A Coding Tips
A.1 Keyboard Shortcuts
There are three things you will do often in this course for which there are keyboard shortcuts that will save you time and energy over the long run.
- Insert a code chunk:
Cmd+Opt+Ion Mac orCtrl+Alt+Ion Windows - Run the current line or selection of code:
Cmd+Returnon Mac orCtrl+Enteron Windows - Knit document:
Cmd+Shift+Kon Mac orCtrl+Shift+Kon Windows
There are many more keyboard shortcuts. Accessing keyboard shortcuts has a keyboard shortcut! It is Opt+Shift+K on Mac or Alt+Shift+K on Windows.
A.2 Specifying Datasets and Variables
A.2.1 Datasets
R can store multiple datasets or objects at a time for you to work with. Therefore, you must tell R the dataset on which to run a function. Suppose I want R to provide a quick preview of a dataset named gapminder. I can use the glimpse() function for this like so:
Rows: 1,704
Columns: 6
$ country <fct> "Afghanistan", "Afghanistan", "Afghanistan", "Afghanistan", …
$ continent <fct> Asia, Asia, Asia, Asia, Asia, Asia, Asia, Asia, Asia, Asia, …
$ year <int> 1952, 1957, 1962, 1967, 1972, 1977, 1982, 1987, 1992, 1997, …
$ lifeExp <dbl> 28.801, 30.332, 31.997, 34.020, 36.088, 38.438, 39.854, 40.8…
$ pop <int> 8425333, 9240934, 10267083, 11537966, 13079460, 14880372, 12…
$ gdpPercap <dbl> 779.4453, 820.8530, 853.1007, 836.1971, 739.9811, 786.1134, …Failure to tell R a dataset that has been loaded in your environment will result in an error. For example, suppose I misspell the dataset so that R looks for an object that does not exist.
A.2.2 Variables
Some functions pertain not to an entire dataset but to a specific variable within a dataset. Suppose I wanted to compute an average using the mean() function. Only specifying the dataset results in an error because the mean of a dataset makes no sense.
Instead, I need to specify a variable for which I want the average. To specify a variable within a dataset, we use the $ operator. Suppose I want to compute the average life expectancy, lifeExp, in the gapminder dataset.
[1] 59.47444A.3 Assignment and Pipe Operators
A.3.1 Assignment Operator
Whenever we run a function we have the option of saving the result as a new object to reference for future use. To save a the result of anything to a new object, we use the assignment operator, <-. Whatever happens on the right side of <- is assigned to whatever name we give the new object on the left side.
I just computed average life expectancy. Suppose I want to save that result. In that case, I would use the following code:
This avg_lifeExp will now show up in my environment pane (top-right) as a single value. This works not just for specific values but for anything we want to save to use later in our code.
Note that R did not print out the result like it did when I ran mean(gapminder$lifeExp) above. This is because R assumes I do not want the printout because I am saving it. If I want R to print the result, I can simply run the object name like so
[1] 59.47444or I can wrap the code originally assigning the new object in parentheses
[1] 59.47444You will use the assignment operator often in this course. The keyboard shortcut for it is Opt + -, that is option and the minus sign key on Mac, or Alt + - on Windows.
A.3.2 Pipe Operator
We do not have to do one thing at a time in a code chunk, nor do we need to save a new object with each function we apply to an existing object.
Suppose I want a dataset that includes only 1952 as well as country and life expectancy. I could use code like so (this involves functions you may not know yet):
The first line of the above code uses the filter function to save a new object named gap_1952 that contains gapminder observations only for which year equals 1952. The second line of the above code uses the select function to save a new object named gap_1952_lifeExp that includes only the country and lifeExp variables from the gap_1952 dataset.
That code includes unnecessary intermediate object/dataset. It is also difficult to read and follow because you have to track which datasets are used in each step. The pipe operator makes this sort of iterative process much easier to code and read.
The pipe operator pipes/feeds the result of what precedes it to the next line and so on. In the code above, I start by naming a new object, gap_1952_lifeExp. This new object is determined by taking the gapminder dataset, then piping it to the filter function that keeps observations for which year equals 1952, then piping the result of that – which is equivalent to the intermediate gap_1952 dataset from before – to the select function that keeps only the country and lifeExp variables.
Now we can compute the average life expectancy in 1952 like so:
[1] 49.05762Note how much easier it is to read the code using the pipe operator compared to the code that does not. With the pipe operator, you can read code from left-to-right to understand what is being done. Also, recall that you must specify the dataset for any function.
If you use the pipe operator, you do not need to specify the dataset in every function included in the pipe because you will have already fed the dataset to the next line containing the function.
For example, the filter function by default requires us to specify the object like so
This is the same code that produced the above intermediate dataset, gap_1952. But note how when using pipes, one does not need to specify the dataset within the filter function. This is because the pipe operator already feeds the gapminder dataset to the filter function.
Forgetting to exclude the dataset from a function when using the pipe operator will result in an error.
gap_1952 <- gapminder %>%
filter(gapminder, year == 1952)
Error: Problem with `filter()` input `..1`.
x Input `..1$country` must be a logical vector, not a factor<bf6dc>.The keyboard shortcut for the pipe operator is Cmd+Shift+M for Mac or Ctrl+Shift+M for Windows.