Tag: variable labels

Intro

Defining variable labels is a useful way to describe and document datasets. Unlike SPSS, which makes it very easy to define variable labels using the data editor, base R doesn't provide any function to define variable labels (as far as I know).

However, Daniel Luedecke's R package sjlablled fills this gap. Let's give an example.

Defining variable labels

First, we load the mtcars data frame and define variable labels for all of the 11 variables:

data(mtcars)
labs <- c("Miles/(US) gallon", "Number of cylinders", "Displacement (cu.in.)", 
    "Gross horsepower", "Rear axle ratio", "Weight (1000 lbs)", "1/4 mile time", 
    "V/S", "Transmission", "Number of forward gears", "Number of carburetors")

Assigning labels to variables

Second, we assign the variable labels to the variables of the mtcars data frame:

library(sjlabelled)
mtcars <- set_label(mtcars, label = labs)

When we have a look at the mtcars data frame using RStudio's data viewer, we find the variable labels placed right underneath the variable names:

Moreover, we may as well save both variable names and labels into a data frame:

library(dplyr) # for data manipulation
library(knitr) # for printing tables
df <- get_label(mtcars) %>%
        data.frame() %>%
          rename_at(vars(1), funs(paste0('var.labs'))) %>%
            mutate(var.names = colnames(mtcars)) 
kable(df, align = 'lc')

var.labs	var.names
Miles/(US) gallon	mpg
Number of cylinders	cyl
Displacement (cu.in.)	disp
Gross horsepower	hp
Rear axle ratio	drat
Weight (1000 lbs)	wt
¼ mile time	qsec
V/S	vs
Transmission	am
Number of forward gears	gear
Number of carburetors	carb