Do cars with big engines use more fuel than cars with small engines?

Included in ggplot2 is a data frame called mpg containing variables in columns and observations in rows collected by the US Environmental Protection Agency for 38 models of cars.

Type mpg in the console window to take a look at this R object:

mpg

## # A tibble: 234 × 11
##    manufacturer      model displ  year   cyl      trans   drv   cty   hwy
##           <chr>      <chr> <dbl> <int> <int>      <chr> <chr> <int> <int>
## 1          audi         a4   1.8  1999     4   auto(l5)     f    18    29
## 2          audi         a4   1.8  1999     4 manual(m5)     f    21    29
## 3          audi         a4   2.0  2008     4 manual(m6)     f    20    31
## 4          audi         a4   2.0  2008     4   auto(av)     f    21    30
## 5          audi         a4   2.8  1999     6   auto(l5)     f    16    26
## 6          audi         a4   2.8  1999     6 manual(m5)     f    18    26
## 7          audi         a4   3.1  2008     6   auto(av)     f    18    27
## 8          audi a4 quattro   1.8  1999     4 manual(m5)     4    18    26
## 9          audi a4 quattro   1.8  1999     4   auto(l5)     4    16    25
## 10         audi a4 quattro   2.0  2008     4 manual(m6)     4    20    28
## # ... with 224 more rows, and 2 more variables: fl <chr>, class <chr>

Try ?mpg to learn more about the data.

1. Engine size in litres is in the displ column.
2. Fuel efficiency on the highway in miles per gallon is given in the hwy column.

To plot engine size (x-axis) against fuel efficiency (y-axis) we do the following:

ggplot(data = mpg) + geom_point(mapping = aes(x=displ, y=hwy))

This plot shows a negative relationship between engine size and fuel efficiency.

What we did here was call the ggplot() function to create an empty graph, where the first argument is the data (here mpg). The we used a geom_point() function to add a layer of points mapping the aesthetics for the x and y axes.

Mapping is always paired to aesthetics aes(). An aesthetic is a visual property of the objects in your plot, such a point size, shape or point colour.

Challenge

Is the trend the same for fuel efficiency in cities?

• Which variable represents city miles per gallon?
• Can you produce a plot of engine size (x-axis) against city fuel efficiency (y-axis) ?
• Is the trend positive or negative between engine size and city fuel efficiency ?

Here we just used the values of x and y as points, but we could add a colour or shape aesthetic too.

But why?

Suppose we also were trying to understand the relationship between the type of car, the engine size and fuel efficiency. To contain all this information in a single plot we need to have a third aesthetic, a third property to describe the car type, contained in the class column.

ggplot(data = mpg) + geom_point(mapping = aes(x=displ, y=hwy, colour=class))

As we might expect, bigger cars such as SUVs tend to have bigger engines and are also less fuel efficient, but some smaller cars such as 2seaters also have big engines and greater fuel efficiency. Hence we have a more nuanced view with this additional aesthetic.

Check out the ggplot2 documentation for all the aesthetic possibilities (and Google for examples): http://ggplot2.tidyverse.org/reference/

So now we have re-usable code snippet for generating plots in R:

ggplot(data = <DATA>) + 
  <GEOM_FUNCTION>(mapping = aes(<MAPPINGS>))

Concretely, in our first example <DATA> was mpg, the <GEOM_FUNCTION> was geom_point() and the arguments we supplies to map our aesthetics <MAPPINGS> were x = displ, y = hwy.

What is the availability of different types of diamond?

Plotting in R can do more than just display raw data, it can transform it at the same time.

Also, we only looked at one type of geometric object, geom_point, last time. Geoms are the type of data display ggplot2 creates e.g. points, box-plots, bar charts. Here we’ll see some of the other geom types.

Let’s look at another built in data frame called diamonds.

diamonds

## # A tibble: 53,940 × 10
##    carat       cut color clarity depth table price     x     y     z
##    <dbl>     <ord> <ord>   <ord> <dbl> <dbl> <int> <dbl> <dbl> <dbl>
## 1   0.23     Ideal     E     SI2  61.5    55   326  3.95  3.98  2.43
## 2   0.21   Premium     E     SI1  59.8    61   326  3.89  3.84  2.31
## 3   0.23      Good     E     VS1  56.9    65   327  4.05  4.07  2.31
## 4   0.29   Premium     I     VS2  62.4    58   334  4.20  4.23  2.63
## 5   0.31      Good     J     SI2  63.3    58   335  4.34  4.35  2.75
## 6   0.24 Very Good     J    VVS2  62.8    57   336  3.94  3.96  2.48
## 7   0.24 Very Good     I    VVS1  62.3    57   336  3.95  3.98  2.47
## 8   0.26 Very Good     H     SI1  61.9    55   337  4.07  4.11  2.53
## 9   0.22      Fair     E     VS2  65.1    61   337  3.87  3.78  2.49
## 10  0.23 Very Good     H     VS1  59.4    61   338  4.00  4.05  2.39
## # ... with 53,930 more rows

This data set has records of about 54,000 diamonds, including price (dollars), quality of the cut (cut) and weight (carat).

Let’s create a bar-plot of diamond cut quality using our code snippet:

ggplot(data = diamonds) + 
  geom_bar(mapping = aes(x=cut))

What’s happened here? Rather than plotting the raw data, ggplot has transformed the data by counting the the number of diamonds in the row of data frame corresponding with each quality.

This statistical transformation of counting the rows is a default of geom_bar(), by reading the help we can find out more ?geom_bar, and how to change it.

We see that ideal cut quality diamonds are most abundant.

The depth measurement describes the relationship of the height to its width as a percentage. Plotting this as a box-plot automatically calculates the the range of depths, and associated statistics for each quality type.

ggplot(data = diamonds) + 
  geom_boxplot(mapping = aes(x=cut,y=depth))

Challenge

As before, we could apply another aesthetic, colour to also look at the same relationship, but also across the clarity types.

Create a box-plot as before, but add colour = clarity to the aesthetics mapping

Comparing our bar chart and box plot we can see that diamonds of the greatest abundance are also generally the most uniform in terms of depth and clarity.

We’ll return to plotting later and look at how to make them suitable for publication by performing tasks such as relabelling axes or adding titles.

Hopefully you are beginning to see how a single line of code can do a lot.

Calling functions

Functions in R are objects followed by parentheses, such as library(). Functions have the form:

function_name(arg1 = val, arg2 = val2, ...)

Let’s use seq() to create a sequence of numbers, and at the same time practice tab completion.

Start typing se in the console and you should see a list of functions appear, add q to shorten the list, then use the up and down arrow to highlight the function of interest seq() and hit Tab to select.

RStudio puts the cursor between the parentheses to prompt us to enter some arguments. Here we’ll use 1 as the start and 10 as the end:

seq(1,10)

##  [1]  1  2  3  4  5  6  7  8  9 10

If we left off a parentheses to close the function, then when we hit enter we’ll see a + indicating RStudio is expecting further code. We either add the missing part or press Escape to cancel the code.

Let’s call a function and make an assignment at the same time. Here we’ll use the base R function seq() which takes three arguments: from, to and by.

Read the following code as *“make an object called my_sequence that stores a sequence of numbers from 2 to 20 by intervals of 2*.

my_sequence <- seq(2,20,2)

This time nothing was returned to the console, but we now have an object called my_sequence in our environment.

Can you remember how to inspect it?

If we want to subset elements of my_sequence we use square brackets [].

For example element five would be subset by:

my_sequence[5]

## [1] 10

Here the number five is the index of the vector, not the value of the fifth element. The value of the fifth element is 10.

And returning multiple elements uses a colon :, like so

my_sequence[5:8]

## [1] 10 12 14 16