In the one console line that it took to create a subset of our data, you encountered your first multi-argument (and variable-argument) function in the R language.

You were first introduced to functions in Chapter 2. There, the date() function received no arguments and output the current date and time in the R console. Shortly after, you used setwd(dir) and getwd(dir) to set and retrieve your R working directory. Both of these functions received a single argument. With subset(data, ...) you have used your first multi-argument function. Further, subset(data, ...) represents a variable-argument function, meaning that the exact number of arguments it receives can be different depending the circumstance. In our example, we used two arguments. However, we could have used more arguments to further specify our subset. For instance, we could have added an additional argument to our subset(data, ...) function that told R to include only certain columns in its output.

Any time that you see ellipsis (...) in an R function, you know that it accepts a variable number of arguments. In contrast, some multi-argument functions, such as cor(x, y, use, method) for correlations, accept no more and no less than a specific number of arguments. However many others, such as plot(x, y, ...) for scatterplots, can accept relatively few or many arguments, depending on the situation.

In the second argument of our subset(data, ...) function, we employed the equivalency operator. It is formed by two consecutive equals signs (==). This operator evaluates the equivalency of two statements, the one to its left and the one to its right. If the statements are equal, then the argument is deemed True. If not, it is considered False.

Conversely, the non-equivalency operator, which is formed by an exclamation point joined with a single equals sign (!=), tests to see if two statements are not equal. If they indeed are not, then the argument is deemed True, otherwise False.

Our implementation of the subset(data, ...) function made use of two arguments. The first referred to our data source, the battleHistory variable. The second specified the exact data that we wanted to pull from that source.

> subsetHeadToHead <- subset(battleHistory, battleHistory$Method
== "headToHead")

In our case, we wanted to include battles only if they employed the head to head combat method. To clarify this operation, let us dissect the second argument.

battleHistory$Method == "headToHead"

You should already be familiar with the left-hand segment, which selects the Method column from the battleHistory dataset. By using the equivalency operator (==) and "headToHead", we are telling our function to select only the rows in the Method column that contain a value of headToHead. In words, this argument can be read as "in the battleHistory dataset, select rows from the Method column only if they have a value of headTohead." Hence, our resulting subset yielded only the 30 rows from our original dataset that contained the head to head combat method.

Now that you are familiar with extracting information from large datasets, use the subset(data, ...) function to create subsets for each of the remaining battle methods surround, ambush, and fire. Save each of these subsets into new variables, named subsetSurround, subsetAmbush, and subsetFire respectively.