Structure of a dataset. Source: R for Data Science by Hadley Wickham & Garrett Grolemund

Get working directory

In most cases, when you open RStudio, the working directory (so where it is currently working) is different than where your dataset is located. To know what is the working directory RStudio is currently using, run getwd(). On MacOS, this function will most likely render a location such as "/Users/yourname/", while on Windows it will most likely render "c:/Documents/". Do not worry if your working directory is different, the most important is to set the working directory correctly (so where your file is located) and not where it is now.

Set working directory

As mentioned earlier, your dataset is most likely located in a different location than your working directory. Without any action from you, RStudio will never be able to import your file as it is not looking in the correct folder (you will encounter the following error in the console: cannot open file ‘data.csv’: No such file or directory). Now, in order to specify the correct location of your file (that is, to tell RStudio in which folder it should look for your dataset), you have three options:

1. the user-friendly method
2. via the console
3. via the text editor (see below why it is my preferred option)

User-friendly method

To set the correct folder, so to set the working directory equal to the folder where your file is located, follow these steps:

1. In the lower right pane of RStudio, click on the tab “Files”
2. Click on “Home” next to the house icon
3. Go to the folder where your dataset is located
4. Click on “More”
5. Click on “Set As Working Directory

Set working directory in RStudio (user-friendly method)

Alternatively, you can also set the working directory by clicking on Session > Set Working Directory > Choose Directory…

Set working directory in RStudio (user-friendly method)

As you can see in the console, any of the two methods will actually execute the code setwd() with the path to the folder you specified. So by clicking on the buttons you actually asked RStudio to write a line of code for you. This method has the advantage that you do not need to remember the code and that you will not make a mistake in the name of the path to your folder. The disadvantage is that if you leave RStudio and open it again later, you will have to specify the working directory again as RStudio did not save your actions via the buttons.

Example of raw data

User-friendly way

As shown below, simply click on the file > Import Dataset…

Import dataset in RStudio

A window which looks like this will open:

Import window in RStudio

From this window, you can have a preview of your data, and more importantly, check whether your data seems to have been imported correctly. If your data have been correctly imported, you can click on “Import”. If this is not the case, you can change the import options at the bottom of the window (below the data preview) corresponding to the information you gathered when looking at the raw data. Below, the import options you will most likely use:

• Name: set the name of your data set (default is the name of the file). Avoid special characters and long names (as you will have to type the name of your dataset several times). I personally rename my datasets with a generic name such as “dat”, others use “df” (for dataframe), “data”, or even “my_data”. You could use more explicit names such as “tennis_data” if you are using data on tennis matches for example. However, the main drawback with using specific names for datasets is that if, for instance, you want to reuse the code you created while analysing tennis data on other datasets, you will need to edit your code by replacing all occurrences of “tennis_data” by the name of your new dataset
• Skip: specify the number of top rows you want to skip (default is 0). Most of the time, 0 is fine. However, if your file contains some blank rows at the top (or information you want to disregard), set the number of rows to skip
• First Row as Names: specify whether the variables names are present or not (default is that variables names are present)
• Delimiter: the character which separate the values. From our raw data above, you can see that the delimiter is a comma (“,”). Change it to semicolon if your values are separated by “;”
• NA: how missing values are specified (default is empty cells). From our raw data above, you can see that missing values are simply empty cells, so leave NA to default or change it to “empty”. Change this option if missing values in your raw data are coded as “NA” or “0” (tip: do not code yourself missing values as “0”, otherwise you will not be able to distinguish the true zero values and the missing values)

After changing the import options corresponding to your data, click on “Import”. You should now see your dataset in a new window and from there you can start analyzing your data.

This user-friendly method has the advantage that you do not need to remember the code (see the next section for the entire code). However, the main drawback is that your import options will not be saved for a future usage so you will need to import your dataset manually each time you open RStudio.

Via the text editor

Similarly to setting the working directory, I also recommend using the text editor instead of the user-friendly method for the simple reason that you can save your import options when using the text editor (and not when using the user-friendly method). Saving your import options in your script (thanks to a line of code) allows you to quickly import your dataset the exact same way without having to repeat all the necessary steps every time you import your dataset. The command to import a CSV file is read.csv() (or read.csv2() which is equivalent but with other default import options). Here is an example with the same file than in the user-friendly method:

dat <- read.csv(
  file = "data.csv",
  header = TRUE,
  sep = ",",
  dec = "."
)

• dat <-: name of the dataset in RStudio. This means that after importation, I will need to refer to the dataset by calling dat
• file =: name of the file in the working directory. Do not forget “” around the name, the extension .csv at the end and the fact that RStudio is case sensitive ("Data.csv" will give an error) and space sensitive inside “” ("data .csv" will also throw an error). In our case the file is named “data.csv” so file = "data.csv"
• header =: are variables names present? The default is TRUE, change it to FALSE if it is not the case in your dataset (TRUE and FALSE are always in capital letters, true will not work!)
• sep =: separator. Equivalent to delimiter in the user-friendly method. Do not forget the “”. In our dataset the separator of the values is a comma so sep = ","
• dec =: decimal. Do not forget the “”. In our dataset, the decimal for the numeric values is a point, so dec = "."
• I do not write that missing values are coded as empty cells in my dataset because it is the default
• Last but not least, do not forget that the arguments are separated by a comma

Other arguments exist, run ?read.csv to see all of them.

After the importation you can check whether your data have been correctly imported by running View(dat) where dat is the name you chose for your data. A window, similar than for the user-friendly method, will display your data. Alternatively you can also run head(dat) to see the first 6 rows and check that it corresponds to your Excel file. If something is not correct, edit the import options and check again. If your dataset has been correctly imported, you can now start analyzing your data. See other articles on R if you want to learn how.

The advantage of importing your dataset directly via the code in the text editor is that your import options will be saved for a future usage, preventing you from importing it manually every time you open your script. You will, however, need to remember the function read.csv() (not the arguments since you can always check them in the help documentation).