R Factors

A factor is a special data type for categorical variables — variables that belong to a fixed set of named groups. Examples include gender (Male/Female), education level (High School/Bachelor/Master/PhD), or product categories. Factors store the possible values efficiently and give them a defined order when needed.

Why Factors Exist

Without Factors:                     With Factors:
─────────────────────────────────    ──────────────────────────────────────
size <- c("S","M","L","M","S","L")  size <- factor(c("S","M","L","M","S","L"))
Just text — R doesn't know           R knows "S", "M", "L" are the only
 there are only 3 valid sizes         valid sizes → more efficient + safer

Creating a Factor

# Unordered factor (nominal)
blood_type <- factor(c("A", "B", "O", "AB", "O", "A", "B"))
print(blood_type)
# [1] A  B  O  AB O  A  B
# Levels: A AB B O

levels(blood_type)   # "A"  "AB" "B"  "O"
nlevels(blood_type)  # 4

Factor Structure Diagram

blood_type factor:
  Values:  A   B   O   AB   O   A   B
           │   │   │    │   │   │   │
  Stored:  1   3   4    2   4   1   3   (integer codes internally)
           │   │   │    │   │   │   │
  Levels: [1]=A [2]=AB [3]=B [4]=O

Factors store the category labels (levels) once, and then store integers internally for each value. This saves memory with large datasets.

Ordered Factors

edu <- factor(
  c("Bachelor", "PhD", "Master", "High School", "Bachelor"),
  levels = c("High School", "Bachelor", "Master", "PhD"),
  ordered = TRUE
)

print(edu)
# [1] Bachelor PhD      Master   High School Bachelor
# Levels: High School < Bachelor < Master < PhD

edu[2] > edu[3]   # PhD > Master → TRUE
edu[4] < edu[1]   # High School < Bachelor → TRUE

Modifying Factors

# Rename levels
size <- factor(c("S", "M", "L", "M", "S"))
levels(size)[levels(size) == "S"] <- "Small"
levels(size)[levels(size) == "M"] <- "Medium"
levels(size)[levels(size) == "L"] <- "Large"
print(size)
# [1] Small  Medium Large  Medium Small

Counting Levels with table()

blood_type <- factor(c("A","B","O","AB","O","A","B","O","A"))
table(blood_type)
# blood_type
# A  AB   B   O
# 3   1   2   3

Factors in Data Frames

df <- data.frame(
  name   = c("Riya", "Arjun", "Meena", "Dev"),
  gender = factor(c("F", "M", "F", "M")),
  grade  = factor(c("A", "B", "A", "C"),
                  levels = c("C","B","A"), ordered = TRUE)
)

# Filter female students
df[df$gender == "F", ]

# Students with grade A or above
df[df$grade >= "A", ]

Converting Factors

f <- factor(c("3", "1", "2", "1"))

as.character(f)   # "3" "1" "2" "1"
as.numeric(f)     # 3 1 2 1  (careful: this uses internal codes sometimes)

# Safe way to convert factor of numbers to numeric:
as.numeric(as.character(f))   # always correct

Dropping Unused Levels

sizes <- factor(c("S", "M", "L", "XL"), levels = c("XS","S","M","L","XL"))
small_sizes <- sizes[sizes %in% c("S", "M")]

levels(small_sizes)          # still has XS, L, XL (unused!)
droplevels(small_sizes)      # removes unused levels → S, M only

Factors are critical for statistical modeling — regression functions, ANOVA, and plotting all treat factors specially. Using them correctly ensures your categorical data is handled with its proper structure rather than as plain text.