Variables

In this lesson you will learn:

• What a variable is.
• What a data set is.

Using R

Importing data from a text file

A data set in #\mathrm{R}# is in the form of a data frame. If you use #\mathtt{read.table()}# to import data from a text file (in your #\mathrm{R}# working directory) into the #\mathrm{R}# workspace, the result is a data frame.

> MyData = read.table("somedata.txt",header=TRUE)

> class(MyData)

[1] "data.frame"

Loading .RData files

If you use the #\mathtt{load()}# command to load an #\mathtt{.RData}# file (in your #\mathrm{R}# working directory) into the #\mathrm{R}# workspace, you can use the #\mathtt{class()}# command as above to see if it is a data frame.

In this course, you will be given many #\mathtt{.RData}# files which, when loaded, will contain a file which is already in the #\mathtt{data.frame}# format. Note that you do not assign a loaded #\mathtt{.RData}# file to a name, as you do with #\mathtt{read.table()}#.

To see what files were contained in the #\mathtt{.RData}# file after you loaded it, you can see what is in your current #\mathrm{R}# workspace using the #\mathtt{ls()}# command.

> load("my_research.RData")

> ls()

[1] "Data_January_8" "Data_February_4" "Data_March_5"

> class(Data_February_4)

[1] "data.frame"

Subsetting data

Each column of a data frame in #\mathrm{R}# is a variable and will have a name at the top of the data. Either the name is provided in the imported #\mathtt{(header=TRUE)}# or loaded file, or #\mathrm{R}# will assign it a name, such as #\mathtt{V1 \text{,} V2 \text{,}}# etc.

If you want to input the data for that variable into another #\mathrm{R}# function, you reference the column by first giving the name of the data file, then the name of the column, joined by a #\mathtt{$}#.

> Data_Februray_4$temperature

Alternatively, if the variable you want is in column #3#, you can reference it using the column reference:

> Data_February4[,3]

The blank space before the comma means that you want all the rows in column #3#.

If you only want rows #10# through #23#, you would use:

> Data_February_4$temperature[10:23]

or

> Data_February_4$[10:23,3]

If you want rows #7#, #14#, #21#, and #30# through #33#, you can use:

> Data_February_4$temperature[c(7,14,21,30:33)]

or

> Data_February_4$[c(7,14,21,30:33),3]

If you want all rows except row #8#, you can use:

> Data_February_4$temperature[-8]

or

> Data_February_4[-8,3]

Extracting variables

You can also extract a variable from the data frame so that it exists both within the data frame and also outside the data frame as a vector, directly accessible within the #\mathrm{R}# workspace. You have to choose a name for the vector when you extract it.

> Temps = Data_February_4$temperature

Then even if you alter the data contained in #\mathtt{Temps}# the original data will still be preserved inside the data frame #\mathtt{Data\_February\_4}#

Sorting vectors

You can sort the vector in order from smallest to largest using the #\mathtt{sort()}# function.

> Temps = sort(Temps)

If you want it sorted from largest to smallest, use:

> Temps = sort(Temps, decreasing=TRUE)

or

> Temps = sort(Temps, d=T)

Reordering columns of a data frame

You can sort the entire data frame by one of its columns using the #\mathtt{order()}# function.

For instance, to sort the data frame #\mathtt{Data\_February\_4}# by the temperature column, use:

> Data_February_4 = Data_February_4[order(Data_February_4$temperature),]

Then the rows of the data frame will be re-ordered from the case with the lowest temperature to the case with the highest temperature.

To sort from highest to lowest instead, use:

> Data_February_4 = Data_February_4[-order(Data_February_4$temperature),]

See #\mathrm{R}# Manual for further information.