height = case_when(height<=150~"small",height<=190~"medium",height>190~"tall"))
  • starwars_data %>% mutate(height = case_when(height<=150~small,height<=190~medium,height>190~tall))
  • starwars_data %>% recode(height = case_when(height<=150~"small",height<=190~"medium",height>190~"tall"))
  • starwars_data %>% recode(height = case_when(height<=150~small,height<=190~medium,height>190~tall))

5. Imagine you want to provide a systematic overview over all hair colors and what species wear these hair colors frequently (not accounting for the skewed sampling of species)? What is a valid approach?

  • starwars_data %>% group_by(hair_color) %>% group_by(species) %>% summarize(count = n()) %>% arrange(hair_color)
  • starwars_data %>% group_by(hair_color, species) %>% summarize(count = n()) %>% arrange(hair_color)
  • starwars_data %>% group_by(hair_color & species) %>% summarize(count = n()) %>% arrange(hair_color)
  • starwars_data %>% group_by(hair_color + species) %>% summarize(count = n()) %>% arrange(hair_color)

Task 2

It’s your turn now. Load the starwars data like this:

library(dplyr) # to activate the dplyr package
starwars_data <- starwars # to assign the pre-installed starwars data set (dplyr) into a source object in our environment

How many humans are contained in the starwars dataset overall? First, solve this task using filter() only. Next, solve it by combining filter() with summarize(count = n()). Finally, try to solve it with count().

Task 3

Use mutate() and case_when() to create a new column height_group with: - “short” if height < 140 - “medium” if height < 180 - “tall” otherwise.

Task 4

How many humans are contained in starwars by height_group? (Hint: You’ll need to chain multiple functions. Remember that summarize() has a best friend that will help you solve this task. :))

Task 5

What is the most common height_group among Star Wars characters? (Hint: You’ll need to chain multiple functions, but one of them should be arrange())

Task 6

What is the average mass of Star Wars characters that are not human and have yellow eyes? (Hint: You’ll need to calculate descriptive statistics and remove all NAs.)

Task 7

Compare the mean, median, and standard deviation of mass for all humans and droids. (Hint: You’ll need to calculate descriptive statistics and remove all NAs.)

Task 8

Create a new variable in which you store the mass in gram (gr_mass). Add it to the dataframe. Test whether your solution works by printing your data to the console, but only show the name, species, mass, and your new variable gr_mass.

When you’re ready to look at the solutions, you can find them here: Solutions for Exercise 3.