R factors

Factors in R is a categorical variables (aka dummy variables). They can be used to keep unique categories from the vector. This can be considered as a analogy of structure set() from Python

It is possible to make factors from any vecotor in R


> test <- c("low", "high", "medium", "high", "medium", "low", "high", "low")
> tt <- factor(test)
> tt
[1] low    high   medium high   medium low    high   low   
Levels: high low medium

To see the content of levels in factors it is possible to use all standard operations, as we use for vectors.


> levels(tt) # Display all levels
[1] "high"   "low"    "medium"
> levels(tt)[2] # Check information about level [2]
[1] "low"
> nlevels(tt)   # Check number of levels
[1] 3

Base function to work with factors are:

  • str() - To check the structure of factor
  • table() - To named vector with all information in our factor
  • as.character() - To give original data vector
  • as.numeric() - To give numerical data of original vector


> str(tt)
 Factor w/ 3 levels "high","low","medium": 2 1 3 1 3 2 1 2
> table(tt)
tt
  high    low medium 
     3      3      2 
> as.character(tt)
[1] "low"    "high"   "medium" "high"   "medium" "low"    "high"   "low"   
> as.numeric(tt)
[1] 2 1 3 1 3 2 1 2

The main interest in factor is the result if table() function, which give us named vector and we van use it as named vector


> unname(table(tt)["low"])
[1] 3
> unname(table(tt)[levels(tt)[2]])
[1] 3
> unname(table(tt)[2])
[1] 3


Published: 2021-11-10 21:36:41