Here are some follow-up exercises for subsetting vectors and data.frames that you may wish to try.

In class, we have used the gapminder dataset. You might remember that this was a data package that we had to manually install and then load into our workspace using library(). However, there are many other datasets that already come with R. You can checkout the list if you would like using data(). For these exercises, we’re going to use the iris dataset.

iris

1. What are the column names and data types of the different columns in iris?

2. How many rows and columns does iris have?

3. Create a single vector called ‘width’ that is the Sepal.Width th column of iris.

4. What is the 100th value in your ‘width’ vector?

5. What is the last value in your ‘width’ vector? Can you write code that returns this value even if you don’t know how long ‘width’ is?

6. Select rows 10 to 20, with all columns in the iris dataset.

7. Select rows 10 to 20 with only the Species, Petal.Width and Petal.Length. Can you do this two different ways?

8. Select rows 1 to 10, 20, and 100 in the iris dataset.

9. Select the first value in the Sepal.Length column of the iris dataset. Bonus - can you do this three different ways?

10. Bonus: In class the other day I pointed out an odd behaviour, that when we extracted a single column of our data.frame, it lost its dimensions - dim(new_object). We had expected that the dimensions would be however many rows there were and then one column. But the output returned NULL. We also noticed that this new object became a vector, not a data.frame like we started with. As a reminder below is a summary of what we saw in class.

The bonus question is: Can you select a single column of the data.frame and retain the dimensions/type of object?

Hint: The unnecessary dimension is ‘dropped’.

library(gapminder)

str(gapminder)

new_object <- gapminder[, 'country']

dim(gapminder)
is.data.frame(gapminder)

dim(new_object)
is.data.frame(new_object)
is.vector(new_object)