Introducing R in three lines

Here is a simple 2 column data file with some (made up) exam results for 4 students:

name    score
Alice   56
Bob     76
Clare   80
David   48
Bob     45
Bob     74
Alice   63
Clare   56
David   64

First of all we read the data into R:

results <- read.table("results.dat", header = TRUE)

R automatically picks up the names of each column from the file (via "header=TRUE"). Unlike many programming languages, there is no need to loop explicitly over each row of data. Neither do we need to specify data types or declare the object "results" which stores the data in R. However, R provides fine-grained control of such details should you want it (see "?read.table" within R for the relevant help page).

In this case, "results" is a data frame, which is a very common data structure in R, resembling a table or spreadsheet of rows and columns of values, with identical data types in each column.

Having read in the data, we might want to calculate the mean score for each student. We can do this easily, using the "aggregate" function:

aggregate(score ~ name, data = results, FUN = mean)

   name score
1 Alice  59.5
2   Bob  65.0
3 Clare  68.0
4 David  56.0

Note here the use of R's formula syntax ("~") for clearly and concisely expressing that we want to perform some action on the variable "score", according to the different values in the corresponding variable "name"; both of these variables are to be found in the object "results".

In this case, we want to calculate the mean value of "score" for each name, but you can supply any suitable function to the argument "FUN", such as sum, min, max etc. or you can write your own function to use.

In addition to summarising the data on screen, you typically also want to plot the results as a graph. There are different ways to do this, but R assigns a default style of graph according to the type of object being plotted, in accordance with its object oriented design:

plot(results)

The first 2 columns of the data frame "results" are automatically used for the X and Y axis data for plotting. Since the "names" column is not numeric data, R reads it in as a factor by default and correspondingly produces a boxplot summary of the results for each student.

Once again, fine control over the plot is achieved with further header arguments to "plot"; see "?boxplot" and "?par" for the relevant help pages.

---

You could plot the same graph explicitly, using the formula syntax from above, as follows:

plot(score ~ name, data = results)

You can find out more about object orientation in R from these help pages: "?class" and "?S3Methods". For example, try the following (anything after a "#" is treated as a comment and ignored by R):

class(results)                  # "data.frame"
methods(class = "data.frame")