The dataset used in examples In R we can use functions whose names start with read. and write. to import and export data for the file types that we want. As main example we will use measurement data from a study (MAS et al, 1999) about the concentration profile of the active substance after oral administration of 125 mg MDMA (better known as Ecstasy or XTC). The plasma concentration \(C\) - time \(t\) data are listed in the table below.

Entering data from the keyboard and writing to a file First we will enter the above data from the keyboard, while omitting on purpose the final row.

First we set the working directory and show the files present here

> setwd("C:/temp") # set working directory
> dir() # print working directory with files
[1] "age.csv"  "age.xlsx"

Next we enter the above dataset except the last row and create a date frame:

> t <- c(0.25, 0.50, 0.75, 1.00, 1.50, 2.00, 
+        3.00, 4.00, 6.00, 8.00, 10.00)
> C <- c(0, 19.2, 74.6, 125.0, 200.0, 223.1, 
+        215.4, 192.3, 157.7, 130.8, 115.4)
> df <- data.frame(t,C)

We export the current dataset via the function write.table() to the file MDMAdata.txt, with values separated by a semicolon, and with only column names in the first row.

> write.table(df, file="MDMAdata.txt", sep=";", 
+             row.names=FALSE, col.names=TRUE)

Let's check that a new file has occurred in the working directory:

> dir()  # extra file in C:/temp
[1] "age.csv"      "age.xlsx"     "MDMAdata.txt"

Now we append more data to the file:

> write.table(data.frame(t=24.00, C=38.5), file="MDMAdata.txt", append=TRUE,
+             sep=";", row.names = FALSE, col.names = FALSE)

In order to be sure that we can import the exported dataset and get something new, we first remove the data frame df from the workpace before reading the dataset from the file MDMAdata.txt. In the function read.table() we only specify the filename and separator, and indicate that the first row is a header containing the column names of the data frame that we create from the imported data.

> rm(df)  # remove existing data frame df
> df <- read.table(file="MDMAdata.txt", header=TRUE, sep=";")
> df
       t     C
1   0.25   0.0
2   0.50  19.2
3   0.75  74.6
4   1.00 125.0
5   1.50 200.0
6   2.00 223.1
7   3.00 215.4
8   4.00 192.3
9   6.00 157.7
10  8.00 130.8
11 10.00 115.4
12 24.00  38.5

Data import via read.table() The above example already shows the main ingredients of the process of importing and exporting data via a delimited text file. Let's focus now on the import data from delimited text files using read.table() This function reads a file in table format and saves it as a data frame. Each row of the table appears as one line in the file. The syntax is

dataframe <- read.table(file, options)

where file is a delimited ASCII file and options are parameters controlling how data must be processed. The table below lists the most common options.

The base-R environment also provides the functions read.csv() and read.delim() for importing rectangular text files. These are simply wrapper functions that call read.table() with specific defaults. For example, read.csv() calls read.table() with header=TRUE, and sep="," while read.delim() calls read.table() with header=TRUE and sep="\t" (a tab). Details are provided in the read.table() help.

The package readr

The package readr provides a powerful alternative to basic R functions for reading rectangular text files. The primary function is read_delim() with helper functions read_csv() and read_tsv() for reading comma-delimited and tab-delimited files, respectively. The main advantage of using the import functions of this package is that they are significantly faster than the basic R functions. Additionally, they are very good at guessing the correct data type of each column (numeric, character, date, and date-time). Functions in the this package return data as tibbles (data frames with some specialized features).

In the example below we use the file grades.csv, the comma-separated-values text file created by saving the Excel file studentgrades.xlsx in this format. This Excel sheet is a fictitious list of student grades for maths and stats exams.

> install.packages("readr")
> library(readr)
> setwd("C:/temp")
> grades <-
+   read_csv(file="studentgrades.csv")

-- Column specification ------
cols(
  studentID = col_double(),
  first.name = col_character(),
  last.name = col_character(),
  maths = col_double(),
  stats = col_double()
)
> View(grades)
> grades

# A tibble: 6 x 5
  studentID first.name last.name maths stats
      <dbl> <chr>      <chr>     <dbl> <dbl>
1       101 John       Smith       7.5   5  
2        34 John       Jones       4.5   5  
3        76 Jane       Smith       8     8.5
4       104 Daniel     Brown       6     6.5
5       111 John       Taylor      9     9  
6        83 Esther     William    10     9.5