The apply family of functions can be a handy way to perform repetitive tasks quickly. Today we will be using apply functions to improve some previous tasks. First we will be revisiting `comma_split()` (for the last time) and learning to use it on several columns at once. Next, we will be tackling a common annoyance when it comes to loading several files into R.

The first thing we will be doing today is combining our `comma_split()` function and `lapply()` in order to split all of our `<DRINK>_day` comma separated columns at once. I've provided the fully generalized version of `comma_split()` we've developed in the past few worksheets below. Run the following to add it to your environment.

Once you've got the function in your environment, we are going to use `lapply()` to apply it over all the relevant columns in our `survey` dataframe. The first argument to `lapply()` will be the thing you want to apply over. In this case it will be the `survey` dataframe. The second argument, `FUN`, is the function you want to apply to the first argument, `comma_split()`.

{{% notice tip %}}

Note that when supplying a dataframe to `lapply()` it interprets this as applying `FUN` to each column in the dataframe.

{{% /notice %}}

This covers what we want to apply (`FUN`) and what we want to apply it to (`survey`), but there is one hangup. How do we pass our arguments to `comma_split()`? The first argument of `FUN` is always assumed to be the current part of your data the apply function is going over. In the case of `survey` and `comma_split()` the first argument to `comma_split()`, `vector_to_split`, will be the columns from `survey`.

But how do we specify the rest of the arguments? This is where some of our advanced function knowledge comes in. The third argument to `lapply()` is `...`, which you may recall means we can *pass* arguments through the `lapply()` function. As long as we put in an argument to `lapply()` that matches the argument names in `FUN`, (`comma_split()`), they will be passed through. So in this case if we supply `lapply()` with an argument called `possible_columns`, that will be passed to `comma_split()`.

Use `lapply()` to apply `comma_split()` to all of the `<DRINK>_days` columns in our `survey` dataframe. Save the results as `drink_dfs`. **DO NOT** include the `()` after `comma_split` when providing it as an argument; it will produce an error.

You will have gotten a list object of length 5 back. Recall that lists are super-vectors. In this case, each element of our list contains an entire dataframe! Each of these dataframes is the normal output from `comma_split()`. We could combine these with our `survey` data if we cared to dome some analyses.

Using `lapply()` does not need to be limited to uses within R. One common way to use it is when loading in data. To try this, we're going to load in a number of `.csv` files containing economic and population data on Massachusetts from the [American Community Survey five-year estimates (ACS5)](https://www.census.gov/data/developers/data-sets/acs-5year.html). You can think of it as a yearly mini-census.

First, we need to have a directory of data files with identical structures. For this example, we can use some data I have hosted on GitHub.

we will use the data from lab 3 (aggregation and merging). First, specify the path to the data folder of lab 3. You have to find the specific path on your computer.